Hybrids reveal the barriers to successful mating between species

It’s a tale as old as wine. Two organisms meet over a barrel of alcohol and decide to mate.

Geneticist Maitreya Dunham didn’t see it happen. But she has molecular evidence that two yeast species produced a hybrid in an old warehouse turned microbrewery. The two species had grown apart, evolutionarily speaking, about 10 million to 20 million years ago, Dunham, of the University of Washington in Seattle, and colleagues calculate. Yet the distant relatives interbred, producing the hybrid that ended up in a barrel of wild beer.

But wait a minute, some people may adamantly argue: Separate species can’t mate and reproduce. Much of the time, those hybrid deniers are right. But hybrids do happen, far more often than scientists used to think, says Molly Schumer, an evolutionary geneticist at Harvard University. Plants are famous cross-species breeders. Yeasts do it all the time, especially in the nutrient-rich microbe “meet market” that is a brewery, Dunham says. “It’s like a yeast hookup zone.”

Researchers have discovered fish, birds, mice, fruit flies and other animals in the wild carrying DNA from parents of different species. Perhaps the biggest shocker of all was the 2010 discovery that humans had interbred with Neandertals after leaving Africa. Humans still carry genetic souvenirs of the encounters (SN: 6/5/10, p. 5). These discoveries would seem to contradict the biological species concept, which holds that separate species can’t mate and produce fertile offspring.

But the blurred lines between species don’t bother many researchers, says evolutionary biologist Glenn-Peter Sætre of the University of Oslo. He and others are learning from that blurry zone, the place where hybrids thrive or sometimes fail. Finding out why some hybrids make it and others don’t may yield molecular details about how reproductive barriers between species are built, says evolutionary population geneticist Graham Coop of the University of California, Davis. Studying fully reproductively separated species is no good; their barriers are well established, and it’s difficult to know which bricks were laid first. With hybrids, researchers can watch the barriers being erected, Coop and others say. Those studies give researchers insight into the process of speciation, the separation of a population into different species.

Mingling fish

The swordtail fish Xiphophorus nezahualcoyotl inhabits certain rivers in Mexico (green shows range). The species is a hybrid of X. montezumae (blue) and X. cortezi (red), found in different rivers. The interbreeding between those species probably took place more than 2,500 generations ago, though swordtails in other rivers have hybridized more recently. 

Scientists have begun to build a list of “speciation genes” — different in different species — that help establish and maintain that parting of the ways. Despite the name, the job of those genes isn’t to draw lines in the sand between species. Those lines are a side effect of the genetic tweaks species gather by chance or while adapting to new environments.

Dead-end hybrids

When Schumer began working with species of swordtail fish in Mexican rivers about six years ago, scientists still thought hybrids were rare among animals. But the fishes’ DNA told another tale. Their genetic blueprints, or genomes, were patchworks of different species’ DNA, indicating interbreeding with at least one other relative in the recent past. And by recent, Schumer means that hybridization became the rage among two species of Mexican swordtails just within the last 20 to 30 years (SN Online: 5/21/14). “We weren’t clear at the time if [hybridization] was something weird about swordtails,” she says, “but it’s becoming clear that many species groups are that way.”

Sometimes, hybridization brings benefits. For example, the famous Galápagos Islands finches named for Charles Darwin picked up survival advantages by interbreeding with other finch species (SN: 3/7/15, p. 7). But more often, hybrids don’t happen.

Geographical barriers (mountains high enough, rivers wide enough) can prevent interspecies contact, as can out-of-sync mating cycles. Even when mating actually occurs, barriers still exist, says Michael Nachman, a population geneticist at the University of California, Berkeley. Fertility in hybrids is often subpar. Sperm from one species may not be able to fuse with eggs from another, and other molecular incompatibilities may cause embryos to fail. Sometimes being a hybrid is a death sentence. Those problems have genetic and physiological roots that scientists are only beginning to understand.

A hybrid combines the genomes of both parents, a process similar to combining parts from two machines built under slightly different measuring systems. “You can take two fully functional complex machines and put them together and the whole thing falls apart,” says evolutionary geneticist Nitin Phadnis. Scientists can learn how species form by studying the wreckage of such hybrid machinery. In 2009, Phadnis, now at the University of Utah in Salt Lake City, and H. Allen Orr of the University of Rochester in New York discovered, in fruit flies, one of the first known speciation genes.

Things don’t necessarily fall apart right away, says Polly Campbell, an evolutionary geneticist at Oklahoma State University in Stillwater. As two species spend time apart adapting to separate environments, many changes accumulate in the species’ DNA. Such changes, or mutations, may alter the function or the structure of proteins produced by the genes. In a first-generation hybrid, those changes may not be visible. Parent Species A’s genes produce cogs that fit its cellular machinery and Parent Species B’s genes do the same. The hybrid inherits components to assemble fully functional versions of both parents’ machinery. But when hybrids go on to breed with each other, their offspring inherit different combinations of the original parent species’ genes. Sometimes that works out fine: A small proportion of the next generation may inherit all A machinery or all B machinery. Another proportion of offspring may get a mix of A and B cogs, but might be able to cobble together a biological machine that works well enough. A third segment of the offspring won’t be so lucky: They will be stuck trying to fit Species A’s cogs in an otherwise Species B machine (or vice versa), like a square peg in a round hole. Over time, enough offspring can inherit unworkable combinations of pegs and holes that the hybrids die out.

Story continues after diagram

Right fit

Hybrids may inherit different versions of genes from each parent species. To be viable, a hybrid’s cellular machinery, for example, a protein and its receptor, must fit smoothly, as a peg in a hole. First-generation hybrids may be fine, but their offspring can inherit combinations of pegs and holes that either make healthy, fertile offspring (green), offspring that may or may not make their pegs and holes work (yellow) or defective offspring (red).  

Britannia Wanstrath has seen hybrid wreckage up-close. She is a technician who oversees mouse breeding and welfare at the University of North Carolina at Chapel Hill.

While trying to create better mouse strains for studying human diseases, she and other researchers may have inadvertently stumbled upon genes that render hybrids as dead ends. A massive effort known as the Collaborative Cross yielded 738 hybrid mouse lines by breeding an original eight strains of mice from three different subspecies, Mus musculus domesticus, M. musculus musculus and M. musculus castaneus (SN Online: 2/17/12). Those subspecies (or maybe species — the dividing line is fuzzy) are genetically similar. Because of small differences among the subspecies, researchers expected that a few of the new strains wouldn’t make it, says geneticist Fernando Pardo-Manuel de Villena, also at UNC Chapel Hill. “We expected some extinction, but very, very minor.”

Instead, 95 percent of the 738 hybrid Collaborative Cross lines have gone extinct, Pardo-Manuel de Villena and colleagues reported in the June issue of Genetics.

Story continues after graphic

Up to speed

Each color below represents DNA on chromosome 2 from an originating parent strain in the Collaborative Cross project, a large panel of inbred mice. Hybrid offspring with red in the black vertical box had DNA from one strain, called PWK/PhJ, and their sperm had faster back and forth motion than mice with DNA from other strains. Sperm motility may have been a problem for some of the strains that went extinct.

One of the mouse lines that expired, dubbed 5262, had particularly wild mice that were “difficult to handle and very vocal when agitated,” Wanstrath says. As researchers began breeding relatives in the line to each other so they would all be genetically identical, 5262’s breeding slowed. The inbred litters died soon after birth. The researchers tried everything they could think of to keep the line going, Wanstrath says. “Nothing worked.” Eventually, the last litter was born: one male and one female pup survived. The researchers were excited that they might be able to keep the line going. But the female died 12 days after birth and her brother was left alone. End of the line.

The researchers examined male mice from 347 of the extinct lines to find out what was going on. About 47 percent of the male mice in the extinct lines were infertile because of gene defects that prevented them from making good sperm. Some made no sperm. Some produced sperm with broken tails, unable to swim. Of the 183 male lines that were fertile, 99 could produce offspring only with distantly related females.

The scale of that extinction was unexpected, Campbell says, but male fertility problems aren’t a big surprise. Sperm production is a fragile process, she says. Still, those problems didn’t show up right away. It took generations of inbreeding to doom some hybrids. In the case of mice from the extinct lines, each round of inbreeding distributed the pegs and holes until the pieces no longer fit. In other lines, the sorting made workable combinations.

The barrier genes

UC Berkeley’s Nachman agrees that it’s not easy to figure out which of the thousands of genes in a hybrid’s genome is a speciation gene. He should know. “I’ve been searching for the last 10 years, and I’ve yet to find one,” he says.

In fact, in mammals, only one speciation gene has been identified so far. Geneticist Jiří Forejt of the Czech Academy of Sciences’ Institute of Molecular Genetics in Prague wasn’t looking for it when he caught wild mice and bred them with lab mice to study diversity of immune system genes. Hybrid males from those liaisons were sterile, he and colleagues discovered. “We had no idea we were working with two subspecies, musculus and domesticus,” he says. In 1974, Forejt narrowed the problem to a gene on one chromosome, then finally revealed the gene’s identity in 2009: PRDM9, which produces a protein that determines where on chromosomes genetic information gets swapped.

When making sperm and eggs, organisms halve the number of chromosomes in those cells in a process called meiosis. That halving is necessary so that when an egg and sperm meet in fertilization, the embryo will have the correct number of chromosomes: half from the mother and half from the father. Cells going through meiosis must pass certain checkpoints. One of the biggest involves pairing the chromosomes to exchange bits with each other. That recombination is an important force in the evolution of sexually reproducing organisms because it allows for new combinations of genes.

Story continues after images

In many mammals, including humans and mice, the PRDM9 protein marks where recombination will take place — the molecular equivalent of the orange flags that utility crews use to mark digging locations. PRDM9 grabs DNA using structures called zinc fingers. During evolution, those fingers develop a touch for different DNA sequences in different species. The sequence grabbed by M. musculus musculus’ PRDM9 zinc fingers is slightly different from the ones preferred by M. musculus domesticus’ zinc fingers. And so, the two subspecies carry out their DNA swaps at different places along the chromosomes. Mismatched recombination sites can hold up egg and sperm production, leaving hybrids infertile.

Details about how important it is to have the right zinc fingers came in a report last year in Nature. By re-engineering PRDM9’s zinc fingers, researchers moved recombination hot spots and restored fertility in male hybrid mice.

But PRDM9 can’t take all the blame for hybrid sterility between those mouse subspecies. “In many cases it’s combinations of many genes that result in this failure,” Forejt says. He and colleagues reported last year in PLOS Genetics that they had tracked another speciation gene that interacts with PRDM9 to a stretch of 4 million DNA bases on the X chromosome. There are at least six other genes in that part of the chromosome, and Forejt doesn’t yet know which one is a speciation gene.

Sometimes, PRDM9 plays no role in incompatibility. Many animals, including dogs, birds, crocodiles and amphibians, don’t have a working version of the gene, or don’t use it.

Story continues after graph

Course correction

Good chromosome pairing happens less often in infertile and semifertile hybrids than in fertile hybrids. Re-engineering the version of PRDM9 in infertile and semifertile hybrid mice corrected chromosome pairing and restored fertility (bottom two rows).

Success for sparrows

Some hybrids, including some of Schumer’s fish and Darwin’s finches, have overcome whatever barriers might have been in the way of hybridization. These organisms have sorted out their parents’ differences and formed viable species of their own. Oslo’s Sætre and colleagues are studying one such species, the Italian sparrow, a blend of Spanish sparrows and house sparrows.

The bird’s origin story starts in the Middle East, where one of its parents, the house sparrow (Passer domesticus), is native. In the last 10,000 years, house sparrows accompanied early farmers on migrations into Europe. There, the house sparrows encountered Spanish sparrows (Passer hispaniolensis), found in Europe and northern Africa. Mating between the species produced the Italian sparrow (Passer italiae), which lives on the Italian mainland and a few Mediterranean islands.

Italian sparrows aren’t simple 1-to-1 mixtures of their parent species’ genes, Sætre and colleagues discovered. On average, 61.9 percent of the Italian sparrow’s DNA comes from house sparrows and 38.1 percent from Spanish sparrows, the researchers reported June 14 in Science Advances.

“The genomes of these species have combined, but they’ve also been sorted,” Sætre says, yielding those unequal proportions. In some parts of the Italian sparrow’s genome, house sparrow genes have been purged, leaving only Spanish sparrow DNA. In other sections, the Spanish sparrow contribution got the heave-ho. Such genetic housecleaning was probably necessary to get a mix of genes that could work together.

Today, Italian sparrows are largely cut off from their parents in the reproductive arena, Sætre says. In southern Italy where Italian and Spanish sparrows cross paths, Sætre and colleagues have tested more than 1,000 sparrows across several studies. Not a single one was a Spanish-Italian first-generation hybrid, indicating that impediments to breeding between the two species are high. In the Alps, Italian and house sparrows can sometimes breed, though genetic evidence suggests they rarely do. That reproductive isolation from its parents gives the Italian sparrow independent species status.

Unequal partnership

When house sparrows migrated from the
Middle East to Europe (geographic range, blue), they mated with Spanish sparrows (red) to form Italian sparrows (yellow).

DNA analysis (below) shows that Italian sparrows have purged genes from both parents. Their genomes average 61.9 percent house sparrow and 38.1 percent Spanish sparrow.

In the Italian sparrow, some genes have been tweaked from versions found in the parent species. These altered stretches of DNA are speciation gene suspects, but how such gene tweaks block reproduction isn’t known. “We’re not in a place where we can say what goes wrong biochemically,” Sætre says.

Checkpoint discoveries

Few researchers can point to a particular molecular wrench in the works that makes hybrids inviable, Phadnis says. “This is still cutting-edge science and an unsolved problem.”

In the few cases in which researchers have a handle on which genes are making hybrids sterile, there’s no guarantee the same genes or processes are involved in every failed species mash-up, he says. But he and colleagues are exploring a notion of what might be going wrong for some hybrids.

Phadnis and colleagues have proposed a solution for an almost 100-year-old question about why crossing Drosophila melanogaster fruit flies with flies from its sister species Drosophila simulans results in dead male offspring. Researchers had already discovered that D. melanogaster has a gene called Hmr that’s involved in divvying up chromosomes. Hmr doesn’t play well with Lhr, a D. simulans gene that also helps make sure chromosomes are doled out properly. Scientists knew a third gene was involved in species’ incompatibility, but researchers had technical difficulties identifying it. Phadnis and colleagues reported in Science in 2015 that the gene called Su(Kpn) encodes a checkpoint protein, one that determines whether a cell has completed certain tasks and can go on to divide.

Some of the molecular details are unknown, but Phadnis and colleagues propose that discrepancies between Hmr and Lhr may mess with the way cells divide their chromosomes. Because of the messed-up chromosomes, Su(Kpn) halts further cell division, arresting development and causing death of male offspring only, the researchers propose.

Arrested cell division may also be a problem for lab-made hybrids of two subspecies of Drosophila pseudoobscura on their way to becoming separate species. About 150,000 to 230,000 years ago, two subspecies — USA and Bogota — started to drift apart. Now, mating Bogota females and USA males produces infertile male hybrid offspring. The hybrid males can become weakly fertile in old age, but they produce mostly daughters. In 2009, Phadnis and Orr described in Science what was happening. The problem is selfishness. Specifically, a gene called Overdrive acts as a “selfish element” and skews sperm production so that hybrid males mainly make sperm carrying X chromosomes. That leaves only female offspring.

But Overdrive doesn’t work alone. At least six other genes are involved in the male sterility, Phadnis reported in 2011 in Genetics. As the USA and Bogota subspecies spend more and more time apart, they may layer on additional barriers. In unpublished work, Phadnis has discovered that Overdrive may also be putting the brakes on development, just as Su(Kpn) does. Its stopping power gets weaker with age, allowing some sperm production. Phadnis doesn’t know all the details yet, but then neither does anyone else.

“Despite a lot of effort, there really isn’t a single system in which you can tell the complete details of any hybrid incompatibility,” Phadnis says.

Even the yeasts that hybridize so readily in Maitreya Dunham’s beer barrel and that she intentionally breeds in her lab don’t have the whole hybridization thing worked out. Hybrid yeast often have fertility problems when reproducing sexually. Luckily for them, yeast can reproduce asexually, essentially making clones of themselves forever, Dunham says. “It doesn’t matter how messed up your genome is if you can make clones of yourself.”

This article appears in the November 11, 2017 issue of Science News with the headline, "Hybrids tell tales: When species manage to mix, they offer clues to reproductive barriers."