4

Beautiful Monsters

MONSTERS LOOM LARGE in speculations about the workings of nature. In the centuries before Darwin, the word monster had an almost technical meaning. Natural philosophers and anatomists crafted taxonomies to describe two-headed goats, multilegged frogs, and conjoined twins. In the sixteenth century, many thought these deformities came about as the result of too much seed during conception or from a pregnant woman’s wandering thoughts.

A new science was heralded in the 1700s when the German anatomist Samuel Thomas von Sömmerring (1755–1830) surmised that monsters reflect alterations in normal development rather than mystical causes. They were, in his words, “disruptions of the generative force.” On the title page of his monograph on the subject in 1791, he depicted duplicated human heads: stillborn infants with two complete heads sprouting from the neck, and others with duplications of only the face. In his view, each case represented an alteration of normal development at different stages. Complete duplicate heads came about from disruptions of early stages of development, while incompletely fused faces arose from later ones.

A few decades later, Geoffroy Saint-Hilaire proposed that monstres, a term he used frequently, reflected the hidden potential for creatures to transform into one another. After his expedition with Napoleon in Egypt and his encounter with lunged fish (see Chapter 1), Saint-Hilaire spent his days trying to mutate chicken eggs, adding various chemicals to perturb their development. He believed that if he added just the right concoction of chemicals to developing embryos, then he could change one creature into another. Following an early notion that chickens went through a fish stage in their normal development, Saint-Hilaire worked for decades trying to make chicken eggs produce fish hatchlings. That attempt failed, but his son Isidore picked up the mantle and produced a three-volume treatise on congenital anomalies that is still in use today. Isidore developed a taxonomy of birth anomalies, categorizing them by type, organ affected, and degree of anatomical effects. For example, he studied conjoined twins, classifying them according to how many organs were involved and the extent to which their anatomical systems were intermingled. This work formed the basis for later researchers to assess the biological mechanisms, as opposed to supernatural causes, involved in producing anomalies.

With the publication of On the Origin of Species, Darwin transformed the study of developmental anomalies. To Darwin, if the motor for evolution is natural selection, then variation among individuals is its fuel. If individuals in a species vary in having traits that look and function differently, and some of those traits enhance the success of those individuals in a particular environment, then over time those creatures and traits should increase. If a trait is harmful, then it will diminish over time. The essence of evolution is variation among individuals. If all individuals in a population are exactly alike, evolution by natural selection could never happen. The differences among individuals are evolution’s raw material for natural selection; the more variation, the faster evolution could work. Only with a rich supply of variation, including the type revealed by monsters, could natural selection lead to major changes over time.

One of the champions of the study of variation after Darwin was William Bateson (1861–1926). Like Darwin, Bateson grew up with a passion for natural history. When asked as a youth what he wanted to be, he famously replied that he wanted to be a naturalist, but if he wasn’t good enough, he would have to be a doctor. Bateson entered Cambridge University in 1878 as a lackluster student. But Darwin’s On the Origin of Species had a profound effect on young Bateson. He became energized to understand how natural selection works. For him, answers lay in understanding how species vary: What were the mechanisms that made organisms look different from one another? Reading the work of Gregor Mendel, who discovered the principles of heredity in pea plants, Bateson had an epiphany: variation that was transmitted from one generation to the next was the essence of evolution. He translated Mendel’s work into English and invented a new term to describe it: genetics, derived from the Greek work genesis, meaning “origin.”

Bateson, like Geoffroy Saint-Hilaire before him, wanted to classify the ways species and individuals differ. But Bateson had one advantage. Armed with new ideas from the growing field of genetics, he looked for the ways that variation among individuals could influence how evolution worked.

Bateson devoted almost a decade to this study, producing the monumental Materials for the Study of Variation in 1894. The book contains a road map of the ways creatures differ from one another and a search for general rules that underlie the production of variation and, ultimately, the path of evolution. In assessing as many species as he could, he described two different modes of variation. One type is a difference in the size or degree of organs, which form a continuous series from smaller to larger. Populations of mice, for example, have differences in the lengths of their appendages, tails, or other organs. This kind of variation can be easily quantified by measurements of length, width, or volume. The other kind of variation is more dramatic, involving the presence or absence of structures. The polydactyly of Hemingway’s cats is one example. Normal individuals have five toes, while polydactylous ones have six or more. These cats differ from normal ones in the number of toes they have, not in, say, the length of their bones. This type of variation is of kind, not of degree or size.

The search for creatures with extra organs became a passion for Bateson. He was struck by oddities in nature—extra organs or organs in the wrong place, such as bees with legs where antennae should be, or humans with extra ribs, or males with extra nipples. In these cases, it was as if organs were being cut and pasted throughout the body. A well-formed organ could be duplicated in toto or moved about to different places in the body. These monsters had a mystery to them, and understanding them might reveal general rules about how bodies are built and evolve.

Natural philosophers from the sixteenth century onward had been correct in their view that monsters reflect something essential about the living world. What was needed was the right kind of monster and the scientific tools to understand it.

The Fly

One of the greatest decisions in the history of biology came about when Thomas Hunt Morgan (1866–1945) decided to work on flies. Morgan began his career by researching sea acorns, worms, and frogs, convinced that inside their cells and embryos lay clues to our own biology. Nor did he choose them esoterically or haphazardly; he focused on small aquatic creatures that could rebuild complete body parts after losing them. Planarian worms, for example, are champions of regeneration: cut them in half, let them regrow, and the end result would be two complete individuals. Many creatures—worms, fish, and amphibians—can rebuild after trauma. We can only be jealous of our animal cousins; somewhere along our evolutionary line, mammals lost this ability.

Morgan entered science at a time when much of what we take for granted today was completely unknown. The Czech monk Gregor Mendel discovered that traits can be passed on from generation to generation, but the source of that heredity was a mystery. People had observed cells, but the notion that chromosomes play a role in that process was not known, let alone the existence of DNA.

Implicit in Morgan’s science was a fundamental shift in thinking about life, something that undergirds virtually all biomedical research today: diverse creatures, from worms to sea stars, can offer insights into general mechanisms of human biology. His work was governed by the tacit recognition that all creatures on the planet share deep connections.

After a few years of performing experiments on regeneration and describing them in his influential book Regeneration, published in 1901, Morgan realized that the tools simply didn’t exist for him to make significant progress. He began a hunt for a new research program. At the heart of it all, from regeneration to anatomy, lies heredity—the passing of information from one generation to the next. Learning what drives heredity would be a key to unlocking many of biology’s mysteries. Morgan was convinced that insights into genetics would come from finding a creature that bred and grew quickly, was small, and could be maintained in huge numbers in a lab. He ideally wanted a species whose chromosomes, by then proposed but not proven to contain genetic material, could be seen microscopically. This was a pretty long checklist, one that excluded the creature he most wanted to understand—humans.

Unknown to Morgan at the time, an insect taxonomist was on a similar mission, albeit from the opposite side of the problem. Charles W. Woodworth (1865–1940), at the University of California at Berkeley, made it his lifework to uncover the arcane details of insect anatomy, with an eye to classifying flies and other insects. This quest made him an expert in fly biology, so much so that he saw one species, the fruit fly, Drosophila melanogaster, as a potential experimental model. Sometime in the early 1900s (the exact year is not known), he reached out to William E. Castle (1867–1962), a biologist at Harvard, and suggested he try some experiments on fruit flies.

Like Bateson, Castle was interested in uncovering the mechanisms of heredity and variation. At the time, Castle was working on guinea pigs to understand how their fur color and body patterns were passed from generation to generation. But guinea pigs were a source of frustration because the females give birth to eight offspring at most and take almost two months to gestate. To study heredity, Castle had to wait months for them to breed enough to make multiple generations. Woodworth’s suggestion to work on flies had an obvious attraction; the average fruit fly lives for forty to fifty days, during which time a female can produce thousands of embryos. Castle realized that he could do more experiments on heredity in one month with flies than he could in years with guinea pigs.

Castle switched to working on flies and established methods to breed and rear them. He published a paper on fly experiments in 1903 that is less memorable for its scientific results than for its impact on the community. Other scientists, including Morgan, saw the beauty and power of studying flies.

Drosophila seems like an unlikely candidate for groundbreaking discoveries. About three millimeters in length, it lives on rotting fruit. Most of us encounter them around garbage as tiny nonbiting flies that annoy by hovering about. But what makes them a pest makes them promising for science.

Morgan’s work followed the tradition of monsters, which meant finding and analyzing mutants. Mutants are keys to the functioning of normal genes. A mutant with no eyes reflects a defect in one or more genes that control eye formation. In this way, mutants are lodestars that can be used to identify the genes involved in the development of different organs. Since mutants are rare, Morgan needed to breed thousands of flies to pick up a single mutant. He and his team kept hundreds of breeding colonies of flies and put each individual under the microscope to look for any anomalies.

Unknown to most of us, the fly body that emerges under a microscope is beautifully complex. Seen at medium power, an entire world of bristles, spines, and appendages emerges from their body segments. Morgan’s team became familiar with this complexity so that any change, no matter how small, served as fodder for their analysis of new mutants. They spent long hours bent over microscopes looking for flies with any odd trait, perhaps differently shaped wings, novel stripe patterns, or an altered appendage.

Genes, as we now know, are sequences of DNA that are bundled tightly to form chromosomes. Chromosomes sit within the nucleus of a cell, and under the right conditions, they are visible under a microscope. Morgan knew nothing about DNA, but he could see chromosomes. They became his window into genes.

Genes are segments of DNA that are wound and packed tightly into chromosomes that lie within the nucleus of a cell. Notice the banding of the chromosomes.

Morgan devised ingenious ways to try to link the anatomy of mutants to their genetic material. His team found that flies have enormous chromosomes inside their salivary glands. Removing them, and treating them with a red dye obtained from a wild lichen, revealed a series of white and black stripes on the chromosome, some thick and others thin. Morgan then mapped the patterns of white and black bands in both normal flies and those with mutations. By comparing the differences in stripes, he could see the location on the chromosome where the two differed, in essence revealing where the genetic change that made the mutation resided.

The flies fed on rotten bananas, so the Morgan lab was permeated by the smell of garbage. Working there meant spending hours peering into a microscope. Because of these conditions, success in Morgan’s group called for a special sort of person—one who could, despite all else, stay focused on fly bodies, chromosome bands, and mutants. At stake was one of the biggest questions of life: How is information passed on from one generation to the next?

Chromosomes of the midge Chieronomus prope pulcher, with black and white stripes

Morgan’s lab was initially in a cramped space at Columbia University, where stocks of flies were stored, bred, and analyzed under the microscope. Known as the Fly Room, the lab would host a who’s who of early twentieth-century biologists, as Morgan attracted some of the best and brightest to his lab. After spending fourteen years at Columbia, he moved the entire operation to Caltech in 1928, winning the Nobel Prize in 1933.

One of Morgan’s early students possessed a legendary ability to work with flies. Calvin Bridges (1889–1938) had not only the best eyes to discern mutant flies but the patience to sit for hours to find them. Bridges discerned tiny differences among flies that were invisible to others. He brought technical advances too: switching to a binocular microscope expanded the range of his vision and led to the discovery that flies feed well on agar. The latter was an important change for the lab—no longer would the Fly Room smell like rotten bananas.

With a shock of erect hair that seemingly defied the laws of physics, Bridges was a restless soul. When he wasn’t working long hours in the lab, he would often disappear for extended stretches of time. He once emerged with pictures of a new automobile he had designed. Rumors of his amorous trysts abounded, and Morgan disapproved of his private life. The buzz about his affairs meant that Bridges never got promoted to a faculty position at Caltech. When he died in his forties, the word in the lab was that he was killed by the spouse of a jealous lover. Sadly, the truth was just as tragic. Recently a genetics colleague of mine asked his brother, a Los Angeles DA, to dig out Bridges’s death certificate. Bridges died from complications of syphilis.

Calvin Bridges and his hair

To the external world, the lab maintained a complete silence about Bridges’s personal behavior. But he had had such an impact on Morgan’s work that Morgan shared his Nobel Prize winnings with Bridges’s family after his untimely death.

While Bridges was known for spotting mutant flies that had subtle differences in coloration, wing shape, or bristle pattern, one of his most famous discoveries was relatively easy to spot. Its difference would have been hard to miss by even a rank amateur. The name, Bithorax, says it all—instead of two thoracic segments and wings, it had four. A whole region of its body was duplicated, wings and all.

Bridges drew the fly’s body and described its anatomy. Then he did what geneticists do when they find a mutant: he raised the stock and kept it going in the Caltech fly lab. He made a colony of these mutants that could be maintained indefinitely.

Bridges wanted to find the location in the chromosome where the change might have occurred. Using Morgan’s technique of staining the salivary chromosomes, he was able to locate a region in the double-winged mutant where the banding was different from that of normal flies. The Bithorax mutant had happened because of a change in a broad region of the fly’s chromosome.

Normal fruit fly on the left, the Bithorax mutant on the right

Morgan and Bridges’s quest to understand a single trait in flies opened up a new world of challenges and opportunities. They and others showed that various traits in flies are heritable. Some kind of biological material is passed from generation to generation that tells the developing embryo of a fly to place wings in the correct part of the body. Bridges’s mutant revealed that this material resided along a stretch of the fly’s chromosomes. But what was this material that builds organs and bodies, and how does it do its magic? Could it tell us how bodies are built and how they evolved over millions of years?

Beads on a String

Edward Lewis’s (1918–2004) passion for flies was kindled when he saw an advertisement in a magazine. Born in Wilkes-Barre, Pennsylvania, he had an intense curiosity that led him to spend long hours in the local library. Seeing an advertisement for fruit flies, he brought it to the attention of his high school’s biology club. The club set up a fly colony, and Lewis began tinkering with flies.

Lewis entered Caltech in 1939, a year after Bridges’s death, to learn the tools of genetics that had been pioneered in the Fly Room. He was a quiet man with a very rigid diurnal rhythm: he spent early mornings in the lab, exercised at eight a.m., did more solitary work, had an afternoon lunch at Caltech’s famed faculty club, the Athenaeum, then returned to work and played his beloved flute until dinner. He had, like Bridges, a prodigious capacity for sitting for long hours over a microscope working on flies. His favorite time, by all accounts, was the quiet of the lab after dinner. Lewis’s work finding and breeding mutant flies was a form of meditation.

Ed Lewis with his flute in the living room of a friend

The stockroom where Bridges made his great technical advances was still functional and housed the famed Bithorax mutants. By the time Lewis started his studies, he knew of the Bithorax mutant and also had a hunch about its structure. Since Bridges’s map showed that the Bithorax mutant spanned several bands on the chromosome, Lewis thought it might be part of a region containing not one but many genes involved in development.

Seeking to isolate the genetic material that made the extra wing, Lewis devised a novel, but time-consuming method to probe Bithorax. He spent decades on this work, not publishing a single scientific paper for over ten years as he dedicated himself to Bithorax. The six-page article that appeared in 1978 was as revolutionary as it was impenetrable. To understand it all, the paper must be read multiple times, because it is crammed with the insights derived from years of a quiet life with flies.

Lewis had developed a powerful new technique: he would remove a large area of the fly’s chromosome and let the fly develop to see the effect on the body in flies lacking this large region. Then he would add small fragments back sequentially, to see those effects on the body. This approach enabled him to determine what individual pieces of a chromosome can do in isolation.

This approach reminds me of a diet that comes in and out of popularity, called a cleanse. People would fast for several days, then add different food groups to their diet sequentially and in combination. By refraining from eating entirely, then adding only dairy products for a few days, they could see how eggs, milk, and cheese affected their energy levels and mood, for example. Then, by fasting and adding foods in different combinations, they could see the interactions, say, between dark leafy vegetables and dairy. Lewis was doing the same with the large region of the chromosome that held the Bithorax mutant—he took it out completely, let the animals develop to record the effect, and then added bits back in isolation and in different combinations in other embryos, noting their impact on the flies’ bodies as they developed into adults.

Lewis’s genetic cut-and-paste revealed that Bithorax was not caused by a single gene but a group of many of them. The genes lay in a row on the chromosome, like pearls on a necklace. These genes, he surmised, worked together to build the embryo, and each gene had its own function. But that wasn’t the most remarkable thing.

A fly’s body is composed of segments from front to back—head, thorax, and abdomen. Each segment carries an appendage: antennae and mouthparts on the head, wings on the thorax, and legs and spines on the abdomen. Lewis found that each gene in the Bithorax region controlled a different segment of the fly’s body. One gene placed the antennae on the head, another the wings on the thorax, and another the legs on the abdomen. These genes played a role in building basic body architecture. The front-to-back organization of the body was encoded genetically. And, to everyone’s great surprise, the structure of the body was mirrored by the position of the genes on the chromosome: the genes that were active on the head lay on one end, those for the abdomen on the other, and the ones for the thorax in the middle. The organization of the body was reflected in the activity and structure of the genes.

While Lewis’s finding was exhilarating, a lot of biology suggested it might pertain only to flies. For one thing, fly segments are different from parts of other animals, such as fish, mice, and humans. Flies lack a backbone, a spinal cord, and other structures seen in bodies like ours. Fish, mice, and people lack antennae, wings, and bristles.

An even greater difference lies in how the fly develops. During development, most animals have millions of different cells, each with its own nucleus. A fly embryo looks like a single cell with many nuclei, like a giant bag of genetic material. You could not imagine a stranger animal than a fly to try to use to say anything about how animals in general develop and evolve.

The Monster Mash

In 1978, when Lewis’s paper on Bithorax was published, the field of biology was undergoing a technological revolution. In Morgan’s day, genes had been a kind of black box—he and his team were able to piece together their effect on the body and their place on the chromosome, but virtually nothing was known about how they worked, let alone that they were regions of DNA.

By the 1980s, a few years after Lewis published his paper, biologists were able to sequence genes as well as see where they were actively making proteins in the body. Mike Levine and Bill McGinnis, working in the lab of the late Walter Gehring (1939–2014) in Switzerland, had access to a mutant fly in which a leg sprouted from the head, where an antenna normally would be. The head developed normally except for the presence of the leg. Much like Bridges’s mutant fly with the extra wings, or Bateson’s cut-and-paste variations, this mutant shuffled body parts, and it had a defect specific to the head segment.

Using DNA technology that Bridges could not have imagined, Levine and McGinnis were able to isolate the gene responsible for the mutation. Then they made a special piece of DNA to test where the gene was active in development. Recall that when genes are active, they make proteins. To manufacture proteins, they use another molecule, RNA, as an intermediary. To test where genes are turned on, you need to see where RNA is being made. So the two attached a dye to a molecule that would find the RNA wherever it was in the fly body. When this concoction was injected into a developing fly embryo, the dye would be brought to the places where the gene was turned on, and the stain would be visible in the embryo under the microscope.

Normal fly on the left, mutant on the right. It was named Antennapedia because it sprouted a leg where an antenna should be.

The gene of the mutant Antennapedia, with the leg growing out of its head, was normally active in a very specific place: the head. Moreover, the gene controlled the kind of organ that formed in the head, whether an antenna or, as in the mutant, a leg. If this situation sounds familiar, it is because it is what Ed Lewis saw in his chromosomal work on Bithorax years before. Recall that he saw a series of genes, one after another on the chromosome, each specific to one body segment, each controlling which organ developed there. Maybe this head gene was a harbinger of discoveries to come, one of a group of genes that controlled what was happening in each of the fly’s body segments.

The result sent Levine to Lewis’s 1978 paper. He began a long interaction with it, reading and rereading it over fifty times, but still, as he said, he “did not completely understand it.”

Lewis’s paper led Levine and McGinnis to chase one of his major predictions: that there should be a string of similar genes lying next to one another on the chromosome. With the gene isolated, they began a hunt to see if there were any others like it nearby. The technique was crude: they mushed fly bodies into a paste, isolated their DNA, put the mixture in a gel, and added their gene with a dye. The idea was that the gene would act like molecular flypaper and attach to every gene with a similar sequence. The dye would allow them to find and isolate these genes.

The result was unmistakable—there were many other genes like it in the genome. Sequencing each one, Levine and McGinnis found that the dyed genes all had a small stretch of DNA inside that was virtually identical. In a stunning coincidence, Matt Scott, at the University of Indiana, made the same discovery independently.

Now, knowing the sequence of the genes, scientists could apply the same techniques on a larger scale to see where they were active in the fly body during development and where they resided on the chromosome. Using the tricks they had deployed on the mutant that started it all, researchers from around the world found something unexpectedly beautiful: these genes lie next to one another on the chromosome, and each one is active in a different body segment of the fly.

In the midst of this frenzy of experiments, Levine was chatting with a scientist in another lab who pointed out that flies aren’t the only animals that have body segments. Earthworms are basically tubes with blocklike segments that run the length of the body. Why not look at them too? Perhaps their genes marked their segments as well.

This casual comment sent Levine and McGinnis running to the garden behind their building to collect every creepy crawly creature they could find: worms, insects, and flies. After extracting each creature’s DNA, they probed whether they, too, had genes in a similar sequence. They did. And they didn’t stop there. Subsequent research would reveal that the DNA of frogs, mice, and even people had this sequence too.

Subsequent work on worms, flies, fish, and mice revealed universal truths about animal bodies. Versions of the body-building genes of flies turned up virtually everywhere, from worms to people. All these genes were set like beads on a string next to one another on the chromosome. And each gene seemed to be active in a specific segment of the body—head, thorax, and abdomen. In addition, as Lewis first saw, the position of each gene on the chromosome matched the order of the segments from front to back.

Hox genes, set like beads on a string, are active in the body segments of flies and mice.

The papers that described these genes were in the stack that kindled my own work in genetics and molecular biology almost four decades ago.

In 1995 the Nobel Prize committee recognized Edward Lewis for opening up a new world of biology. As he accepted the prize, he was classically circumspect. In his acceptance speech, he said prizes were nothing compared to his first loves, “flies and doing science.”

The world of bugs, flies, and worms is a mash of creatures with different numbers of segments and different types of appendages emerging from them. Think of a lobster with antennae in front, followed by big claws, small claws, and legs. Each of these appendages emerges from a single segment of its body. In centipedes, each body segment has an identical leg emerging from it. Flying insects have wings instead of legs in certain segments. People have vertebrae, ribs, and limbs that lie along the body. With these genes, scientists could now ask how the basic body architecture of animals developed and evolved.

Calvin Bridges identified the general chromosomal region that made an extra set of wings; Ed Lewis revealed that the region contained many genes, each active in a specific part of the body; and Levine, McGinnis, and Scott showed that those genes are deeply ancient among all animals. A new generation was now inspired and poised to understand how these genes worked.

Cut and Paste

When my children were toddlers at the beach on Cape Cod, they used to find little shrimp-shaped animals in the sand. Poking them, and watching their response, led to their nickname, “jumpies.” These creatures, more commonly known as scuds or sand fleas, are about half an inch long, have clear bodies, and usually burrow in beach sand. When provoked, they can contract their bodies and jump a foot or so into the air. The familiar beach variety is only one of the eight thousand known species. All of these species have a remarkable ability to move about by using a diverse array of swimming, digging, and hopping behaviors. They accomplish this with a virtual Swiss army knife of legs: some are large, others small, some face forward, still others face backward. Their name, amphipod, is a Greek reference to having backward- and forward-facing legs: amphi means “dual” and pod, “leg.”

Starting his own independent lab in 1995 in Chicago, biologist Nipam Patel wanted to find a perfect animal to explore how genes work to build bodies. Since amphipods sport so many different kinds of legs, he had a hunch that they could make an excellent creature to study Lewis’s genes. He spent years scouring nineteenth-century German monographs to identify the perfect amphipod to bring into the lab. The 1800s were the apex of anatomical illustration and description, and entire rooms in library stacks are dedicated to different groups. Armed with insights from the descriptions and lithographic plates, Patel developed a plan that also fit nicely into his long-standing hobby.

A visit to Patel’s house in Chicago meant navigating a giant saltwater aquarium in the center of his living room. Since he was a dedicated amateur aquarist, his experience with the filtration system in his home tank gave him an idea. Keeping the system clean was a regular problem, especially keeping the filter clear of the small invertebrates that collected and grew on it. He couldn’t help but notice that amid the grime were small invertebrates burrowing in the muck. Apparently they loved the nutritious particles that flowed by and made it a happy home.

That gave Patel an idea. If tiny creatures liked his small filtration system, imagine the diversity of creatures he might find in the filtered mud of the massive saltwater tanks at Chicago’s Shedd Aquarium. These tanks housed sharks, skates, over fifty species of large fish, and even a human docent in scuba gear from time to time. Patel dispatched a graduate student with a bucket to see what he could find in the filtration system. He had a hunch that the muck would harbor robust little animals that he could use in the lab.

The filters at the Shedd proved to be an Eden for small invertebrates. Patel’s student spent his days scraping the filters, looking at the creatures that lived there under the microscope. One of them—an amphipod known as Parhyale—was extremely promising for research. It was small, bred fast, and grew to adulthood quickly. It also had appendages, lots of different kinds of them. It looked like a perfect experimental animal. Patel worked to breed them in the lab and get the experiments going. Morgan had used flies to understand the mechanisms of heredity; Patel was determined to use amphipods to explore how genes build bodies.

Soon after obtaining Parhyale from Chicago’s Shedd, Patel moved to the University of California at Berkeley to establish a research program centered on the creatures. Berkeley, Patel, and Parhyale proved to be an auspicious fit, because at Berkeley was Jennifer Doudna, one of the scientists who discovered a new way to edit the genome, CRISPR-Cas. With this technique, scientists can target regions of the genome with two kinds of tools: a molecular scalpel to cut DNA and a guide to bring the scalpel to the right place. In 2013 Doudna and her colleagues from around the world had shown that the DNA of different species could be cut and edited with great precision. Their CRISPR scalpel could be used to cut genes out of the genome. Rearing the embryos would let scientists see the effects of removing one of their genes. Other more complicated experiments involved substituting or editing the sequence of genes.

The power of this technology spawned an idea for Patel: What if you could edit Parhyale’s genes to make the genetic activity of one body segment look like that of another? Could you move limbs and body parts around?

Parhyale has limbs along its body length, and each segment of the body contains a different appendage. The front segments of the head have antennae and are followed by segments that contain pieces of the jaw. (We call the jaws and mandibles of invertebrates limbs because, like appendages, they extend from a body segment.) The thorax holds larger limbs, some facing forward with others facing backward. Tiny limbs extend from the abdomen, too, with bushy ones in the front abdominal segments and short stubby ones in the rear.

Six of Lewis’s genes are active during the development of the body axis of Parhyale. As in flies, different body segments can be identified by the kinds of limbs that develop in them and by determining which of the genes are active in the segment during development. What if you could change the pattern of gene activity in the segments—say, make the thorax segment have the abdomen’s genes active inside it? Would that change the kinds of limbs that emerged from the segment? Patel turned genes off one by one, using the gene editing technique developed by his Berkeley colleague.

The elegance of Patel’s experiments emerges in the details. Three of Lewis’s genes, called Ubx, abd-A, and Abd-B, are active in the rear end of Parhyale during development. Their activity in the body marks four regions: one toward the head, where only Ubx is active, followed by another where both Ubx and abd-A are active, one with abd-A and Abd-B active, and one where only Abd-B is active. You could think of each of these four regions as having a genetic address defined by which genes are active inside them. It turns out that the pattern of gene activity corresponds to the kind of appendage that forms. Where only the Ubx gene is active, a backward-facing limb forms, the combined Ubx/abd-A segment produces a forward-facing limb, the abd-A/Abd-B one a bushy limb, and the Abd-B segments, a stubby one.

Patel’s plan was to delete genes to change the addresses of different body segments. What happens when you change the pattern of activity in each body segment?

When Patel deleted the abd-A gene, the parts of the body that formerly had a Ubx/abd-A address now had only Ubx. The part that had had an abd-A/Abd-B now had only an Abd-B address. With the change in addresses came a beautiful experimental monster: a creature with backward-facing limbs where forward-facing ones should have been, and stubby ones where bushy ones would normally be. Shifting the patterns of gene activity in the body segments changed which appendage formed in each segment.

Regular pattern of gene activity (top, shaded areas). Deleting genes to change the patterns of activity in the segments (bottom) changes the kinds of limbs that develop inside.

Patel found he could change genetic addresses and move appendages around the body at will. In doing so, he wasn’t just creating monsters; he was mimicking the diversity of life in nature.

Compare amphipods to their cousins the isopods. Most of us know isopods from one of their most common species: pill bugs. As the name isopod (Greek: “same legs”) implies, they have only forward-facing legs, unlike amphipods, which have both forward- and rear-facing legs. When Patel deleted the abd-A gene in the amphipod, he made creatures that looked like isopods: they had only forward-facing limbs. He copied nature as well: isopods lack abd-A in their normal development.

Changes to these genes explain the differences between creatures as distinct as lobsters and centipedes. The combination of genes active where a lobster’s big claw is made is different from those that make a leg. And in creatures like centipedes, where each segment has the same kind of leg, similar genes are active in each body segment. In insects, worms, and flies, these genes form a road map to the body.

The Monster Within

Parhyale, lobsters, and flies are only the start of the story. Frogs, mice, and people have versions of these genes too. They have different names in people and other mammals. Instead of names like abd-A, Abd-B, and others, they are called Hox genes, followed by a number, such as Hox1, Hox2, and so on. Also, where flies, worms, and insects have only a single string of these genes on one chromosome, we have four sets of these strings on four different chromosomes.

These genes are active along the axis of the bodies of mice and people, and much like flies and Parhyale, they are active in different body segments. Our body segments don’t sprout wings, or legs that face in every direction. Ours hold vertebrae and ribs. Despite these differences, the question becomes: Does our development work the way it does in Parhyale and flies? If you changed the activity of the genes in development, could you make mutants with different numbers of ribs and vertebrae?

Mammalian backbones follow a formula that rarely changes: seven neck vertebrae, followed by twelve thoracic vertebrae, each with a rib, then five lumbar vertebrae. This set is followed by the sacrum and the tail, which in humans is retained as a set of small fused vertebrae called the coccyx.

Just as in flies and Parhyale, our different body segments have different addresses of gene activity. For example, one combination of Bithorax-like genes marks our cervical region, another the thoracic. Likewise, the boundaries between the thoracic and lumbar regions and between the lumbar and sacral vertebrae both have different genes active inside.

What happens when one genetic address is changed into another? Making mutants is far more difficult in mice than in flies or Parhyale. It can take years, largely because the generation time is longer and there are more genes to mutate. But the results are worth the wait.

Take the situation for the lumbar and sacral vertebrae. The region that becomes the lumbar vertebrae has activity of a gene known as Hox10. It is followed by the sacral region, which has a genetic address of two genes, Hox10 and Hox11. In a mutant in which the Hox11 genes are deleted, the segments that would normally form the sacrum have the lumbar genetic address. What happens to the body segments? The end result is a mouse in which the entire sacrum has been transformed into lumbar vertebrae.

Further experiments show this pattern can be repeated with different genes and body parts. Thoracic vertebrae carry ribs. By deleting genes, the entire rear end of the vertebral column can be given the genetic address of thoracic vertebrae. The result: mice with ribs that extend all the way to the tail. As Patel did with Parhyale, modifying the genes changes body segments and the organs that develop inside them.

One could call the products of these experiments monsters, but that would hide how beautifully they reveal the mechanisms behind life’s diversity. A nineteenth-century observation of life, a discovery in the Fly Room, and modern-day genomic biology combine to reveal beauty inside animal bodies. The genetic architecture that builds the bodies of flies, mice, and people reveals that we are all variations on a theme. From a common toolkit come the many branches of the tree of life.

Changes to the activity of Hox genes can predictably change sacral vertebrae into lumbar vertebrae.

Reuse, Recycle, Repurpose

As the ubiquity of Lewis’s genes in different species was revealed, long-forgotten arcane monographs from the nineteenth century came under renewed scrutiny. In the early 1990s, the observations and ideas of classical natural philosophers such as William Bateson were fodder for cutting-edge experiments. Bateson had observed that some of the most common kinds of variation entailed changing the number of body parts or having body parts sprout in odd places. Calvin Bridges, Edward Lewis, and the molecular biologists who came later were following a path that had been set nearly a century before. And just as in the nineteenth century, monsters and mutants, whether they were made in the lab or found in the wild, were at the center of it all.

My training was in a world of fossils, museum collections, and expeditions. But one result sent me scurrying to learn molecular biology as quickly as I could.

As teams of researchers around the world explored the activity of Hox genes in mice, they found something completely unexpected. Mouse Hox genes do not merely control the formation of the vertebrae and ribs along the body axis; they are active in different organs of the embryo, from the head and limbs to the guts and genitalia. It is almost as if these genes are redeployed across the body to build any organ that has its own segmented structure. The patterns of gene activity were pointing to a kind of biological cut-and-paste: a genetic process used to form the main body axis was redeployed to make other bodily structures.

A number of experiments in the early 1990s revealed that the activity of these genes in the limbs is much like that in the body axis; they are active at different times in development and appear to provide a genetic address to different parts of the limb. All limbs, from frogs’ legs to whales’ flippers, have a similar skeletal pattern. Each has a single bone at the base, the humerus. Then two bones, the radius and the ulna, extend from the elbow. At the end are the bones of the wrist and digits. While the sizes, shapes, and numbers of bones may differ in creatures that use wings to fly, flippers to swim, or hands to play piano, this one bone–two bones–little bones–digits pattern is always there. It is a grand anatomical theme, an ancient pattern that underlies the diversity of every creature with a limb skeleton.

What’s more, these three anatomical regions—upper arm, forearm, and hand—correspond to three zones where different Hox genes are active. Each region corresponds to a different address of gene activity, much as in the body of a fly, Parhyale, or a mouse.

Now researchers could ask, What happens when you change the pattern of genetic activity in the different segments of limbs? We saw in Parhyale, and in the body axis of mice, that changing the pattern of gene activity of different body segments could have predictable effects on the organs that develop from them.

In the 1990s a French team of scientists made mutants by deleting Hox genes in mice, much as Patel had done with Parhyale. When they deleted the Hox genes active in the tail, they made a mutant mouse lacking a tail. But now they did the same experiment in the limb. The same Hox genes that make the tail are also active in the limb. They define the most terminal segment of the limb—the hand or the foot. When the French team deleted those same genes active in limbs, they made a population of mice with only the one bone–two bones skeleton in their limbs. The mice that developed with the missing genes lacked hands.

I’ve spent most of my career looking at how hands and feet came about from the fins of fish. My colleagues and I spent six years studying the fossil record to find a fish with arm bones and wrists. Here, suddenly, we had evidence showing the genes that were necessary to make hands.

This result led me to pursue a new path in my own research. In addition to collecting fossils, I realized that I needed to be able to do experiments on genes. Having that toolkit would give me the ability to ask new kinds of questions. Did fish have these genes? If so, what were they doing in fish fins? Could these hand genes help explain how fins were transformed into limbs?

Fish you see at the market, on a dive, or in an aquarium do not have fingers and toes; the fin is made up mostly of a large set of rays with webbing between. The bone in the fin rays is different from the bone of digits. Digits initially form from cartilage precursors, while fin rays develop directly underneath the skin. As we know from the fossil record, the transition from fins to limbs involved two big changes: a gain of digits and a loss of fin rays.

Because the French team revealed the genes that were necessary to make the hands and feet of mice, you might think that those genes are unique to creatures with limbs. But that would be wrong. Fish have these genes too. What are the genes that make hands and feet doing in the fins of fish?

Two young biologists spent four years exploring this question in my Chicago laboratory. First Tetsuya Nakamura worked to duplicate mammalian gene experiments with fish fins. He diligently removed the genes, but the animals lacking these genes did not easily thrive. Remember, these genes are also active in making vertebrae, so the mutant animals could not easily swim. After three years of making mutants, and helping them thrive, Nakamura found something remarkable: when these genes were deleted from the genome, the mutant fish were missing the fin rays.

I first met the second young scientist in 1983, when my anatomy professor, Lee Gehrke, brought his brand-new infant son to a lecture. Little did I know that two decades later the baby, Andrew Gehrke, was going to end up doing a Ph.D. in my laboratory. Gehrke, like Nakamura, would be in the lab until three a.m. most nights devising experiments. A lab in Canada showed that when you marked the hand genes in mice and traced their development, almost all the cells ended up in the wrist and fingers. No big surprise there. The surprise was in fish fins. One late night Gehrke traced the activity of these genes in fish fins and snapped a picture. The resulting figure made the front page of The New York Times for the simple reason that it told a big story. The genes that are necessary to build the hands of mice and people are not only present in fish, but they make the bones that sit at the end of the fin skeleton, the fin rays.

The transformation of fins to limbs is a world of repurposing at every level: genes that make hands and feet are present in fish, making the terminal end of their fins, and versions of these same genes help build the terminal end of the bodies of flies and other animals. Great revolutions in life do not necessarily involve the wholesale invention of new genes, organs, or ways of life. Using ancient features in new ways opens up a world of possibility for descendants.

The pattern of gene activity that is needed to make hands (left) is present in fish making the terminal end of their fins. The light area shows where similar Hox genes were active during development.

Modifying, redeploying, or co-opting ancient genes provides fuel for evolutionary change. Genetic recipes do not need to arise from scratch to make new organs in bodies. Existing genes and networks of them can be pulled off the shelf and modified to make remarkably new things. Using the old to make the new extends to every level of the history of life—even to the invention of new genes themselves.