The oxygen-sensing system (green glow) in the nematode Caenorhabditis elegans

For humans, living as we do in an atmosphere with a rich and stable supply of oxygen, maintaining the correct levels in our bodies is a relatively routine task. For organisms that live in soil or water, however, where oxygen levels can fluctuate wildly, the task is far more challenging — and pretty much of a mystery, until now.

A multi-institutional collaboration of researchers that includes a scientist with the U.S. Department of Energy's Lawrence Berkeley National Laboratory (Berkeley Lab) has learned how one soil dweller, the nematode Caenorhabditis elegans, is able to sense oxygen levels in its environment and feed in areas where the concentration of oxygen is just right.

"It's surprising that more than 200 years after Antoine Laurent Lavoisier and Joseph Priestly discovered oxygen, we're still finding out how organisms sense and use it," says Michael Marletta, a chemist who holds a joint appointment with Berkeley Lab's Physical Biosciences Division and the University of California at Berkeley.

Marletta is one of the co-authors of a paper on the oxygen-sensing system, which appears in the July 15 issue of the journal Nature. Other authors are Cornelia Bargmann, who holds a joint appointment with the University of California at San Francisco and the Howard Hughes Medical Institute, plus David Karow of UC Berkeley, Jesse Gray, Hang Lu, and Andy Chang of UCSF, Jennifer Chang of the University of Michigan, and Ronald Ellis of the University of Medicine and Dentistry of New Jersey.

An enzyme points the way

The discovery of the oxygen-sensing mechanism in nematodes actually began as an investigation into a human enzyme, guanylate cyclase, which performs signaling interactions with nitric oxide molecules critical to regulating blood pressure. When nitric oxide enters a human cell it activates guanylate cyclase, which catalyzes the formation of cyclic GMP, a protein that relaxes and dilates blood vessels. The ability of guanylate cyclase to sense and interact with nitric oxide also plays a prominent role in the central nervous system, in particular in the brain. Nitric oxide is also important in the immune system, where it is used as a cell-killing agent.

"When we took apart the guanylate cyclase protein to study nitric oxide signaling, we found that the binding site is a heme molecule almost identical to the heme molecule that gives hemoglobin its red color and binds to oxygen," says Marletta. "However, whereas the heme molecule in hemoglobin cannot discriminate between oxygen or nitric oxide, the heme molecule in guanylate cyclase only binds with nitric oxide. Somehow, nature engineered a way for the guanylate cyclase to screen out the oxygen, which is usually present in much higher concentrations than nitric oxide."

To understand how this screening-out of oxygen was accomplished, Marletta and his colleagues cloned the genes that code for the heme-binding site of the guanylate cyclase and searched the genome databases for homologues in other organisms. A similar guanylate cyclase enzyme was discovered in Caenorhabditis elegans, the tiny primitive worm that has become a staple of biological investigations. However, when the researchers ran tests to study the ability of the nematode's guanylate cyclase, called GCY-35, to bind with nitric oxide, the results were confusing.

"We found no evidence that the GCY-35 enzyme was binding to the nitric oxide, as we thought it would," says Marletta. "However, we were able to isolate the enzyme's location in sensory neurons that are involved in communication with the outside world. From ongoing research with bacterially-related, cyclase-like sensor proteins, we got the idea that maybe the heme domain in GCY-35 was binding to oxygen."

At the time, Marletta and his group were carrying out parallel studies with UCSF's Bargmann, a leading authority on behavioral studies with nematodes. The collaboration hypothesized that the binding of the GCY-35 enzyme to oxygen might be mediating the known behavioral responses of C. elegans to changes in the oxygen levels of its environment. To test this oxygen hypothesis, Gray and Karow began a new set of experiments.

In the Petri dish on the left the oxygen concentration is 21 percent; here nematodes display bordering and clumping behavior. In the dish on the right the oxygen level is 6 percent, and the nematodes are dispersed.

Lu, a postdoctoral engineer in the Bargmann lab, designed a small chamber (about five square centimeters in area and about 100 microns thick) that could be placed over Petri dishes containing nematodes feeding on bacteria. The chamber allowed the researchers to create precisely controlled gradients of oxygen, which for this study ranged from 0 percent to 21 percent, the concentration of oxygen in ambient air.

A little oxygen goes a long way

Systematic genetic engineering and molecular studies confirmed the researchers suspicion that the GCY-35 enzyme was binding to oxygen rather than nitric oxide. However, the discovery of the oxygen-sensing mechanism in the nematodes led to another surprise. It was presumed that the nematodes would prefer the 21-percent oxygen levels, the condition under which nematodes in the laboratory are routinely kept. But the worms actually much preferred oxygen concentrations of only about 6 to 7 percent. This preference for a lower concentration of oxygen led to an explanation of aggregate behavior, long observed in laboratory nematodes, called "bordering and clumping," in which large numbers of the worms gather around the border of a Petri dish.

"Bordering and clumping turns out to be a laboratory artifact," says Marletta. "The bacteria that the worms feed on concentrate at those borders. Since the bacteria are also consuming oxygen, the concentration of bacteria and the swarm of worms combine to lower the oxygen concentration in the immediate environment. We found that when we reduced the oxygen concentration in a Petri dish to 6 percent, the worms dispersed in three minutes."

In that sense, Marletta says, the nematode's ability to sense oxygen is like a dinner bell. As Bargmann explains, "Oxygen sensing may be a hunting strategy, like smelling your food, except that in the case of C. elegans it's actually smelling the activity of the food. Since bacteria are incredible oxygen hogs, the food is consuming oxygen faster than oxygen can diffuse."

Marletta and Bargmann speculate that the preference of the nematodes for lower oxygen concentrations gives them a selective advantage; perhaps it is a protective response against levels of oxygen that would be damaging to the worms or would shorten their lifespan. This same mechanism should also be present in fish and other animals that live in environments where the oxygen levels fluctuate. It may also be similar to the mechanism by which the human body deals with hypoxia (oxygen deficiency).

Says Bargmann, "We do monitor oxygen levels in our bloodstream, using a rice-sized organ called the carotid body that sits at the fork of the carotid artery. When there's even a small drop in oxygen concentrations, as might happen in vigorous exercise, these neuron-like cells evoke a hyperventilation or rapid-breathing response. It might well be that we use a similar oxygen-sensing mechanism as the nematode."

Michael Marletta is a chemist with Berkeley Lab's Physical Biosciences Division and an Aldo DeBenedictis Distinguished Professor of Chemistry at UC Berkeley.

Marletta and his research team have already begun structural chemistry studies that they believe will enable them to predict whether a given guanylate cyclase enzyme will be an oxygen or a nitric oxide sensor. A better understanding of how these enzymes are able to selectively bind to nitric oxide or oxygen will have implications across a wide swatch of biology, including research into human cardiovascular diseases.

Says Marletta, "What I love most about this project is that we started out simply trying to understand how the mammalian cyclase enzyme is able to bind to nitric oxide. It appears to be a rather narrow question, but in searching for the answer, look at all that we've learned about oxygen sensing."

This research was supported by funding from Lawrence Berkeley National Laboratory and the Howard Hughes Medical
Institute. Both Marletta and Bargmann are part of the California Institute for Quantitative Biomedical Research (QB3), a consortium of three UC campuses at Berkeley, San Francisco, and Santa Cruz.

Berkeley Lab is a U.S. Department of Energy national laboratory located in Berkeley, California. It conducts unclassified scientific research and is managed by the University of California. Visit our website at www.lbl.gov/.

Additional information

• "Oxygen sensation and social feeding mediated by a C. elegans guanylate cyclase homologue," by Jesse M. Gray, David S. Karow, Hang Lu, Andy J. Chang, Jennifer S. Chang, Ronald E. Ellis, Michael A. Marletta, and Cornelia I. Bargmann, appears in the 15 July 2004 issue of Nature.
• Michael Marletta can be reach at (510) 643-9325, marletta@berkeley.edu
• Cori Bargmann can be reached at (415) 476-3558, cori@itsa.ucsf.edu