Symbiosis in the Deep Sea

The remarkable density of life at deep-sea hydrothermal vents is explained by die mutually beneficial symbiosis of in vertebrate animals and sulfide-oxidizing bacteria that colonize their cells

by James J. Childress, Horst Felbeck and George N. Somero

Biologists categorize many of the world’s environments as deserts:

regions where the limited avail­ability of some key factor, such as wa­ter, sunlight or an essential nutrient, places sharp constraints on the exis­tence of living things. Until recently the deep sea was considered to be such a desert, where the low abundance of organisms stems from the extreme limitation of the food supply.

Yet there is one habitat in the deep sea where the density of life equals, if it does not surpass, what is found in any other marine ecosystem. It is the system of hydrothermal vents, or deep-sea hot springs, situated at sea­floor spreading centers. The vents, dis­covered only 10 years ago, are found along ridges at the bottom of the ocean where the earth’s crustab plates are spreading apart.

It was at a spreading ridge in the Pa­cific Ocean, 320 kilometers northeast of the Galapagos Islands, that geolo­gists on the research submarine Alvin in 1977 discovered an oasis of densely packed animal life 2,600 meters below the surface. Here were species previ­ously unknown to science, living in to­tal darkness in densities enormously higher than had ever been thought pos­sible in the deep sea. Giant tube worms as much as one meter long, large white clams 30 centimeters in length and clusters of mussels formed thick ag­gregations around the hydrothermal vents. There were smaller but nonethe­less significant numbers of shrimps, crabs and fishes.

Such biological density was com­pletely unexpected, and also very puz­zling. Any ecosystem depends on the presence of primary producers: au­totrophic (self-supporting) organisms that synthesize their own reduced car­bon compounds, such as carbohy­drates, from carbon dioxide. Green plants that convert carbon dioxide into carbohydrates in the presence of sun­light are called photoautotrophs, and they are the primary producers in most marine and terrestrial ecosystems.

Yet clearly photosynthesis is impos­sible at the depth of the vents. Not enough light penetrates beyond the upper 200 or 300 meters of the ocean’s surface to support photosynthesis. Be­low the sunlit surface the density of life falls off rapidly because there is less food. The organisms that live in the deep sea depend on organic matter that drifts down from the sunlit “eu­photic” zone to the ocean floor. Most organic matter is synthesized, con­sumed and recycled in the euphotic zone, and only a small fraction sinks to the deep sea. Yet at the vents life was blooming 2,600 meters below the surface of the ocean. What unique fea­ture of the hydrothermal-vent system could explain the abundance of life at these depths?

It seemed possible that the density of life at these deep-sea hot springs might be explained on the basis of temperature. In contrast to most of the deep sea, which is very cold (from two to four degrees Celsius), the vent wa­ters have average temperatures rang­ing from 10 to 20 degrees. Obser­vations made over the past six years have shown, however, that the warm temperatures do not account for the unique fauna of the vent ecosystem.

Instead there is a better explanation for these teeming oases of life at the bottom of the ocean. Water samples taken from the vent sites and analyzed by John M. Edmond of the Massa­chusetts Institute of Technology and his colleagues indicated that these un­derwater hot springs, bike many hot springs on land, are rich in hydrogen sulfide. This energy-rich but highly toxic compound is found at high lev­els in the water that flows from cracks in the ocean floor [see “Hot Springs on the Ocean Floor,” by John M. Ed­mond and Karen Von Damm; SCIEN­TIFIC AMERICAN, April, 1983].

For years biologists had known that sulfide-rich habitats, such as terrestrial hot springs, support large numbers of free-living bacteria. Like green plants, they are autotrophs, but they get en­ergy for carbon fixation—the conver­sion of carbon dioxide into organic molecules that serve as nutrients—not from the sun but from the oxidation of hydrogen sulfide.

Holger W. Jannasch of the Woods Hole Oceanographic Institution and David Karl of the University of Ha­waii at Manoa isolated bacteria from the vent waters and did a series of experiments demonstrating that some species of vent bacteria are indeed au­totrophic and depend on hydrogen sul­fide and other reduced forms of sulfur for their metabolic activities. These sulfide-oxidizing bacteria might form the base of the vent food chain by pro­viding food for the animal species.

Although they are very distantly re­bated, green plants and sulfur bacteria are functionally similar in an impor­tant way: both are primary producers. Whereas green plants are photoau­totrophs, sulfur-oxidizing bacteria ca­pable of using an inorganic energy source to drive carbon dioxide fixation are called chemolithoautotrophs: they are literally consumers of inorganic chemicals that can exist autonomous­ly—without an external source of re­duced carbon compounds. Both fix carbon dioxide through a series of bio­chemical reactions collectively known as the Calvin cycle. The Calvin cycle functions in much the same way for bacteria as it does for green plants; the end product for both is reduced car­bon.

It appeared, then, that the sulfur bac­teria were in effect serving as the “green plants” of the vents and that their “sunlight” was hydrogen sulfide and other reduced carbon compounds. A major complication of this simple model was soon recognized: one of the dominant vent animals, the tube worm Riftia pachyptila, seemed to lack any means of harvesting the abundant crop of sulfur bacteria proliferating in the waters all around it.

Riftia is a strikingly unusual crea­ture by conventional anatomical stan­dards. It is essentially a closed sac, without a mouth or a digestive system and with no other means of ingesting particulate food. At its anterior tip there is a bright red branchial (gill­like) plume where oxygen, carbon di­oxide and hydrogen sulfide are ex­changed with the ambient seawater.

Below the plume there is a ring of mus­cle, the vestimentum, that anchors the worm in its white tube. Most of the rest of the animal consists of a thin-walled sac that contains the worm’s in­ternal organs. The largest of them is the trophosome, which occupies most of the body cavity. As its name (“the feeding body”) suggests, the tropho­some contributes significantly to the worm’s nutrition—but it lacks a chan­nel through which particulate mate­rials from the outside world can en­ter the worm. The big question was how, in view of its unusual anatomy, Riftias manages to obtain the nutrients it needs for survival.

Microscopic studies done by Colleen M. Cavanaugh of Harvard Uni­versity and her colleagues and parallel biochemical studies done by us pro­vided the first clues. Examination of the trophosome of Riftia revealed that it is colonized by vast numbers of sul­fur-oxidizing bacteria. We recognized that the bacteria and Riftia had estab­lished what is known as an endosymbi­otic relation.

Symbiosis is the co-occurrence of two distinct species in which the life of species is closely interwoven with the life of the other. Symbioses vary from relations that are beneficial to one partner and harmful to the oth­er (parasitism) to relations from which both the partners benefit (mutualism). When one species, known as the sym­biont, lives within the body of the oth­er, known as the host, the relation is called endosymbiotic.

The Riftia-bacteria endosymbiosis is mutualistic. The tube worm receives reduced carbon molecules from the bacteria and in return provides the bacteria with the raw materials need­ed to fuel its chemolithoautotrophic metabolism: carbon dioxide, oxygen and hydrogen sulfide. These essential chemicals are absorbed at the plume and transported to the bacteria in the trophosome by the host’s circulatory system. The worm’s trophosome can be thought of as an internal factory, where the bacteria are line workers producing reduced carbon compounds and passing them to the animal host to serve as its food.


The ability of Riftia to absorb sul­fide from vent water and transport it to the bacteria in its trophosome without either poisoning itself or de­grading the sulfide presented a major puzzle. Hydrogen sulfide is a highly toxic compound, comparable to cya­nide in its ability to block respiration, the process whereby the animal uses oxygen. In most animals sulfide inhib­its respiration in two ways: by blocking oxygen’s binding sites on its major car­rier, the hemoglobin molecule, and by poisoning an important respiratory en­zyme, cytochrome c oxidase. Studies of Riftia showed, however, that sulfide has no effect on oxygen binding and that the worm’s respiration rate is sub­stantial even in the presence of sulfide concentrations lethal to most animals.

We wanted to know how aerobic respiration in Riftia is possible in the presence of high sulfide concentra­tions. Clearly Riftia had three obsta­cles to overcome. First, it needed to evolve a special transport system to extract sulfide from vent water. Sec­ond, it needed to transport sulfide in its blood without allowing the sulfide to compete with oxygen for binding sites on the hemoglobin molecule or to re­act with oxygen. (In the presence of oxygen, sulfide is highly unstable, de­composing rapidly to oxidized forms such as thiosulfate and elemental sul­fur.) Third, it needed a mechanism to prevent sulfide from diffusing into its cells and poisoning respiration.

We collaborated with Mark A. Pow­ell (who is now at the University of California at Davis) and Steven C. Hand (now at the University of Colorado at Boulder) to isolate cytochrome c oxidase from plume cells and exam­ine its behavior in the presence of sul­fide. Cytochrome c oxidase is respon­sible for the final step in the chain of metabolic reactions known as oxi­dative phosphorylation, the most im­portant process by which adenosine E AND triphosphate (ATP), the major energy currency of the cell, is synthesized in aerobic (oxygen-using) organisms. In most animals minute concentrations of sulfide are enough to inhibit cyto­chrome c oxidase. We first hypothe­sized that Riftia might have evolved a sulfide-insensitive form of the en­zyme. Experiments showed this is not the case: highly purified cytochrome c oxidase from Riflia is just as sensitive to sulfide poisoning as cytochrome c oxidase from other animals.  We noted in experiments with the enzyme that its sensitivity to sulfide depended on the extent to which we isolated it from other proteins of the plume. As we went through successive purification steps we observed a sub­stantial decrease in the bright red color of our preparation, and with it an in­crease in the sensitivity of cytochrome c oxidase to sulfide poisoning. The col­or change suggested that something in the blood was protecting cytochrome c oxidase from the toxic effects of sulfide. We proved this by adding a minute amount of whole blood to a poisoned enzyme system. Almost im­mediately the cytochrome c oxidase activity returned to a normal, unin­hibited level. The observation showed that a blood component—perhaps he­moglobin—was binding sulfide more strongly than the sulfide-sensitive cytochrome c oxidase system was, there­ by preventing poisoning of respiration. This discovery and simultaneous stud­ies of the worm’s hemoglobin done during a 1982 expedition to the East Pacific Rise vent site indicated that he­moglobin is indeed a key player in the worm’s internal transport system.  Riflia has a rich blood supply: the deep red of the branchial plume is   due to the presence of a large volume   of hemoglobin-rich blood, which ac­counts for more than 30 percent of the worm’s total volume. The total con­centration of hemoglobin per liter­ approximately half what it is in human blood—is extremely high for an invtebrate. Moreover, Riftia  hemoglobin is very different from human and  other vertebrate hemoglobins.

 It is large with a molecular weight of as much as two million daltons (human hemoglobin  has a molecular weight of 64000 daltons). Rather than being contained within red blood cells, it circulates. freely in the serum. It also has an usually high affinity for oxygen as well as  as an unusually high carrying capac­ity for oxygen. Riftia hemoglobin is thus well adapted for extracting oxy­gen from the vent waters and trans­porting it to the cells of the tube worm and to its symbiont.

There is an even more important and striking difference between the he­moglobin of Riftia  and other hemoglo­bins: the tube-worm molecule can bind both oxygen and hydrogen sulfide si­multaneously. This discovery, made by Alissa J. Arp (who is now at San Francisco State University) and us, suggested that the site where sulfide binds to the molecule is different from the site where oxygen binds. Charles R. Fisher, Jr., of the University of Cal­ifornia at Santa Barbara showed that the Riftia hemoglobin molecule stabi­lizes mixtures of oxygen and sulfide, preventing the spontaneous oxidation of sulfide. We therefore concluded that hemoglobin plays a dual role in the tube worm: it prevents sulfide from poisoning respiration and it also protects sulfide by allowing it to be transported to the trophosome without being oxidized.

After it is unloaded in the cells of the trophosome, the sulfide is oxidized. The oxidation takes place in the bacte­rial symbionts, as was demonstrated by a special staining procedure that signals the presence of hydrogen sul­fide. Further investigation by our group showed that the sulfide oxida­tion drives the synthesis of ATP and the fixation of carbon dioxide. More­over, we showed that the activity of the Calvin cycle is approximately the same in the trophosome bacteria as it is in the leaves of a green plant.

Studies of other vent animals reveal that Riftia pachyptila is not the only hydrothermal-vent species that has evolved a symbiotic relation with sul­fur bacteria. The large white clam, Ca­lyptogena magnifica, and the mussel, Bathymodiolus thermophilus, also de­pend on chemosynthetic endosymbi­onts for food. But these species have evolved quite different approaches to the same problem.

In Calyptogena the bacteria are not in an internal organ but in the gills, where they can readily obtain oxygen and carbon dioxide from the respira­tory water flow. The basic metabolic plan is the same, however: the bacte­ria oxidize sulfide and supply the clam with fixed carbon compounds. Like other invertebrates harboring sulfur bacteria as endosymbionts, Calypto­gena has a greatly reduced ability to feed on and digest particulate foods.

Still, we were puzzled at first by the data from water and blood samples. The clams must be concentrating sul­fide in their blood, because the sulfide level there is orders of magnitude higher than the concentration in the ambient water. Because of the way the clams are oriented in the vent waters, however, the sulfide concentration in the water bathing their gills is low. From what source, then, do these giant clams obtain sulfide to feed their en­dosymbionts? The path turns out to be indirect. Apparently the clams absorb sulfide through their large, elongated feet, which extend into the hydrother­mal vents, where the concentrations of sulfide are highest. Once it is absorbed through the clam’s feet, sulfide is transported by the blood to the bacte­ria in the gills.

Transport of sulfide in the blood of Calyptogena is a very different process from the one described for Riflia. The clam’s hemoglobin (which is incor­porated within the animal’s red blood cells) is irreversibly poisoned by sul­fide, and so it cannot serve as a sul­fide-transport protein. Calyptogena has overcome the problem of sulfide poi­soning by evolving a special sulfide-transporting protein. It is an extremely large protein and it circulates in the se­rum rather than in the red cells. The protein has a dual function: it protects the animal’s hemoglobin and cyto­chrome c oxidase from sulfide poison­ing and also protects sulfide against oxidation on the way to the gills. The binding of sulfide to the protein is re­versible: it is off-loaded to bacteria in the gill (by mechanisms not yet under­stood), where it is oxidized to provide energy for the Calvin cycle.

Much less is known about symbio­sis in Bathymodiolus thermophilus. Like the giant clam, the mussel has bacteri­al symbionts in its gills, but the path­ways of sulfide transport and metabo­lism are not yet understood. Kenneth L. Smith, Jr., of the Scripps Institution of Oceanography has, however, car­ried out an interesting series of experi­ments demonstrating that the relation between these deep-sea mussels and the vent water is obligatory. When he moved mussels away from the im­mediate vicinity of a vent to a more peripheral region, they showed clear signs of starvation. Such starvation is a naturally occurring process at hy­drothermal-vent sites. Individual vent sites are active for only a few decades at most. Large numbers of dead mus­sels and clams litter sites where water flow has ceased, indicating that for these animals survival is not possible without a supply of sulfide.

Growth rate serves as another in­dication that symbiosis is effective in meeting the nutritional requirements of the animal hosts. The rapid turn­over of life at these vent sites is reflect­ed in the accelerated growth and the quick attainment of reproductive age that are observed in these animals. Work done by Richard A. Lutz of Rut­gers University show that the clams and mussels of the vents grow as fast as the fastest-growing bivalves found in shallow waters.

The tube worm, the clam and the mussel, the largest and most dominant numerically of the vent species, owe their ecological success to their symbi­osis with sulfur-metabolizing bacteria. Many of the smaller and less conspicu­ous vent animals lack symbionts. They obtain their nutrients either by filter­ing particulate food, such as bacteria, from the water or by feeding on the an­imals that do contain symbionts. We have observed vent crabs, for exam­ple, in the act of feeding on the respira­tory plume of Riftia.

The symbiont-free animals are in­teresting objects to study in their own right. Because sulfide is readily ab­sorbed across the surface of an inver­tebrate’s body, the symbiont-free an­imals, like the symbiont-containing species, have had to evolve mech­anisms to prevent sulfide poisoning. With the collaboration of Russell Vet­ter of Scripps and Mark Wells of Santa Barbara, we examined these mecha­nisms in the vent crab.

The animal’s heart rate and the beat­ing of its scaphognathites (appendages that drive the flow of respiratory wa­ter) did not change when we raised sul­fide concentrations well above normal levels in ambient seawater, and the level of sulfide in the crab’s blood in­creased only minimally. Because we could not isolate from the crab’s blood a sulfide-binding protein comparable to the tube worm’s hemoglobin, we as­sumed this species must have evolved some different strategy for detoxifying sulfide. We found that the crab can de­toxify the sulfide it absorbs by oxidiz­ing sulfide to thiosulfate, a much less toxic form of sulfur. The process is carried out in the crab’s hepatopancre­as, a tissue similar in function to the vertebrate liver.

We wondered whether, in the symbi­ont-containing animals, each host was colonized by only a single type of bac­terial symbiont.  Recent studies of the sequences of a particular ribonucleic acid (the 165 RNA) of bacterial ribo­somes, done by Daniel Distel and our group in collaboration with Norman R. Pace of Indiana University, indicate that the relation between a sulfur-me­tabolizing bacterium and its animal host is species-specific, that is, each host species harbors a unique strain of bacteria. In spite of the fact that many types of sulfur bacteria have entered into this kind of symbiosis, no one bac­terium seems to have adapted itself to more than one species of host. Pre­sumably endosymbiosis with sulfur bacteria originated independently and repeatedly in diverse animal groups.


Our discoveries at deep-sea vents have stimulated surveys of a wide range of sulfide-rich environments, including mangrove swamps, petrole­um seeps, sewage outfall zones and marshes. The surveys have revealed that the kinds of sulfur-based symbi­oses first discovered in the deep sea are widespread, and that various other energy-rich inorganic molecules, such as methane, can be exploited in analo­gous symbioses between other inver­tebrates and bacteria. These diverse symbioses have been described in dis­parate animal groups, including some smaller tube-worm relatives of Riftia, clams of a number of different fam­ilies, various mussels and several groups of small marine worms.

As our studies progress we expect find further instances in which animals worm rely on chemolithoautotrophic symbi­onts , thereby gaining the ability to tol­erate and exploit habitats the animals could not colonize successfully with out bacterial partners




SULFIDE IS HIGHLY TOXIC to most animals (a). It poisons respiration at two levels in the blood, where it binds to hemoglobin, and in cells, where it inhibits the respiratory to enzyme cytochrome c oxidase (not shown). Animals associated with sulfide-rich hydro- thermal vents have evolved different strategies to avoid sulfide poisoning. The tube Riftia pachyptila (b) has a separate binding site on its hemoglobin molecule for sulfide

and so can transport oxygen and sulfide simultaneously in its bloodstream. The vent clam onts (]alyptogena niagnifica (c) has a special transport protein to carry sulfide to bacteria in crate its gills. The vent crab Bythograca therinydron (d) lacks endosymbiotic bacteria; it is able -to detoxify sulfide by oxidizing it to nontoxic thiosulmate in its liverlike hepatopancrens..



PHOTOSYNTHESIS and chemosynthesis are compared. The energy sources differ, but the conversion process and end products are the same. In photosynthesis light is ab­sorbed by the chloroplasts of plants and drives carbon fixation by the Calvin cycle, a process yielding sugars, fats and amino acids that enter the food chain, passing from herbivores to carnivores. In chemosynthesis energy is provided by hydrogen sulfide issu­ing from vents in the ocean floor. It is taken up by free-living bacteria and also absorbed by vent animals such as the tube worm, which transport it to endosymbiotic bacteria. In the bacteria it is oxidized, providing energy for the Calvin cycle. The end products en­ter the food chain, passing directly from lower-order carnivores to higher.order ones.

SYMBIOSIS IN THE DEEP SEA/questions with reading sheet


1.  Why was the deep sea considered to have a low abundance of organisms?


2.  Where are the hydrothermal vents situated? (overall)


3.  What organisms were discovered near the vent by the crew of the Alvin?

4.  Why is photosynthesis impossible at the vent?

5.  What information did the water samples shed on the mystery of why life was around the vents?

6.  What do hot springs bacteria use for energy?

7.  Define chemolithoautroph.

8.  What product is produced by both plants and these bacteria?

9.  What is strange about the anamotical studies of Riftia?

10.  What is the trophosome responsible for?

11.  What did examination of the trophosome reveal?

12. What type of symbiotic relationship exists between the bacteria and worm?

13.  How do both parties of this relationship benefit?

14.  What effect did sulfide have on oxygen bonding in the worm?

15.  What 3 obstacles did Riftia have to overcome to carry on respiration?

16.  What is responsible for the deep red color of the bronchial plume?

17.  How is Riftia hemoglobin different than human hemoglobin?

18.  What happens to the sulfides after it is unloaded to the cells of the trophosome?

19.  What other vent animals are chemosynthetic endosymbionts?

20. Where is sulfide absorbed in the clam (Calyptogena)?

21.  What does the accelerated growth rates of the organisms around the vent indicate?

22.  How does the vent crab detoxify the sulfide it absorbs?