Human Errors

by Nathan H. Lents

  The entire genome—functional or not—gets copied every time a cell divides. This consumes cellular energy and requires time, energy, and chemical resources. The best estimates are that a human body experiences at least 1x1011 cell divisions each day. That’s over a million cell divisions per second. Each one of those divisions involves copying the entire genome, junk and all. You expend a few dietary calories every single day just to copy your largely useless DNA.

  Oddly, cells meticulously spot-check this junk DNA for errors. Each time a cell copies this irrelevant DNA, it engages the same proofreading and repair mechanisms that it does when it copies the most important genes in the genome. No region is ignored; none gets special attention. This is perplexing because a replication error in a stretch of gibberish DNA is inconsequential, while the same mutation in a gene can be lethal, as we’ll see shortly. The machinery for copying and editing DNA can’t seem to differentiate between the two types—genes and gibberish—any more than a chimpanzee can differentiate between a poem by a preschooler and one by Maya Angelou.

  We are in an exciting new era of biomedical research. Scientists can now read the entire sequence of someone’s genome, all 4.6 billion letters spread across the forty-six chromosomes, in a process that takes just a couple of weeks and costs about a thousand dollars. (The first complete sequencing of a human genome took over a decade to finish and cost nearly three hundred million dollars.) But although we are finding many surprising new functions for some regions of DNA formerly referred to as junk, these are still overshadowed by the nonfunctional junk. What’s more, even the apparently functional junk probably began as pure junk.* Given all of the nonsense encoded in the human genome, it’s amazing we’ve turned out as well as we have.

  Although the enormous tangled masses of useless DNA in the genome may be the biggest flaw of all, even the functional parts of it—the genes—are rife with errors. These errors generally come from mutations, which are what we call the changes that get made to a DNA sequence. There are two common ways that the genome can experience sudden changes. (Three if you count retroviral infections, but we’ll talk about those soon enough.) One is through damage to the DNA molecule itself. This can occur due to radiation, UV light, or harmful chemicals called mutagens, such as those abundant in cigarette smoke. (Mutagens are often called carcinogens due to the mutations’ tendency to cause cancer.)

  A second way that the genome can experience changes is through copying errors made when DNA is duplicated in preparation for cell division. Each cell has 4.6 billion letters of DNA code, and each day the average person experiences somewhere around 1x1011 cell divisions, so that is literally 1020 (100,000,000,000,000,000,000 or 100 quintillion) chances a day for a cell to make a mistake while copying DNA. Cells are terrific copyeditors, making fewer than one mistake in a million letters and immediately catching and correcting 99.9 percent of those rare errors. But even with that incredibly low error rate, with so many chances to make a mistake, sometimes mistakes do happen and don’t get corrected. Those become mutations. In fact, every day, you experience millions of mutations throughout your body.

  Fortunately, most of those mutations occur in unimportant regions of the DNA so they don’t really matter. Further, mutations in cells that are not sperm or egg cells have no real consequences for evolution because they don’t get passed on. Only the DNA in the so-called germ cells contributes to the next generation.

  However, both copying errors and DNA damage can and do occur in important regions of the genome of sperm or egg cells. When this happens, these mutations will probably not affect that individual so much as his or her children. These are thus called heritable mutations and are the basis of all the evolutionary changes and adaptations in living things. But it’s not all happy accidents when it comes to heritable mutations. While most are inconsequential (considering how much of the genome doesn’t do anything anyway), many mutations are harmful because they disrupt the function of a gene.

  The poor offspring that inherit a gene mutation from their father or mother are almost always worse off for it. That’s what natural selection is all about—keeping the gene pool clean—but sometimes the harm that a mutation brings is not immediate. If a mutation causes a human or animal no short-term loss in health or fertility, it won’t necessarily be eliminated. It could even spread throughout a population. If this mutation causes harm only far down the road, natural selection is powerless to immediately stop it.

  This is evolution’s blind spot, and its ramifications can be seen throughout our species—and deep inside each of us. For the human genome contains thousands of scars from harmful mutations that natural selection failed to notice until it was too late.

  Broken Genes

  Of the useless DNA in the human genome, one particular kind stands out: pseudogenes. These stretches of genetic code look like genes but do not function as such. They are the evolutionary remnants of once-functional genes that became mutated beyond repair at some point in our species’ deep past.

  We saw one such pseudogene in the previous chapter, the GULO pseudogene, which in its functional form allows nearly all nonprimate animals to synthesize their own vitamin C. In some common ancestor of all living primates, the GULO gene was damaged by random mutation. Because this ancestor happened to have a diet rich in vitamin C, the mutation didn’t cause any harm to that individual. However, since it was passed on to all primates, they—we—are subject to the horrors of scurvy.

  You might wonder why nature wasn’t able to fix this problem the same way it created it: through mutation. That’d be nice, but it’s nearly impossible. A mutation is like a lightning strike, a random error in the process of copying 4.6 billion letters of DNA. The odds of lightning striking the same place twice are so infinitesimally tiny as to be nonexistent. What’s more, it’s exceedingly unlikely that a mutation will fix a broken gene because, following the initial damaging mutation, the gene will soon rack up additional mutations. If the first mutation doesn’t kill or harm the bearer, future mutations won’t either. These mutations will not be eliminated by natural selection.

  This is why, over the scale of evolutionary time, the mutation rate of pseudogenes is dramatically higher than that of functional genes. Mutations in functional genes don’t usually persist across generations; typically, the lightning strike will cause such harm to the cell or organism in which it occurs that the individual will be less likely to reproduce successfully, thereby limiting the spread of the faulty genetic material. Pseudogenes, however, are free to accumulate mutations without harming the entity that carries them—and that’s exactly what happens. The pseudogenes get passed on and on, continuing their deterioration down through the generations, and it doesn’t take long before the gene is mutated beyond all hope of repair.

  That’s what happened to the human GULO gene. Compared to the functional version that most other animals have, our GULO gene is littered with hundreds of mutations. Yet it is still easily recognizable. Our GULO gene is more than 85 percent identical to the DNA sequence of the functional GULO gene found in carnivores such as dogs and cats. It’s mostly all there, just sitting useless, like a car rusting in a junkyard—except that humans have continually refashioned this rusty old gene, billions of times every day, ever since the gene was initially broken tens of millions of years ago.

  Thanks to scurvy, GULO may be humans’ most famous pseudogene—but it is hardly the only one. We humans have quite a few broken genes lurking in our genomes—actually, well over a few; more than a hundred or even a thousand. Scientists estimate that the human genome contains the intact remnants of nearly twenty thousand pseudogenes. That’s almost as many broken genes as functional ones.

  To be fair, the majority of these pseudogenes are the result of accidental gene duplications. This explains why the disrupting mutations and subsequent death of the gene didn’t have any deleterious effects on the individual—these were extra copies of genes anyway. Their function was redundant with that of other genes, so
losing these genes didn’t put anyone at a disadvantage. Of course, it’s still pointless to keep them around and duplicate them constantly. Pointless, and a waste of energy—but not directly harmful.

  But when the only copy of this or that functional gene gets broken by a mutation and becomes a pseudogene, it can really hurt. Besides GULO (and its gift of scurvy), another pseudogene whose breakage had a deleterious impact on our species’ health is a gene that once helped our ancestors fight infections. This gene created theta defensin, a protein still found in most Old World monkeys, New World monkeys, and even one of our fellow apes, the orangutan. However, in an ancestor common to humans and our African ape relatives, gorillas and chimpanzees, it was deactivated, then mutated beyond repair. Without a working version of it, humans are more susceptible to infections than our more distant primate cousins.

  Admittedly, we have probably evolved other defenses to take the place of these theta defensin proteins—but not enough, it seems. For instance, cells that lack theta proteins appear more susceptible to HIV infection. We really could have used these proteins in the late 1970s and 1980s as HIV ravaged human populations around the globe. Were it not for this broken gene, the AIDS crisis might never have happened, or at least it might not have been so widespread and so deadly.

  Pseudogenes are a lesson in the cruel habits of nature, which takes no thought for the morrow. Mutations are random, and natural selection operates merely from one generation to the next. Evolution, however, operates on very long timescales. We are the long-term products of short-term actions. Evolution is not—indeed, cannot be—goal-oriented. Natural selection is affected only by immediate or very short-term consequences. It is blind to long-term ones. When GULO or the gene that produces theta defensin die by mutation, natural selection can protect the species only if the lethal effects are felt immediately. If the bearers of the mutation continue to flourish and pass it down to their progeny, evolution is powerless to stop it. The death of the GULO gene probably had no consequence at all for the first primate that endured it, yet his or her distant progeny suffer still, several tens of millions of years later.

  The GULO gene and the theta defensin gene are not unique in having suffered such debilitating mutations. Every single one of our other twenty-three thousand genes has been, and still is being, struck and killed by the lightning bolts of mutation. The only reason humans haven’t lost more genes to mutation is that the first unlucky mutant usually dies or is made sterile and is thus not able to pass on the pseudogene. A tragic fate for her or him, but lucky for the rest of us.

  Some scientists refer to pseudogenes as dead rather than broken, because nature has managed to “resurrect” some of them to serve new functions. This reminds me of something a friend of mine once did when his refrigerator broke. Instead of undertaking the hassle of hauling it to a junkyard, he turned it into an armoire for his bedroom. He didn’t buy a refrigerator intending to turn it into an armoire; he just repurposed a broken fridge as one because it was much easier than getting rid of it. He resurrected his dead refrigerator for a brand-new purpose. It was a nifty trick—but as far as we know, resurrected genes are as rare as armoire-refrigerators.

  Alligators in the Gene Pool

  As we’ve just seen, the process of copying DNA is not perfect. The machinery that our bodies have developed for this purpose occasionally makes mistakes, and these mistakes can cause problems. But those kinds of mutations are sporadic—freak accidents, like the sudden death of the GULO gene in one primate’s genome, which just so happened to spread to an entire population of organisms. Like the mutations themselves, the diseases that sometimes result from these errors, such as scurvy, are sporadic. But there is an entire class of genetic diseases that are more insidious than these, precisely because the mutations that cause them weren’t fixed by the accident of genetic drift. They were actually favored by natural selection.

  Humans have a whole host of persistent genetic diseases that have been with our species for generations, millennia, even millions of years. Each of them has an interesting story, and collectively they can teach us some valuable lessons about the sloppy and sometimes cruel process of evolution.

  Probably the most well-known and widespread example of a genetic disease that has frustrated humans for ages is sickle cell disease, or SCD. Three hundred thousand babies are born with this condition every year. In 2013 alone, at least 176,000 people died from it. The disease is caused by a mutation in one of the genes for hemoglobin, the protein that carries oxygen around in the bloodstream and delivers it to all cells.

  Normally, red blood cells are packed with hemoglobin and take on a shape that facilitates both maximum oxygen delivery and optimal folding so that the cells can squeeze through tiny blood vessels called capillaries. The mutant versions of hemoglobin found in SCD patients, however, do not pack together as tightly, resulting in poorly shaped red blood cells. These malformed cells do not deliver oxygen as efficiently and, worse, cannot fold and squeeze through small vessels. They tend to get stuck in tight spaces, creating a sort of sanguinary traffic jam that leads to an intensely painful and sometimes life-threatening sickle cell crisis as the tissue downstream of the traffic jam becomes starved for oxygen. In the developed world, the danger of sickle cell crises can usually be managed with close monitoring and modern medicine. In underdeveloped regions of Africa, Latin America, India, Arabia, Southeast Asia, and Oceania, it is often fatal.

  The strangest aspect of SCD is that it is caused by a single-point mutation, a simple switch of one DNA letter for another (though there are many possible point mutations that can cause the disease, with different mutations common to different geographical ethnic groups). This is truly strange because a point mutation with such a drastic negative effect on survival is usually eliminated from the population rather quickly. Research in population genetics has shown convincingly that a mutation that causes even a slight disadvantage will be eliminated from a population in a matter of a few generations, not thousands of years. To be sure, genetic diseases caused by the interaction of multiple genes or those that give only a slight predisposition toward illness are sometimes difficult for natural selection to sort out. But SCD should be easy. It’s a single mutated gene with a disastrous effect. There is simply no way that it should have persisted for very long.

  The shape of normal red blood cells (left) and those showing signs of sickle cell disease (right). Whereas normal red blood cells easily fold in half in order to squeeze through tight capillary vessels, sickle-shaped cells are much less flexible and often get stuck in tight spots.

  Yet the mutant coding that causes SCD is hundreds of thousands of years old, and it has appeared and spread—spread!—in many different ethnic groups. How could a mutation that causes a horrific, debilitating disease, one that can easily be fatal without modern medical intervention, not only pop up multiple times and in multiple places throughout human history but also appear to often have been favored by natural selection? Further, how could it have spread so relatively far in the populations it affects?

  The answer is surprisingly simple. Like many genetic diseases, sickle cell disease is recessive. This means that you need to inherit two copies of the mutant allele, one from each parent, in order to develop the disease. If you inherit only one copy, you will not be affected in any noticeable way—although you will be a carrier, able to pass the gene to your children, and they may develop the disease if the other parent also gives them a bad copy. If two carriers of SCD procreate, approximately one-fourth of their children will develop the disease, even though both parents appear to be healthy. For this reason, recessive traits sometimes give the appearance of skipping generations. Nonetheless, with SCD being so deadly, the early deaths of the sufferers of the disease should have eventually eliminated it from the population.

  The reason the SCD mutations were not eliminated is that the carriers of SCD—the individuals who have only one copy of the code and are thus not afflicted with the symptoms—are more re
sistant to malaria than noncarriers. Malaria, like SCD, is a disease affecting red blood cells. However, malaria is caused by a parasite that is passed to humans through mosquito bites. It turns out that people with just one copy of the mutant SCD allele do have a slight difference in the shape of their red blood cells. It’s not enough to cause the sickle cell disease, but it is enough to make the cells inhospitable to the parasite that causes malaria.

  Sickle cell disease is often discussed in introductory biology courses as an example of something called heterozygote advantage—heterozygote being the term for someone who has two different copies of a certain type of allele. A carrier of SCD is a heterozygote for the affected gene because she has one copy of the mutant allele and one normal copy. To see why being a carrier could be an advantage, first consider the fact that, if you get two copies of the mutant SCD allele, you are in serious trouble. However, if you get just one copy, you are better off than those who have no copies, because you’re both SCD-symptom-free and you have a lower chance of contracting malaria. In areas of the world where malaria is and has been a real problem, the mutant SCD gene is being pulled in two directions by natural selection. On the one hand, sickle cell disease can be deadly; on the other hand, malaria can also be deadly. Evolution had to weigh one threat against the other, and the result is a compromise in which the mutant alleles that cause SCD but protect against malaria are present in up to 20 percent of the population in the most malaria-infested regions of Central Africa.

  A map comparing global distribution of the genes for sickle cell disease with the range of the Plasmodium parasite that causes malaria. Because sickle cell disease confers resistance to malaria, their geographical reaches are markedly overlapping.