When we visit a zoo and peer at our closest living relatives, the great apes, two things reliably captivate us. One: they look so very much like people. The expressive faces and the grasping hands of chimpanzees, bonobos, orangutans and gorillas are eerily similar to our own. The other: these creatures are so clearly not us. Our upright walking, capacious and clever brains, and a list of other traits sharply set us apart. What were the key defining events in evolution that make us uniquely human? Why did they happen—and how? Anthropologists and evolutionary biologists have toiled at such questions for decades and increasingly are turning to modern genetic technologies to help crack the mystery. We have found that some of the most important human characteristics—features that set us apart from our closest relatives—may have come not from additions to our genes, as one might expect. Instead they have come out of losses: the disappearance of key stretches of DNA. Several research laboratories, including mine, have traced some of this lost DNA across time, comparing human genomes with those of other mammals and even archaic humans: the Neandertals and our lesser known cousins, the Denisovans. We have learned that during the roughly eight million years since our lineage split from chimps, our ancestors’ genomes were stripped of DNA “switches” that activate key genes during development. Neandertals share our loss, making it clear the vanishing act occurred early in our evolutionary path. In fact, loss of these DNA sequences appears to be linked to quintessentially human traits: big brains, upright walking and our distinctive mating habits. (The last part of the project led me, in the course of my experiments, to learn a surprising amount about the structure of primate penises.) Losers I first developed a keen interest in human evolution during my Ph.D. years with noted anthropologist C. Owen Lovejoy of Kent State University, where I studied the difference in skeletons of males and females in extinct human relatives. I wanted to continue this kind of work to learn what, in our genes and developmental processes, had changed as humans progressed along our unusual evolutionary path. I was fortunate to obtain a postdoctoral position with David Kingsley of Stanford University, who was bearing down on just the kind of questions that fascinated me. Among other work, Kingsley’s lab had identified DNA changes involved in the evolution of stickleback fishes—including the deletion of a stretch of DNA in freshwater sticklebacks that, it turned out, caused the spiny pelvic fins to be lost in those species. That lost DNA piece contained a “switch” that was needed to activate a gene involved in pelvic spine development, at the right time and place. If this kind of process had happened in sticklebacks, why not in human beings, too? It seemed reasonable to suppose that subtle changes in when and where genes are turned on during development might be one way our genome had evolved to generate our unique anatomy. Inspired by that fishy example, we set out to see if we could find important switches that had disappeared in human beings over evolutionary time. Today’s availability of completely sequenced human and ape genomes, as well as the computational tools needed to analyze them, made our experiments possible. A group of us in Kingsley’s lab teamed up with Stanford computational scientist Gill Bejerano and then graduate student Cory McLean to plan the experiments.
Credit: Jen Christiansen
Finding missing switches is not easy, because genomes are vast. Ours contains 3.2 billion bases (the individual letters of a DNA sequence), and about 100 million of these differ between humans and chimps. How could our experiment be done? To understand our approach, a bit of background is in order. We know that in a creature’s genome, stretches of DNA that are doing important jobs are preserved during evolution with high fidelity. We also know that the more closely related two species are, the more similar their genetic sequences will be. In the case of chimps and humans, for example, our genomes are 99 percent identical in the tiny portion—less than 1 percent—that carries instructions for making proteins. And they are 96 percent identical in the much larger portion of the genome that does not contain these protein-coding genes. Searching the Junk Pile We were interested in this much larger area—stretches that, in the past, were written off as “junk” DNA but are now known to be stuffed with switches that turn genes on and off. The work of these switches is crucial. Although pretty much all human body cells contain the same 20,000 or so genes, they are not all turned on everywhere or at all times and places. Only certain genes are needed to build a brain, for example, and others for bones or hair. Because chimps and humans, despite their differences, have the same basic bodily structure, it is not surprising that the vast, switch-containing terrain in our genomes has a lot of similarities. The differences were what mattered to us. Specifically, we wanted to find sequences that had been preserved across evolutionary time in many species (indicating that the sequences were important) but were no longer present in humans. To do this, our computational genomics collaborators first compared the chimpanzee, macaque and mouse genomes. They pinpointed hundreds of DNA chunks that remained nearly unchanged among all three species. The next step was to scour this list to find chunks that did not exist in the human genome and thus had been lost sometime after our lineage diverged from the chimp’s. We found more than 500. Which of them to study? Because we wanted to find switches that might alter mammalian development, we focused on deletions near genes with known roles in that process. One of my colleagues pursued a deletion near a gene that regulates formation of neurons; another worked on a deletion near a gene involved in skeletal formation. For my part, because of my interest in the evolution of the differences in male and female body forms, I was excited by a deletion near the gene for the androgen receptor. Androgens such as testosterone are hormones needed for the development of male-specific traits. Made in the testes, they circulate through the bloodstream. In response, cells that actively make androgen receptors will then follow a male pattern of development: formation of a penis instead of a clitoris, for example, or (later in life) beard growth and an enlarged larynx for a deep voice. We needed, first, to test if those chunks of DNA really contained on-switches. To do this, we extracted them from both chimp and mouse DNA and attached them to a gene that turns cells blue—but only when that gene is activated. We injected this stitched-together piece of DNA into fertilized mouse eggs to see if any parts of the embryos were blue as they developed—indicating a functional switch in the piece of DNA—and, if so, where. Male Turnoffs My results were exciting: they really seemed to show that I was working with a true on-switch for the androgen receptor, one that human beings had shed. In mouse embryos, the genital tubercle (which develops into either a clitoris or penis) stained blue, as did the developing mammary glands and spots on the mouse face where sensory whiskers called vibrissae form. All these tissues are known to make the androgen receptor respond to testosterone. Looking more closely, I saw that the staining on the developing genitals was situated in places where small, tough protein spikes later form on the mouse penis. Neither sensory whiskers or spiny penises are human features, of course. But they occur in many mammals, including mice, monkeys and chimps. It is also known that a loss of testosterone results in shorter whiskers in male rodents and a lack of penile spines in rodents and primates. Penile spines and whiskers might similarly disappear if a crucial DNA switch were lost and the androgen receptor were no longer made in these tissues.
MOUSE TEST: To see what a genetic switch did, scientists injected it into mouse embryos with DNA that turned cells blue where the switch was “on.” Blue showed up in spots that develop into sensory whiskers and other hair follicles (1) and cells that form the penis or clitoris and mammary glands (2). Other techniques showed the switch is very near a gene that lets cells respond to sex hormones such as testosterone (3), and in adult male mice the switch is very active in cells that create penile spines (4). Credit: Courtesy of Philip L. Reno
As I pursued my experiments, others were busy with their own deletions of choice, with intriguing results as well. Then graduate student Alex Pollen found that his chunk of DNA activated the neural gene it was near, at precise spots in the developing brain. The gene is involved in a key process: it helps to kill off surplus neurons, which are overproduced during embryonic development. That offers a tantalizing thought: because the human brain is far larger than the chimp’s (1,400 versus 400 cubic centimeters), might loss of this switch have contributed to that evolutionary ramp-up, by releasing brakes on brain growth? Vahan B. Indjeian, then a postdoc in the lab, similarly found that his switch turned on the gene involved in skeletal growth—in developing hind limbs, specifically the toes of the foot. Toes two through five in humans are shorter than in apes and mice, alterations that improve the foot for upright walking. It is easy to see how brain and bone switches fit into the pattern of human evolution. Loss of both appears linked to hallmarks of humanity: a big brain and walking on two legs. Loss of sensory whiskers is fairly easy to rationalize because we no longer root around in the dark with snouts to grub out food or capture prey but use hands, in daylight, to find nourishment. Despite their reduced importance, though, it is unclear how we would be better off without these whiskers. Sensitive Relationships The penile spine story is less intuitive, but it is potentially more powerful and also fits neatly into the adaptive history of our species. Loss of spines, we believe, is one of a suite of changes that together had far-reaching effects on our evolutionary path. Together these changes altered the ways we mate, the physical appearance of males and females, our relationships with one another and the ways we care for offspring. Made of keratin, the same stuff as our fingernails, these spines occur in many mammals, including primates, rodents, cats, bats and opossums, and range from simple microscopic cones to large barbs and multipronged spikes. They may serve varied functions depending on the species: heightening stimulation, inducing ovulation, removing sperm deposited by other males, or irritating the vaginal lining to limit female interest in mating with others. The copulation time of spine-sporting primates is remarkably brief: in the chimp, typically less than 10 seconds. And historical experiments in primates show that removal of penile spines can extend copulation by two thirds. From such observations we can surmise that loss of penile spines was one of the changes in humans that have made the sex act last longer, and thus be more intimate, compared with that of our spine-bearing forebears. That sounds pleasant, but it could also serve our species from an evolutionary perspective. Our own reproductive strategy is unlike that of any apes, which all have intense male-male competition at their core. In chimps and bonobos, males compete to mate with as many fertile females as possible. They produce copious quantities of sperm (chimp testicles are three times larger than human ones), have penile spines and, like all male great apes and monkeys, have deadly, fanglike canines to discourage rivals. They leave rearing of offspring entirely to the female. Thus, for her, successful mating results in considerable commitment—gestating, nursing and rearing each infant to independence—and the female does not reproduce again until the weaning is completed. Humans are different. They form fairly faithful pair bonds. Men often help to rear offspring, enabling earlier weaning and increasing reproductive rate. Male-male competition is not as intense. We believe that loss of penile spines went along with loss of other traits associated with fierce competition (such as dangerous canines) and gain of others that promote bonding and cooperation. Bipedalism, as Lovejoy proposed, could be one of these features. Early male help probably initially took the form of procuring foods rich in fat and proteins, such as grubs, insects and small vertebrates, that required extensive search and transport. Males would need to travel far with hands free for carrying, which likely provided the initial selective advantage for walking on two legs. Gene Loss and Feature Gains And there is more. Cooperation and provisioning would also allow parents to rear dependent offspring for longer and thus lengthen the juvenile period after weaning. This would offer a longer time for learning and therefore enhance the usefulness of a large, agile brain—indeed, perhaps set the stage for its evolution. In that sense, the individual stories of all three of our deletions are deeply intertwined. When I came to Kingsley’s lab, I did not anticipate the turn my work would take—that I would find myself poring over fusty 1940s texts on mammalian genital structure. My lab is continuing research into this and other genetic and developmental changes with big consequences: the evolutionary shaping of the delicate bones in the human wrist to perfect them for toolmaking. There is much we may never know about all this distant history, no matter how keen we may be to find out. But even if we cannot be sure about the why of an evolutionary change, with the tools of modern molecular biology we can now tackle the how—a critical and fascinating question in its own right.
The other: these creatures are so clearly not us. Our upright walking, capacious and clever brains, and a list of other traits sharply set us apart. What were the key defining events in evolution that make us uniquely human? Why did they happen—and how? Anthropologists and evolutionary biologists have toiled at such questions for decades and increasingly are turning to modern genetic technologies to help crack the mystery. We have found that some of the most important human characteristics—features that set us apart from our closest relatives—may have come not from additions to our genes, as one might expect. Instead they have come out of losses: the disappearance of key stretches of DNA.
Several research laboratories, including mine, have traced some of this lost DNA across time, comparing human genomes with those of other mammals and even archaic humans: the Neandertals and our lesser known cousins, the Denisovans. We have learned that during the roughly eight million years since our lineage split from chimps, our ancestors’ genomes were stripped of DNA “switches” that activate key genes during development. Neandertals share our loss, making it clear the vanishing act occurred early in our evolutionary path.
In fact, loss of these DNA sequences appears to be linked to quintessentially human traits: big brains, upright walking and our distinctive mating habits. (The last part of the project led me, in the course of my experiments, to learn a surprising amount about the structure of primate penises.)
Losers
I first developed a keen interest in human evolution during my Ph.D. years with noted anthropologist C. Owen Lovejoy of Kent State University, where I studied the difference in skeletons of males and females in extinct human relatives. I wanted to continue this kind of work to learn what, in our genes and developmental processes, had changed as humans progressed along our unusual evolutionary path. I was fortunate to obtain a postdoctoral position with David Kingsley of Stanford University, who was bearing down on just the kind of questions that fascinated me.
Among other work, Kingsley’s lab had identified DNA changes involved in the evolution of stickleback fishes—including the deletion of a stretch of DNA in freshwater sticklebacks that, it turned out, caused the spiny pelvic fins to be lost in those species. That lost DNA piece contained a “switch” that was needed to activate a gene involved in pelvic spine development, at the right time and place.
If this kind of process had happened in sticklebacks, why not in human beings, too? It seemed reasonable to suppose that subtle changes in when and where genes are turned on during development might be one way our genome had evolved to generate our unique anatomy.
Inspired by that fishy example, we set out to see if we could find important switches that had disappeared in human beings over evolutionary time. Today’s availability of completely sequenced human and ape genomes, as well as the computational tools needed to analyze them, made our experiments possible. A group of us in Kingsley’s lab teamed up with Stanford computational scientist Gill Bejerano and then graduate student Cory McLean to plan the experiments.
Finding missing switches is not easy, because genomes are vast. Ours contains 3.2 billion bases (the individual letters of a DNA sequence), and about 100 million of these differ between humans and chimps. How could our experiment be done? To understand our approach, a bit of background is in order.
We know that in a creature’s genome, stretches of DNA that are doing important jobs are preserved during evolution with high fidelity. We also know that the more closely related two species are, the more similar their genetic sequences will be. In the case of chimps and humans, for example, our genomes are 99 percent identical in the tiny portion—less than 1 percent—that carries instructions for making proteins. And they are 96 percent identical in the much larger portion of the genome that does not contain these protein-coding genes.
Searching the Junk Pile
We were interested in this much larger area—stretches that, in the past, were written off as “junk” DNA but are now known to be stuffed with switches that turn genes on and off. The work of these switches is crucial. Although pretty much all human body cells contain the same 20,000 or so genes, they are not all turned on everywhere or at all times and places. Only certain genes are needed to build a brain, for example, and others for bones or hair. Because chimps and humans, despite their differences, have the same basic bodily structure, it is not surprising that the vast, switch-containing terrain in our genomes has a lot of similarities.
The differences were what mattered to us. Specifically, we wanted to find sequences that had been preserved across evolutionary time in many species (indicating that the sequences were important) but were no longer present in humans. To do this, our computational genomics collaborators first compared the chimpanzee, macaque and mouse genomes. They pinpointed hundreds of DNA chunks that remained nearly unchanged among all three species. The next step was to scour this list to find chunks that did not exist in the human genome and thus had been lost sometime after our lineage diverged from the chimp’s. We found more than 500.
Which of them to study? Because we wanted to find switches that might alter mammalian development, we focused on deletions near genes with known roles in that process. One of my colleagues pursued a deletion near a gene that regulates formation of neurons; another worked on a deletion near a gene involved in skeletal formation.
For my part, because of my interest in the evolution of the differences in male and female body forms, I was excited by a deletion near the gene for the androgen receptor. Androgens such as testosterone are hormones needed for the development of male-specific traits. Made in the testes, they circulate through the bloodstream. In response, cells that actively make androgen receptors will then follow a male pattern of development: formation of a penis instead of a clitoris, for example, or (later in life) beard growth and an enlarged larynx for a deep voice.
We needed, first, to test if those chunks of DNA really contained on-switches. To do this, we extracted them from both chimp and mouse DNA and attached them to a gene that turns cells blue—but only when that gene is activated. We injected this stitched-together piece of DNA into fertilized mouse eggs to see if any parts of the embryos were blue as they developed—indicating a functional switch in the piece of DNA—and, if so, where.
Male Turnoffs
My results were exciting: they really seemed to show that I was working with a true on-switch for the androgen receptor, one that human beings had shed. In mouse embryos, the genital tubercle (which develops into either a clitoris or penis) stained blue, as did the developing mammary glands and spots on the mouse face where sensory whiskers called vibrissae form. All these tissues are known to make the androgen receptor respond to testosterone. Looking more closely, I saw that the staining on the developing genitals was situated in places where small, tough protein spikes later form on the mouse penis.
Neither sensory whiskers or spiny penises are human features, of course. But they occur in many mammals, including mice, monkeys and chimps. It is also known that a loss of testosterone results in shorter whiskers in male rodents and a lack of penile spines in rodents and primates. Penile spines and whiskers might similarly disappear if a crucial DNA switch were lost and the androgen receptor were no longer made in these tissues.
As I pursued my experiments, others were busy with their own deletions of choice, with intriguing results as well. Then graduate student Alex Pollen found that his chunk of DNA activated the neural gene it was near, at precise spots in the developing brain. The gene is involved in a key process: it helps to kill off surplus neurons, which are overproduced during embryonic development. That offers a tantalizing thought: because the human brain is far larger than the chimp’s (1,400 versus 400 cubic centimeters), might loss of this switch have contributed to that evolutionary ramp-up, by releasing brakes on brain growth?
Vahan B. Indjeian, then a postdoc in the lab, similarly found that his switch turned on the gene involved in skeletal growth—in developing hind limbs, specifically the toes of the foot. Toes two through five in humans are shorter than in apes and mice, alterations that improve the foot for upright walking.
It is easy to see how brain and bone switches fit into the pattern of human evolution. Loss of both appears linked to hallmarks of humanity: a big brain and walking on two legs. Loss of sensory whiskers is fairly easy to rationalize because we no longer root around in the dark with snouts to grub out food or capture prey but use hands, in daylight, to find nourishment. Despite their reduced importance, though, it is unclear how we would be better off without these whiskers.
Sensitive Relationships
The penile spine story is less intuitive, but it is potentially more powerful and also fits neatly into the adaptive history of our species. Loss of spines, we believe, is one of a suite of changes that together had far-reaching effects on our evolutionary path. Together these changes altered the ways we mate, the physical appearance of males and females, our relationships with one another and the ways we care for offspring.
Made of keratin, the same stuff as our fingernails, these spines occur in many mammals, including primates, rodents, cats, bats and opossums, and range from simple microscopic cones to large barbs and multipronged spikes. They may serve varied functions depending on the species: heightening stimulation, inducing ovulation, removing sperm deposited by other males, or irritating the vaginal lining to limit female interest in mating with others.
The copulation time of spine-sporting primates is remarkably brief: in the chimp, typically less than 10 seconds. And historical experiments in primates show that removal of penile spines can extend copulation by two thirds. From such observations we can surmise that loss of penile spines was one of the changes in humans that have made the sex act last longer, and thus be more intimate, compared with that of our spine-bearing forebears. That sounds pleasant, but it could also serve our species from an evolutionary perspective.
Our own reproductive strategy is unlike that of any apes, which all have intense male-male competition at their core. In chimps and bonobos, males compete to mate with as many fertile females as possible. They produce copious quantities of sperm (chimp testicles are three times larger than human ones), have penile spines and, like all male great apes and monkeys, have deadly, fanglike canines to discourage rivals. They leave rearing of offspring entirely to the female. Thus, for her, successful mating results in considerable commitment—gestating, nursing and rearing each infant to independence—and the female does not reproduce again until the weaning is completed.
Humans are different. They form fairly faithful pair bonds. Men often help to rear offspring, enabling earlier weaning and increasing reproductive rate. Male-male competition is not as intense. We believe that loss of penile spines went along with loss of other traits associated with fierce competition (such as dangerous canines) and gain of others that promote bonding and cooperation.
Bipedalism, as Lovejoy proposed, could be one of these features. Early male help probably initially took the form of procuring foods rich in fat and proteins, such as grubs, insects and small vertebrates, that required extensive search and transport. Males would need to travel far with hands free for carrying, which likely provided the initial selective advantage for walking on two legs.
Gene Loss and Feature Gains
And there is more. Cooperation and provisioning would also allow parents to rear dependent offspring for longer and thus lengthen the juvenile period after weaning. This would offer a longer time for learning and therefore enhance the usefulness of a large, agile brain—indeed, perhaps set the stage for its evolution.
In that sense, the individual stories of all three of our deletions are deeply intertwined.
When I came to Kingsley’s lab, I did not anticipate the turn my work would take—that I would find myself poring over fusty 1940s texts on mammalian genital structure. My lab is continuing research into this and other genetic and developmental changes with big consequences: the evolutionary shaping of the delicate bones in the human wrist to perfect them for toolmaking.
There is much we may never know about all this distant history, no matter how keen we may be to find out. But even if we cannot be sure about the why of an evolutionary change, with the tools of modern molecular biology we can now tackle the how—a critical and fascinating question in its own right.
