Some river pollution spawns body-altering steroids
On the way back from a field trip to one of Alabama's barrier islands in the late 1970s, W. Mike Howell and the students in his vertebrate biology class detoured over the state line into Florida to collect some fish.
En route to Pensacola, they passed a shady stream, Elevenmile Creek. It was a searing hot day, so it seemed like a good idea to stop and cool off. The consequences of that detour continue to expand.
"Pulling off the interstate," Howell remembers, "we didn't have any earthly idea that this creek was polluted." But as the students climbed down its banks, the smell of chemicals hit them. "The water was dark," he notes, "but we assumed it was just due to the tannins and resins that you see in many swampy southern streams."
An ichthyologist at Samford University in Birmingham, Howell had been studying the genetics of mosquitofish (Gambusia holbrooki). During the Florida rest stop, he and a graduate student started to look for mosquitofish specimens in the creek's murky shallows.
With each dip of their nets, they harvested a generous haul of the roughly 2-inch-long fish. It didn't take long to realize that something was strange. Every mosquitofish looked to be a male, which the scientists recognized by its gonopodium--a thin, elongated anal fin that males use for copulation.
As Howell and his student began speculating where the females could be, the pair realized that one of the males looked pregnant. On its side, it even bore the black spot signaling pregnancy.
"This suggested many of the others might also be females," Howell recalls. Further investigation confirmed that suspicion and launched the scientist on a 23-year hunt to find out what had so convincingly masculinized the physique of every female mosquitofish in this stretch of the creek.
The answer came from investigations of the Fenholloway River, an even more polluted stream 225 miles from Elevenmile Creek. Outside Perry, just east of Florida's panhandle, the southern Fenholloway bears the same reek and coffee-colored stain that Howell's team had encountered at the creek. In both cases, the water's aromatic pollution traced to wastes from nearby pulp-and-paper mills.
In its upcoming paper in Environmental Toxicology & Chemistry, Howell's group reports finding trace contamination of the lower Fenholloway with androstenedione. Known among weight lifters as andro, this is the over-the-counter anabolic steroid that St. Louis Cardinals' slugger Mark McGuire took flack for consuming throughout his record-breaking 1999 home run streak.
The Fenholloway's andro, however, doesn't come out of any bottle but arises from the biotransformation of pollutants. It also represents the first environmental androgen--a pollutant that functions like a male sex hormone--ever discovered in nature, says John A. McLachlan, a toxicologist who pioneered studies on environmental hormones (SN: 7/3/93, p. 10).
In people, andro binds to receptors on cells and triggers the production of testosterone, the body's primary androgen. In the Fenholloway River and Elevenmile Creek, Howell now believes, andro helped give the female mosquitofish their masculine fins.
But he and other scientists worry that mosquitofish may be the proverbial canary in the coal mine--a delicate species that signals threats to others.
"Most vertebrates respond to hormones in pretty much the same way," observes Stephen A. Bortone, environmental science director of the Conservancy of Southwest Florida in Naples. "I'm not saying that humans would be masculinized" by the hormone doses fish are getting, he says. "But if an androgen affects fish one way, it will likely affect humans in a similar way."
Five years ago, Bortone, who has collaborated with the Samford team, reported data showing that adding androgens to water in the laboratory could masculinize female mosquitofish (SN: 7/15/95, p. 44). However, paper-mill effluent didn't contain andro, testosterone, or any other of the compounds that he had tested.
To discover what was transforming the fish, the Birmingham researchers turned to the espresso-colored lower span of the Fenholloway. It was an ideal site, notes Robert Angus of the University of Alabama at Birmingham, because the local mill uses so much of the upstream flow in its operations--at times, 100 percent--that "essentially the whole of the [lower] river is effluent."
A complex brew of pulp-derived chemicals, including cholesterol-like pine sterols, travels along the lower Fenholloway. Together with Angus, Ronald L. Jenkins of Samford ran various samples of the river through a test-tube assay for compounds that bind to the androgen receptor. That's how they found andro, which turned out to be present in the river at 15 nanograms per liter. That's similar to concentrations of various estrogens detected in undiluted sewage effluent from a Canadian site (SN: 6/17/00, p. 388).
The river's bottom proved a far richer source of andro than the water itself, so the scientists probed the sediment's chemistry. At an October 2000 environmental hormones symposium, hosted by the Center for Bioenvironmental Research (CBR) at Tulane and Xavier Universities in New Orleans, Jenkins and Angus reported finding at least three more compounds that bind to the androgen receptor. So far, they've identified only one: progesterone.
Though this female sex hormone is normally associated with pregnancy, it can have androgenlike effects by opposing the actions of the body's primary female-sex hormone, estrogen. Jenkins told Science News, that bacteria in the sediment seem to chemically alter plant sterols to make the androgens.
Water downstream of pulp-mill operations on the Fenholloway is strongly masculinizing, or androgenic, whereas water upstream of the plant and in a nearby river is not, Environmental Protection Agency scientists reported at a November meeting in Nashville of the Society of Environmental Toxicology and Chemistry (SETAC). Their findings confirm those of Jenkins and Angus. Gerald T. Ankley of the agency's Duluth, Minn., ecology lab is now working to identify the water's androgens.
Androstenedione alone wouldn't appear "sufficient to explain the degree of androgenicity that we see," observes Ankley's coworker L. Earl Gray Jr., an environmental-hormones toxicologist at EPA's laboratory in Research Triangle Park, N.C. He expects a soup of androgens to emerge in further studies.
The EPA team is collaborating with Edward F. Orlando and his colleagues at the University of Florida in Gainesville on studies of hundreds of female mosquitofish from the river. So far, Gray says, the Florida contingent has found at least some masculinization of all females collected downstream of pulp-mill operations.
As Howell noted in 1978, however, these females remain fertile.
Heather B. McNatt of the University of Alabama has been comparing pregnant females from the Fenholloway River with those in nearby Spring Creek, which is free of mill wastes. She's found that those at Spring Creek grow about two-thirds bigger than their counterparts in the Fenholloway. More importantly, Spring Creek moms produce bigger babies and more of them.
The researchers also have seen differences in Fenholloway dads. Though male mosquitofish are normally smaller than females, Howell notes that males in the lower Fenholloway are dramatically stunted--evidence of being "precociously masculinized," he says. In mosquitofish, once the testes mature, a male's testosterone production soars, a gonopodium develops, and then growth ceases.
Howell's group has found males little more than a half-inch long sporting gonopodia. "These very, very tiny males were adults," he notes.
Masculinizing American eels
From the mill-polluted rivers, Bortone and William P. Davis of the EPA in Gulf Breeze, Fla., have pulled up two other small species of fish that have been masculinized--the least killifish (Heterandria formosa) and sailfin molly (Poecilia latipinna). As with mosquitofish, effluent-exposed females develop fins resembling gonopodia.
Davis found that the masculinized female killifish also developed an unusual belligerence. Ordinarily, he notes, "I could raise these tiny fish in tight quarters, 10 to a jar." Once he doctors their water with a little androgen, however, the females suddenly "turn aggressive--like little sharks," he says.
American eels (Anguilla rostrata) migrating up the lower reaches of Elevenmile Creek from the Gulf of Mexico exhibit a different form of masculinization. Born at sea, eels enter the creek as 4-inch juveniles. Males remain at the base of the creek in brackish or salt water while females continue traveling up into freshwater.
Ordinarily, these fish would inhabit their respective portions of the creek for 4 to 10 years. Then, they'd develop gonads, change skin color, evolve huge eyes adapted for vision in the deep sea, and cease to eat--all changes in preparation for their return to the Caribbean for spawning. Yet studies 20 years ago in the lower Elevenmile Creek found that some males exposed to pulp-mill effluent were developing testes and big eyes and undergoing related changes several years early. Females returned from their upstream haunts unusually early, and underwent similar changes at the base of the creek.
This precocious maturation appeared to be signaling the tiny eels to migrate long before they have the size and energy stores needed to survive the long swim to their spawning grounds, says Davis.
These changes also can be triggered in the lab by adding androgens to the young eels' water. Presumably, Davis says, bacterial production of andro and related pollutants in the waterways foster the early maturation.
Elevenmile Creek's bluegill sunfish (Lepomis macrochirus) are also sensitive to pulp-mill pollutants, according to preliminary data that a Louisiana team reported at the CBR symposium. Henry L. Bart Jr. of Tulane University and his colleagues found that those fish living downstream of the mill are developing higher-than-normal androgen concentrations in their blood and growing unusually rapidly.
Pulp-mill residues may be a concern in other countries, as well. A Swedish study of eelpout (Zoarces viviparus) points to yet another masculinizing influence.
This eel-like marine fish--like mosquitofish, killifish, and sailfin mollies--bears its young live. Most eelpout broods contain a 50:50 mix of males and females, D.G. Joakim Larsson and his colleagues at Goteborg University found. However, the sex ratio at two Swedish coastal sites in 1997 and 1998 that were contaminated with pulp-mill wastes "was significantly shifted toward fewer females," the scientists report in the December 2000 Environmental Toxicology & Chemistry. One site produced 55 percent males, the other 58 percent.
In 1999, a brief plant shutdown occurred. All of a sudden, sex ratios for the fish returned to normal, reported Larsson at the SETAC meeting last November.
Androgens can trigger devastating changes in another hormone system important to aquatic life. Gerald A. LeBlanc of North Carolina State University in Raleigh has found in laboratory tests that androgens can derail the function of ecdysteroids--hormones that choreograph development in many invertebrates, including the molting of crustaceans. However, few natural waterways contain androgens at concentrations high enough to disrupt this development.
LeBlanc works with tiny crustaceans known as water fleas (Daphnia magna). Agents that perturb their ecdysteroid system can really "mess up embryonic development," he says. In some cases, the body will be grossly deformed, preventing if from swimming well. Other times, the young suffer what LeBlanc calls "embryonic arrest." They hatch and live briefly, even though they still "half resemble an egg," he says.
After attending the CBR symposium in October, LeBlanc became interested in testing samples of Fenholloway sediment to see if its androgens might disrupt development in water fleas there.
LeBlanc notes that some aquatic species throughout the world are experiencing yet another pollution-induced masculinization, known as imposex. Affecting various populations of some 100 species of marine snails, this condition saddles both genders with grossly exaggerated male sex organs.
The males go into a reproductive mode that never shuts off. Affected females can develop a huge penis that blocks their release of eggs. The resulting egg-engorged females eventually die.
The one agent now known to trigger this excessive masculinization is tributyl tin. It's not an androgen, but a toxic compound used to retard barnacle development on ship hulls. How this pollutant masculinizes the snails remains a mystery.
Some female fish change into males in response to local sex ratio or social cues. Scientists have been puzzling over those animals for decades (SN: 10/ 21/95, p. 266). The chemistry behind the switch has remained elusive. The Birmingham team's report on andro and related compounds now indicates that besides natural factors, pollution can have gender bending effects.
"This work is important and something we should pay attention to," says McLachlan, CBR's director.
For instance, he argues, andro's production by sediment-dwelling microbes bolsters "the potential for bacteria to be active players in any type of pollution," an idea to which he says few people have paid much attention.
He's also impressed by the growing list of hormonal signals that have been identified in the environment. Even 5 years ago, scientists recognized only a few instances of environmental hormones, all estrogens, McLachlan notes. More recently, agents that masquerade as melatonin (SN: 10/17/98, p. 252), anti-androgens (SN: 4/3/99, p. 213), and thyroid hormones have emerged (SN: 10/2/99, p. 212).
"I now suspect," he says, "that for every hormone that we know of, we will eventually find things in the environment that mimic or block it." McLachlan adds that by knowing what environmental hormones can do--even through such serendipitous findings as the macho mosquitofish--ecologists and health researchers can begin to look for subtle harm in other populations, including people.
Gerald T. Ankley
USEPA Environmental Effects Research Laboratory
Mid-Continent Ecology Division/ORD
6201 Congdon Boulevard
Duluth, MN 55804
Henry L. Bart
Museum of Natural History
Building A-3, Wild Boar Road
Belle Chasse, LA 70037
Stephen A. Bortone
The Conservancy of Southwest Florida
1450 Merrihue Drive
Naples, FL 34102-3449
Web site: [Go to]
William P. Davis
USEPA Environmental Effects Research Laboratory
Gulf Ecology Division/ORD
Sabi8ne Island Drive
Gulf Breeze, FL 32561-5299
L. Earl Gray
Research Triangle Park, NC 27711
Louis J. Guillette Jr.
Department of Zoology
University of Florida
223 Bartram Hall
P.O. Box 118525
Gainesville, FL 32611
W. Mike Howell
Department of Biology
800 Lakeshore Drive
Birmingham, AL 35229
Ronald L. Jenkins
Department of Biology
800 Lakeshore Drive
Birmingham, AL 35229
Marine Science Institute
University of Texas, Austin
750 Channel View Drive
Port Aransas, TX 78373
Department of Toxicology
North Carolina State University
1600 Blue Ridge Road
Reproductive Physiology Laboratory
Raleigh, NC 17695
John A. McLachlan
Tulane-Xavier Center for Bioenvironmental Research
1430 Tulane Avenue
New Orleans, LA 70112
Bortone, S.A., and W.P. Davis. 1994. Fish intersexuality as indicator of environmental stress. BioScience (March):165.
Raloff, J. 1999. Thyroid linked to some frog defects. Science News 156(Oct. 2):212.
______. 1999. Hormone mimics: New assessments air. Science News 156(Aug. 14):101.
______. 1999. Common pollutants undermine masculinity. Science News 155(April 3):213.
______. 1998. Does light have a dark side? Science News 154(Oct. 17):252.