Sticky speciation

A limnetic male stickleback (top) would rather mate with a female of his own species than with the benthic female (bottom).

Courtesy of Ernie Cooper

Zoologist Jennifer Gow had a hunch. So she booted up her laptop, launched Google Earth, and zoomed in on Nelson Island, a small, largely unpopulated island along the British Columbia coastline. Gow, a postdoc at the University of British Columbia, had studied the handful of lakes where pairs of closely related fish species called threespine sticklebacks are found, and thought this region might have more. From the satellite images, one isolated, small lake on the island stood out. Little Quarry Lake sat at the right elevation above sea level. Its size, range and lack of navigable water connections matched other stickleback-containing lakes, too.

In June 2007, Gow, together with local freshwater ecologist Michael Jackson and professional guide John Dafoe, visited Nelson Island, where, says Gow, they "made a bee-line" for Little Quarry Lake. There, "lo and behold," they found exactly what they were looking for - the fifth - known threespine stickleback species pair, a rare coupling of two closely related but morphologically distinct forms that coexist in the same body of water.

Although threespine sticklebacks (Gasterosteus spp.) are commonly found throughout the northern hemisphere, species pairs exist in only a handful of coastal lakes, all along British Columbia's Strait of Georgia. Each species pair arose independently of one another an estimated 12,000 years ago as the glaciers of the last ice age retreated. In the melting glaciers' wakes, a marine stickleback was trapped behind, where it adapted to two separate freshwater niches and diversified into a benthic bottom-dweller and a limnetic open-water feeder.

Often compared to Darwin's finches, the stickleback species pairs provide a powerful and tractable comparative model to test natural selection's role in the speciation process. Although benthic and limnetic sticklebacks usually refuse to mate with each other, two benthics or two limnetics from different lakes that evolved in complete isolation for thousands of years will happily interbreed. "Repeatedly we see the same mechanism of reproductive isolation evolve in lock step with adaptation to their environment," says UBC evolutionary biologist Dolph Schluter. "Natural selection is the only process that can do that."

Scientists fear, however, that this so-called "natural laboratory system" doesn't have enough replicates, especially as two of the five known species pairs have already been effectively wiped out by introduced species. Gow's discovery - the first in more than two decades - is good news for conserving the critically endangered sticklebacks. But it's also a boon for biological research.

Soon after bringing the fish back to her UBC lab, Gow walked down the hallway to Schluter's office, Little Quarry Lake sticklebacks in hand. When Schluter saw the fish, he was immediately taken aback. "Oh my god, they sure looked like limnetics and benthics," he recalls. Using genetic markers and morphological features, Gow and Schluter showed that the Little Quarry Lake sticklebacks were, indeed, a true species pair (Can J Zool, 86:564-71, 2008).

Gow and Schluter also found a fairly unique characteristic in the Little Quarry Lake benthic species: The fish lacked a pelvic girdle, the bony modified pelvic fin that is the equivalent of tetrapod hind limbs in most sticklebacks. Schluter had found this only once before in the five other British Columbia species pairs.

In 2004, together with geneticist David Kingsley at Stanford University, Schluter showed that regulatory changes in the Pitx1 gene led to pelvic girdle loss in both the benthics from Paxton Lake on Texada Island, BC, and in another population of nonpaired sticklebacks from Iceland (Nature, 428:717-23, 2004). Could the same mutation be at work in the Little Quarry Lake fish?

Kingsley and Schluter have already teased apart the genetic basis of another post-glacial stickleback feature: the loss of numerous protective armor plates. In 2005, they showed that a single mutation lurking at low frequencies in the marine ancestor was responsible for the parallel evolution of armor plate reductions in dozens of freshwater stickleback populations around the world (Science, 307:1928-33, 2005).

But armor plate changes are as "common as cockroaches," says Kingsley. Since pelvic reductions are much rarer, they probably don't stem from ancient, preexisting mutations, or we would see fewer freshwater girdles by now, he adds. Rather, each mutation causing pelvic reductions probably arose anew, including in the Little Quarry Lake benthic species. "That makes every single population of great interest," says Kingsley.