The Edge of Evolution
Page 6
For chloroquine to kill malaria, it has to get inside the parasite’s “stomach” and stay there for a while. In fact, the parasite itself grabs the drug and concentrates it ten-thousand-fold in its digestive vacuole. The process is complex, and there are many ways it could short-circuit, yielding resistance to the drug. Let’s speculate about the possibilities, without worrying yet about the mutational complexity.
Perhaps the digestive vacuole’s environment could be changed a bit to make it less congenial to the drug. Or possibly a protein pump that ordinarily removes other things from the cell could be altered a bit, to toss out the poison. Or maybe some of the ordinary repair machinery of the cell could be tweaked to chemically damage the chloroquine. Or possibly P. falciparum could change some of the components of its membrane to stop the entry of the chemical. Or it might develop an alternative way to deal with waste. The large number of potential ways for the parasite to counter chloroquine makes it difficult for scientists to track down what really is going on in drug resistance. It has only been in the past few years that the mutation that makes P. falciparum resistant to chloroquine has been unmasked. Even now researchers are unsure if it’s the whole story, but they are confident that at least it is a large chunk of the story.
A group led by Thomas Wellems of the National Institutes of Health made the discovery through a series of genetic studies of the parasite. First they narrowed it down to just one of the parasite’s fourteen chromosomes. More sophisticated (and laborious) studies further narrowed the region containing the resistance gene to a four-hundred-thousand-nucleotide region of that chromosome,3 and then eventually to a thirty-six-thousand-nucleotide region.4 Painstaking analysis of the details of this area of the parasite’s genetic map uncovered a previously unnoticed gene.
When that gene was sequenced, workers were able to determine the amino acids in the protein they were seeking. It was a needle in a haystack: one of the approximately fifty-three hundred proteins that the parasite’s DNA encodes.5 With the progress that biology has made over the past few decades, scientists are able to tell a great deal about what role a protein is likely to play in a cell just by looking at its amino acid sequence, even before conducting any experiments. They have learned to recognize patterns in amino acid sequences that are reliably found in proteins that do particular kinds of jobs. As an analogy, if an engineer saw an unfamiliar machine that had wheels, he would guess that it was probably used for transportation of some sort; if the machine had a sharp blade, that part probably was used for cutting something; and so on.
The sequence of the protein—dubbed PfCRT, for P. falciparum chloroquine resistance trait—revealed that the protein contained ten separate stretches of amino acids that were all hydrophobic (water-hating), or oily. This suggested that in the cell, those regions would be stuck in a membrane, which is itself made of oily molecules that prefer contact with other oily molecules. The other regions of PfCRT were not so hydrophobic and probably stuck out into the water on one side of the membrane or the other. Other proteins known to have such features help form pumps and portals. The membranes of all living cells contain many different kinds of protein machines that act as gateways, allowing such molecules as foodstuffs or nutrients to pass in and waste products to pass out. Sometimes the gateways are passive, simply allowing the right-shaped molecules to float through on their own. Often, however, the portals are active, grabbing the right molecules and pushing them through the membrane. Because the cell has to deal with many different kinds of molecules that pass in both directions, it has many separate pumps and portals.
CHANGING THE PUMP
The amino acid sequence of PfCRT suggested that it was a protein pump, but that suspicion needed to be confirmed. Using clever laboratory techniques, Wellems and his coworkers demonstrated that the protein was located in the membrane of the parasite’s digestive vacuole—its stomach. Thanks to this protein, the stomachs of mutant parasites accumulate a lot less chloroquine, and the bug survives to reproduce. Exactly why the mutant stomachs collect less drug is currently unclear, but it may well be that the mutation allows the chloroquine to leak out through the PfCRT pump.6
The staggering complexity of modern biology is a challenge for anyone to understand, but in order to find the edge of evolution, we need to get to the bottom of it—and we aren’t quite there yet. The PfCRT protein has 424 amino acids. Just as sickle hemoglobin exhibits a change in its amino acid sequence from normal hemoglobin, the mutant PfCRT also has changes in its sequence.7 And just as different hemoglobin mutants can be found in different areas of the world (such as HbC in Africa, HbE in Asia, and thalassemias around the Mediterranean Sea), different mutations have been found in PfCRT from different regions of the globe. Scientists have analyzed the protein from P. falciparum from patients in South America, Asia, and Africa. The mutant PfCRTs exhibit a range of changes, affecting as few as four amino acids to as many as eight. However, the same two amino acid changes are almost always present—one switch at position number 76 and another at position 220. The other mutations in the protein differ from each other, with one group of mutations common to chloroquine-resistant parasites from South America, and a second clustering of mutations appearing in malaria from Asia and Africa. This suggests that chloroquine resistance in malaria probably arose at least twice, separately in South America and Asia, and that the Asian resistance was transmitted to Africa. Later work suggested that there had actually been four separate origins.8
Since two particular amino acid changes occur in almost all of these cases, they both seem to be required for the primary activity by which the protein confers resistance. The other mutations apparently “compensate” for side effects caused by these two primary mutations.
FIGURE 3.1
Schematic drawing of the PfCRT protein. Each circle represents an amino acid position. Arrows point to black circles that represent two positions (76 and 220) where mutations are almost always found in resistant proteins. (Reproduced from Bray, P. G., Martin, R. E., Tilley, L., Ward, S. A., Kirk, K., and Fidock, D. A. 2005. Defining the role of PfCRT in Plasmodium falciparum chloroquine resistance. Mol. Microbiol. 56:323–33. Courtesy of Blackwell Publishing.)
In the last chapter we saw that changes in human genes in the wake of malarial attacks were diminishments—beneficial only in dire circumstances, but detrimental in normal times. P. falciparum, however, greatly outnumbers humans, and reproduces much more rapidly, and therefore has many more opportunities for lucky genetic accidents. By standard Darwinian theory, it ought to make the next step in the arms race very early. Standard Darwinian logic predicts that malaria will mutate more, and sift its mutations more effectively, than humans. So are the changes in the mutated PfCRT an improvement? Is the parasite strengthening in an absolute sense, and evolving new “advanced and complex machinery,” as Richard Dawkins might expect? It appears not. When chloroquine is no longer used to treat malaria patients in a region, the mutant strain of P. falciparum declines and the original strain makes a comeback, indicating that the mutant is weaker than the original strain in the absence of the toxic chloroquine.9 Apparently, much like human thalassemia or sickle hemoglobin or G6PD deficiency, the mutant malarial protein is a net plus only in desperate circumstances—in trench warfare.
TIGER BY THE TAIL
As a teenager I was a big fan of science fiction, and I remember reading a short story entitled “Tiger by the Tail.”10 In the story an opening to another dimension somehow popped up in…a pocketbook! Someone had the bright idea to toss a grappling hook through the opening. If the grappling hook pulled stuff from the other dimension into ours, or vice versa, then the dimension that lost the tug of war would be destroyed. (There was a scientist on hand to explain it all.) This, the humans thought, was a great way to blackmail the aliens on the other side. As the story ended, however, the chain holding the grappling hook, which had been slowly emerging from the pocketbook, reversed direction as the aliens pulled harder on their end. The tables had turned, a
nd now all humanity was threatened.
There is an analogy to the human-malaria struggle. The malarial parasite is turning, too, pulling harder on the grappling hook of synthetic drugs. In many ways chloroquine was a dream drug—not only effective but cheap, and with few side effects. It lasted for decades before resistance to it became widespread. Newer drugs and methods to combat malaria fall short on one or more of these features. Not only is malaria almost wholly resistant to chloroquine, but it is becoming increasingly adept at shrugging off the newer drugs that have followed. Malaria that has developed resistance to one drug seems to develop resistance to new drugs at an accelerated rate compared to “initial” malaria.11 Appendix B details the specifics of several rounds of this trench war.
The development of drug resistance in malaria, like the development of the sickle cell gene and thalassemia in humans, is a crystal clear example of Darwinian evolution in action. We see it all right there—the selective pressure exerted on malaria by toxic drugs, the occasional mutations that make one bug more fit than its kin, the spreading of the mutation through the population. Yet malaria beautifully illustrates both the strengths and the shortcomings of the sort of blind search that Darwinian evolution demands. And with the help of mathematics, we can finally begin to achieve some precision about the limits of random mutation.
Answer: The obstacle that malaria hasn’t been able to mutate around.
Question: What is sickle hemoglobin?
In our grudging admiration of P. falciparum’s wizardry at quickly mutating past our wonder drugs (even if the changes are ultimately diminishments), it’s easy to lose sight of its failure to deal with sickle hemoglobin. Sickle has been around for thousands of years, not for mere decades like antimalarial drugs. Resistance to one recent drug, atovaquone, arose in the lab scant weeks after a small culture of malaria was exposed to it. Almost a hundred thousand times as many ticks of the clock have passed since Sickle Eve was born. About that much time has passed since C-Eve lived, too, and since thalassemia first appeared. Yet they are all still effective against malaria.
How can that be? Why should a single amino acid change in sickle hemoglobin checkmate malaria’s million-murdering death when the best rational efforts of chemists are brushed aside in short order? Is the answer the complexity of the chemical? No. Chloroquine is no less complex than the new amino acid in sickle hemoglobin. Is it the specificity of the target? No. The problems chloroquine causes are no less specifically targeted to the stomach of the parasite than the problems sickle creates in the red blood cell; arguably chloroquine is more specific. Is it the inexperience of the parasite in dealing with the arena of attack? Hardly. The parasite routinely eats hemoglobin; it is made to deal with the stuff. Chloroquine is an artificial chemical. Malaria never saw the drug before the 1930s. Yet the parasite conquered chloroquine but is stumped by sickle.
I will get to the reason why sickle is such a challenge. At this point, let’s simply take note of an exceptionally important implication of the widely varying success of malaria at dealing with the challenges that come its way. One simple yet crucial conclusion that we can already draw is this:
Darwinian evolution can deal quickly and easily with some problems, but slowly if at all with others.
Are there problems that are even harder for evolution than dealing with sickle hemoglobin? Problems that are for all intents and purposes beyond the reach of random mutation?
POWERBALL
Almost every Monday night my wife and I go to a sports bar a few miles from home to have a couple of hours alone together, away from the kids. (Buffalo wings are half price on Monday night.) On our way there I usually stop at a convenience store to buy a Powerball lottery ticket. It surely won’t be long until I hit the jackpot, but so far I haven’t managed to match more than a single number on any ticket. Luck of the Irish.
In the multistate Powerball lottery five white balls are drawn from a drum that contains fifty-three consecutively numbered balls, and then one (the fabled Powerball itself) from a separate container with forty-two red balls. The lottery ticket has five numbers listed together and then a separate number. There are various ways to win prizes. If the separate number on your ticket matches the Powerball, you get three dollars. If both the Powerball number and one other match, you’re up to four dollars. To win the grand prize (at least ten million dollars) all the numbers have to match.
The odds of winning the grand prize are officially listed as one in 120,526,770. The odds of winning lesser prizes are better: one in 260 for matching three white balls, one in 12,248 for matching four white balls, one in 2,939,677 for matching five white balls. Not surprisingly, the odds of winning get worse the more balls that have to match, and they get worse rapidly. Matching four balls isn’t just, say, 33 percent harder than matching three—it’s about fifty times harder. Matching five balls is hundreds of times less likely than matching four, and ten thousand times less likely than matching three.
FIGURE 3.2
The odds against winning a lottery such as Powerball increase rapidly the more numbers that have to match. The same principle holds true for mutations to DNA.
Often in a Powerball drawing no one wins the jackpot, because no one bought a ticket that matched the winning numbers. When that happens the accumulated money is rolled over and the jackpot for the next drawing is that much bigger. How long will it likely take before someone (not a particular person—anyone at all) wins the grand prize? Besides the odds, the length of time depends on two other things—the number of players and the frequency of drawings. If the odds of winning are one in a hundred million, and if a million people play every time, then it will take on average about a hundred drawings for someone to win. So if there are two drawings per week (about a hundred per year), then it would take about a year before someone won. But if there were only one drawing per year, on average it would take a century to hit the jackpot. If the number of players were different, the average time to produce a winner would also change. If ten million people played each time instead of one million, then on average only ten drawings would be necessary to get a winner. If a billion people played each time, then it would be very likely that each drawing would have multiple winners.
The very same three simple considerations that regulate how often the Powerball lottery is won—the odds of winning, the number of players, and how often the lottery is held—also govern how fast malaria develops resistance to an antibiotic. Just as the odds of winning at Powerball depend on the range of possible numbers (1 to 53) and how many balls you have to match, the odds of developing antibiotic resistance depend on the number of nucleotides in the parasite’s genome (millions) and how many mutations have to accumulate before there’s a beneficial effect. Just as the time to get a winner depends on how frequently Powerball lotteries are held, the time to produce resistance in P. falciparum depends on the organism’s mutation rate and generation span. And just as the time to a jackpot is shorter the more people play Powerball, the time to antibiotic resistance is shorter the more malarial parasites are exposed to a drug.
BACK STOP
Malaria scientists are acutely aware of these factors and take them into account when planning the battle against the parasite. One team of researchers notes that “the ease with which a population of parasites survives exposure to a drug depends on…the frequency at which the parasites develop resistance…and the size of the parasite population at the onset of treatment.”12 For example, if the chances that a cell will be resistant to a given drug are one in a million, and if there are a hundred million parasitic cells in a patient, then almost certainly there will be some resistant cells in the patient. Administering the drug will kill off the 99.9999 percent of cells that are sensitive, but the approximately one hundred resistant cells will survive and multiply. Soon the patient will be filled with resistant cells.
If the chances for resistance to a different drug are one in a billion, then odds are good that a person suffering from malaria who carrie
s just a hundred million parasitic cells will be cured by the drug, which would likely kill all the cells. However, if ten people each have a hundred million cells, then altogether there are a billion P. falciparum cells, and there’s an even chance that one of those cells in one of those patients will be resistant. If a clinic treats a hundred such patients a day, then on average one in every ten of those patients will harbor a resistant cell. When those patients are treated, the resistant cells will survive and replicate, the patients will be bitten by mosquitoes, and the mosquitoes will spread resistant cells to other people. In short order the drug will be useless.
To greatly increase the chances of successful treatment, one strategy is to use a cocktail of drugs, each component of which is able to kill a sizeable chunk of cells. For example, in urging that several drugs should be used simultaneously against malaria, one researcher explained:
Resistance to antimalarial drugs arises when spontaneously occurring mutants…which confer reduced drug susceptibility are selected, and are then transmitted. Simultaneous use of two or more antimalarials…will reduce the chance of selection, because the chance of a resistant mutant surviving is the product of the parasite mutation rates for the individual drugs, multiplied by the number of parasites in an infection that are exposed to the drugs.13
Suppose a cocktail contains two drugs, A and B, and that one in a million parasite cells are resistant to drug A, and one in a million to drug B. Assuming resistance to A is due to a different mutation than resistance to B, then the odds that a single individual cell is resistant to both drugs at the same time are multiplied, a million times a million, which is one in a trillion.