A Blog by

Placenta Evolution and a Sexual Cold War

With their impressive fins and stunning colours, the poeciliids—a group of small fish that includes guppies, mollies and swordtails—are understandably popular in aquariums. Some have beautiful fan-shaped tails that look like flamenco dresses. Others resemble Kandinsky paintings given life.

But some poeciliids are rare in aquaria, because they are relatively drab—silver-and-black oddities in a family known for extravagance. They also tend to share another weird and less obvious trait: they have placentas.

Unlike most fish, which lay eggs, all poeciliids give birth to live young. Mothers nurture their offspring inside their own bodies. Some produce eggs, but keep them inside their ovaries until the young are ready to enter the world. Others have evolved organs that bring the mother’s tissues so close to her baby’s that she can pass nutrients over—in other words, a placenta. Anatomically, this organ is very different to the placentas that human mothers use to nourish their babies, but it does the same job.

Mammals only evolved placentas once. But the poecilids have evolved these organs on at least eight separate occasions, and in a very short span of time. “The placenta is a very complex organ. Imagine if the eye evolved several times in the hominids. It’s that kind of complexity,” says Bart Pollux from the University of California, Riverside.

Now, Pollux has shown that the rise of the placenta was accompanied by drastic changes in the bodies and lifestyles of the poeciliids. The species with these organs are less ornate. They lack obvious courtship rituals. The males tend to be smaller, but their genitals are bigger. And he thinks that this seemingly unrelated constellation of traits arose because the placenta radically changes the relationship between mothers and their developing young.

Consider an egg-laying species. A female fish loads her eggs with nutritious yolk before they are fertilised. At this point, her investment is set. The most important decision she can now make is to choose a good mate, so her eggs get fertilised by the best possible sperm. In this scenario, the battle of the sexes plays out through discerning females and competitive males. You see bright colours, flashy courtship displays, and perhaps large size differences between the sexes.

In species with placentas, things are different. Mothers continue to provide nutrients to their young long after fertilisation. That opens up a new type of conflict between mother and child. The developing embryo does best if it gets as much nutrition from mum as possible. The mother, however, may do badly at raising future children if she invests too much energy in her current one. So in these species, evolution should drive embryos to want more and mums to hold back.

What about fathers? If the species in question doesn’t form life-long partnerships, a father does best if his mate invests heavily in her current child (which he sired) than in future children (which he won’t). And he can influence her investment because his genes are inside the embryo that’s growing within her. These paternal genes act against her maternal ones in an evolutionary tug-of-war with the baby as the rope.

So, what looks like a conflict between mother and offspring is really still a conflict between male and female. But instead of a showy battle of courtship rituals and bright colours, it’s a covert one that takes place in the womb. It’s a Cold War of the sexes. And that is why the placental species evolve away from showy, flashy traits towards sneaky, understated ones.

Pollux first came across this hypothesis in a paper by David and Jeanne Zeh, published in 2000. “They didn’t know how to test these predictions, but we found a way,” he says. He realised that he needed a large family of animals that have a variety of sexual traits and behaviours, and have evolved placentas over and over again. The poeciliids were perfect.

By studying 110 species, his team showed that the placental ones were less likely to have exaggerated traits like sword tails, wavy fins on their backs, bright colours, or elegant courtship. By contrast, they were more likely to have smaller males and relatively longer genitals. They have moved away from obvious displays that help females to choose mates, and towards traits that make it easier for males to sneak up on females and stealthily mate with them.

The team showed that almost all the poeciliids with placentas have an unusual ability: they can get pregnant while they’re already pregnant. This talent is called superfetation, and it stops any single male from monopolizing an entire litter by fertilising every available egg. Rather than being careful about choosing mates, a female can now mate with many males and let their genes duke it out for investment in the womb.

“It all fit!” says Pollux. All the traits he saw pointed to the same shift from choices and competitions that take place before sex, to subtler battles that are waged after it. Zeh, for one, is pleased to see evidence that supports her hypothesis. “The results of this study fundamentally advance our understanding of sexual selection,” she says.

Of course, this leaves a chicken-and-egg problem. Did the placenta drive the evolution of these other traits, or did the other traits drive the evolution of the placenta? The authors think it’s the former—that’s certainly what the title of the paper implies.

But David Haig from Harvard University, who has studied parent-offspring conflict, thinks it was the other way round. He thinks the first change was a move away from formal courtship and towards forced mating. Now, the males, rather than competing for a female’s attention are competing inside her via their sperm. Haig sees superfetation in the context of this sperm competition, as an adaptation that allows sperm to “get in first” and fertilise eggs before they are fully loaded with yolk. The female continues to add nutritious yolk to what is now an embryo, paving the way for the evolution of a placenta.

Reference: Pollux, Meredith, Springer & Reznick. 2014. The evolution of the placenta drives a shift in sexual selection in livebearing fish. Nature http://dx.doi.org/10.1038/nature13451

More on poecilids:

11 thoughts on “Placenta Evolution and a Sexual Cold War

  1. What is the exact reason for the claim the placenta like organs in poeciliids developed at least eight different times?

  2. Independent originations of a trait are typically identified because the trait appears in the lineage after the last common ancestor of that organism. So the fish that do have placentas likely had ancestors without placentas, but all of the placenta-having ones did not come from the same single ancestor. In this case it seems like 8 different lineages of fish gained placentas from different ancestors.

  3. So much I could even have gathered without asking.

    I could reformulate:

    a) what says these have common ancestors with poeciliids without placentas;
    b) and no one common ancestor since then that did develop one?

  4. I was not very much in doubt that would be the area where you got the arguments from.

    I was more interested in knowing how they look concretely.

    Such two poeciliids with placenta have a common ancestor, because …?

    That common ancestor lacked placenta, because ….?

    There was no more recent common ancestor developing placenta for both, because …?

  5. Does either placental support or differences in males have to come first? Is it possible these are pleiotropic traits (controlled by the same gene or linked set of genes), or perhaps one of the male differences was pleiotropic and the others developed later? Or is the genetic basis of each already known? Otherwise, another possibility is that the placentas may be due to several different genes (epistatic) and could have developed gradually (originally just boosting the egg early on, then more and more replacing the yolk) at the same time as the male traits changed.
    Seems like there are other possibilities we can’t really rule out — if the placentas could originate independently, maybe the fish had independent origins entirely, or maybe they came from a population with mixed traits which broke up into specialized species…?

  6. Hans,
    If I may attempt an answer,

    Such two poeciliids with placenta have a common ancestor, because …?….Molecular and morphological analysis like Haley said.

    That common ancestor lacked placenta, because ….?…We know that the placenta is an exceptional trait in poecillids (uncommon) and in fish in general. Therefore, given a poeciliid phylogeny, it is more parsimonious to assume that the placenta evolved independently 8 times then the inverse situation, a common ancestor evolved the placenta and subsequently all non-placental species (the majority) lost the placenta. Yes we do not “know” which scenario was true, but given parsimony and what we know about how complex traits evolve we can reasonably state which scenario is most likely to be true.

  7. The problem with your explanation is: the morphological analysis of the poeciliids with placenta leading to them having a common ancestor, would it not also thereby include the very notorious fact they both have placentas, especially as it is common for fish not to have such?

  8. No that is likely not an issue. I suppose it was misleading for me to even include morphological analysis, considering it is most certainly due to molecular analysis (on DNA sequence data) that is providing the Poeciliid phylogeny. Although it certainly still has its merits, analysis of morphology is not usually relied upon to provide accurate phylogenies nowadays, it can be misleading like in the example you put forth. Analysis of sequence data from poeciliids gives us our best guess of the true phylogeny and allows us to infer where and when morphological traits evolved. Hope that helps.

  9. Ah, you admit it is a guess?


    One thing about the guess, would a molecular reason be that:

    * poeciliids with placentas have, like mammals, immune systems eliminating foeti with way too many chromosomes
    * and that some poeciliids with placentas have more chromosomes than would be expected from the guessed common ancestor?

    Or are the genetic encodings for placentas just too different, using no common loci?

  10. Mr. Lundahl — I’m not sure what you have invested in your arguments here, but you’re acting like a troll — you might want to sit down and read the paper. If it’s over your head [or if your answer MUST be right], take it that this paper passed peer review for a reason.

Leave a Reply

Your email address will not be published. Required fields are marked *