Influenza One Flu Two Flu Your Flu Bird

nfluenza. Before the mid-20th century, people gave this disease the respect it deserved. Now we call it the flu; act fairly cavalier about the symptoms unless we need a break from work; and poo-poo the vaccine unless we're in a high-risk group populated by babies, octogenarians, and hypochondriacs. Periodically, though, we get reminded that the flu can be serious business — a fact that epidemiologists and virologists have been trying to beat into our thick skulls forever — and then we fly into a panic.

Well, here's news for you, some good, some bad:

^ Many strains of influenza exist, and courtesy of evolution, even more are on the way. Their rapid rate of evolution makes them a constant problem, and every so often, an especially nasty strain catches scientists by surprise.

^ Although some of us may succumb, we aren't completely helpless. Scientists have been fighting back, even using the process of evolution to turn the tables on these pesky viruses. Thanks to techniques for vaccine design, scientists are getting closer to being prepared for these new and improved strains as well.

ln This Chapter

^ Discovering where new flu strains come from

Predicting next year's flu today ^ Understanding vaccines

This chapter has the details on all the bad things the flu can do and all the reasons by you don't need to panic — at least not yet.

The Flu and Your Immune System

Most likely, your immune system has already been introduced to the flu. You got sick, felt lousy for a while, and eventually got better. Why? Because your body has an immune system. To understand the rather eventful history of influenza in the human population, you need to know how the flu virus infects humans and how the human immune system fights back. Consider it a play in three acts.

Act 1: The virus attacks and spreads

The influenza virus, which causes the illness commonly referred to as the flu, is most often transmitted from one person to the next in small droplets drifting through the air — possibly the result of somebody's sneeze or cough.

One day, you're unlucky enough to inhale one of these droplets, and a flu-virus particle attaches to one of your cells. Once inside the cell, the particle takes over the machinery necessary to make more copies of itself. Then your cell bursts and releases all these copies, which go looking for more cells to infect. At this point, you're sick, and you'll probably help the virus get to new cells in other people when you start coughing. End Act 1.

Act 2: The body fights back

You're sick, but you don't stay sick. Instead, your body responds to the invading flu, fights back, and overcomes the infection. How? Keep reading.

Your body has numerous, quite complicated systems for fighting off microscopic foreign invaders such as the flu virus. Combined, these various mechanisms make up your immune system. I won't get into the details here except to note that your body has two general classes of response:

^ Nonspecific responses: Nonspecific responses are reactions, such as tissue inflammation and fever, that are generally bad for all invading microbes. When your body senses the presence of invading microbes, it sometimes turns up the heat in its fight against them. Indeed, having a fever may make you feel bad, but it makes the invading microbes feel even worse.

^ Specific responses: Specific responses include such things as antibodies, which are special proteins that seek out and destroy invaders. When your body becomes aware of the flu, the first thing it notices is that the flu virus is foreign. It's not part of you and, as far as your body is concerned, doesn't belong in you. That's why as soon as the flu has been identified as "not self," your body begins to produce antibodies that specifically target this strain of flu.

As your body responds to the invader in both nonspecific and specific ways, it starts winning the battle against the flu. After a week of two, all the flu particles in your body have been killed, and you're on the road to recovery.

Act 3: Building up the guard jjjjWHB? Your immune system has a very special property called memory. Although it takes a little time for your body to produce enough antibodies to beat back this strain of flu, memory means that after your body has responded once, it's immediately ready to respond again.

That's great news for you, because the flu strain you just recovered from can no longer invade your body; now you are immune to it. And since your body's antibodies are able to attach to different but similar flu strains, you also have resistance to those as well.

Which leads to a sequel: The return of the flu

The evolution of influenza is strongly affected by the immune system's response. The strain of flu from which you recovered was clearly quite good at being the flu. It managed to get not only from some other person into you, but quite possibly into a third person as well. In fact, it's possible that the strain of the flu you got also spread through your workplace, your kid's school, your town, the rest of the country, and even the world. Indeed, flu strains do this all the time. But after everyone who had a particular strain of flu gets better, all of them are immune to that strain. Thus, the next time you get the flu, it will be a genetically different flu strain.

The Three Types of Influenza: A, B, C

Three different strains of flu infect humans. These strains are conveniently named influenza A, influenza B, and influenza C. They all have genomes that use RNA instead of DNA as the genetic material:

1 Influenza A: This flu sweeps through the human population each flu season. It evolves so fast that strains are sufficiently different to avoid existing antibodies every year. Influenza A occurs in humans and a variety of other animals, which makes the evolution of influenza A somewhat more complicated, as well as more interesting for this discussion. The remainder of this chapter concentrates mostly on influenza A.

1 Influenza B: Influenza B is just like influenza A, except that influenza B exists only in humans. Beyond this single difference, you just need to remember that influenza B evolves similarly to influenza A.

1 Influenza C: Because most people are immune to influenza C, this strain of flu is not characterized by epidemics that periodically sweep through the human population. Instead, it is primarily a disease of the young, because it has a chance to reproduce only in children who have not yet developed an immunity to this generally mild flu. I mention it here in case you're curious, but it really isn't important to the discussion in this chapter.

Unlike influenza A and B, influenza C evolves very slowly. We don't understand exactly why it evolves so slowly, but as result of this slower evolution, all influenza C strains are basically the same. About 97 percent of Americans have antibodies to influenza C, which means it's likely that at some point in your life, you were infected with it and developed a resistance to it. Now you are immune to the strain that infected you, and as a result of the high level of similarity among strains, you are also immune to all other influenza C viruses.

The Evolution of Influenza A

Influenza A is a very small virus with a genome made of RNA instead of DNA. Its genome is comprised of eight different RNA segments, and it has only ten genes. (For comparison, consider that people have over 25,000 genes.) Influenza A is a remarkably compact organism, and these ten genes enable it to do everything it needs to do to infect you and reproduce.

Mechanisms of evolution: Mutation, recombination, or reassortment

The influenza A virus (and the influenza B and C viruses too, for that matter) can evolve in three ways:

1 Mutations: Influenza is a virus with an RNA genome, and RNA replication has a high error rate. For this reason, the virus mutates rapidly. That's why this year's flu is different from last year's flu. The strains may be similar enough for the immunity you developed to last year's flu to protect you (or at least partially protect you) but don't count on it.

1 Reassortment: Because influenza has a genome divided into segments, new variants can also arise via genetic reassortment, the process whereby two different viruses infect the same cell, and the progeny have segments that are a combination of these two strains. (For an explanation of how viruses can reproduce this way, refer to Chapter 12.)

When reassortment involves the segments that code for the H or N proteins that cover the outside of the flu virus, the new strains (or the strains produced) can be so different from the parent strains that no host has even partial immunity. As a result, these strains can sweep rapidly through entire populations.

1 Recombination: In recombination, one of the existing influenza genes is incorrectly replicated and ends up with a new section of RNA spliced into it. This sequence could come from another influenza strain or even from a host cell's RNA. Although such events are almost always likely to be harmful to the virus, every once in a while a new sequence with foreign RNA spliced into just the right part of one of the surface proteins may make the new flu strain resistant to existing host antibodies.

Genes to know

For the purposes of this discussion, you need to pay attention to only four genes: three on the outside and one that determines which species a particular strain can infect.

The three genes on the outside are

1 Hemagglutinin (abbreviated H) 1 Neuraminidase (abbreviated N) 1 Matrix

The fourth is the Nucleoprotein gene (also called Nucleocapsid) and is abbreviated NP.

Hemagglutinin, neuraminidase, and matrix code for proteins that are on the surface of the influenza particle. These surface proteins are important because when the flu enters your body, these proteins are the ones that your immune system can potentially see — and guard against.

The Nucleoprotein gene is important is determining the host-specificity of a particular flu strain — a fancy way of saying which critters that particular strain can infect.

The importance of H and N

As noted in the preceding section, hemagglutinin and neuraminidase are abbreviated H and N, respectively. If you've been reading about the flu or hearing about it on TV, you may have heard a particular strain described in terms of H and N and some numbers — for example, avian influenza H5N1.

There are 16 different hemagglutinin proteins and 9 different neuraminidase proteins, but each individual influenza A virus has only 1 type of each protein. The numbers in a strain's name (H5N1, for example) indicate which neu-rominidase and hemagglutinin proteins that particular strain has.

All the H and N types occur in waterfowl, but only some of them occur in humans. At this time, two types of influenza A are circulating in the human population: H3N2 and H1N1. Ideally, avian strain H5N1 — the bird flu you occasionally hear about in the news — won't mutate to easily infect humans and be transmittable between them, and make it three. Right now we can catch this strain from handling infected birds, but we don't seem to pass it to other humans, so it can't spread in the human population. At least not yet.

NP: The host-specificity gene

From the earlier section as well as the reports in the newspapers these days, you know that humans aren't the only ones plagued by the flu. Birds get the flu, too, specifically waterfowl like ducks and geese. But people and birds don't usually get the same strain of the flu because flu strains tend to be species-specific. A human cell is different from a goose cell, and the flu strains that replicate in geese don't usually infect humans. In short, flu strains tend to be specific to a particular groups of animals. The gene that controls (in large part) which species of animal a flu strain can best infect is the NP gene.

The H and N types determine how readily your immune system can see the flu. If a particular strain is an H-N type that your body has seen before, then you've got a jump on it. If it's one your body hasn't seen before, then the flu has the jump on you. The NP gene determines whether the flu can see you, that is, whether it can reproduce in your body. The worst possible combination is an NP gene that lets the flu chow down on you wrapped in surface proteins that take your immune system by surprise.

Who gets the flu and from where

There are five major groups of Influenza A strains. In addition to the strains that infect humans and waterfowl (ducks and geese), there is a group of flu strains that infects horses, one that infects pigs, and one that infects seagulls (which, contrary to what you may reasonably assume, are not waterfowl).

By taking a closer look at the genetic sequences of the NP gene (the one that determines which species a flu strain can best infect), scientists find that it's usually easy to tell the human flu strains from the waterfowl flu strains from horse flu strains and so on because flu strains are species specific. Seagulls tend to get seagull flu, horses tend to get horse flu, pigs tend to get pig flu, and so on. But not always.

One of the most interesting conclusions that can be drawn from an analysis of the sequences of Influenza A is that, much as we tend to think of this little beast as a human virus, its home base appears to be waterfowl. It seems that every so often, a flu strain jumps from waterfowl to some other susceptible species where it persists for a while and then goes extinct and/or is replaced by another strain. All the seagull, human, pig, and horse influenza A strains currently in circulation were probably acquired from waterfowl within the last 100 years. Understanding what causes these events is particularly important today, when public health officials and governmental agencies are coping with the very real possibility of another devastating flu pandemic.

The evolutionary history and relatedness of different flu strains

The RNA sequence information allows scientists to reconstruct the family tree of the flu and figure out how the five major groups of the flu are related to each other. Humans, for example, have influenza strains similar to those in pigs; in fact, it's not unheard of for pigs to catch the flu from people, and vice versa. (The Swine flu scare of 1976 is one example).

In addition to helping scientists formulate hypotheses about the relationship between the different flu groups, surveying the RNA sequence variation for a collection of flu strains allows them to measure the amount of variation in the different groups of the flu. This information can provide information about underlying evolutionary processes.

An important finding is that not all groups have the same amount of variation in the host-specificity protein. A collection of human flu strains has quite a lot of variation, as do the flu strains that attack horses, pigs, and seagulls. But the flu strains infecting waterfowl are much more uniform at the protein level. The waterfowl strains have the greatest diversity of surface proteins — all the different H and N molecule types — but when it comes to the proteins important for actually chowing down on ducks and geese, most of the different waterfowl flu strains are about the same. There just isn't very much variation at the protein level, even though there's variation at the RNA level. (Remember that there could be changes at the nucleotide level that don't affect what protein is made.)

Scientists know that mutations are common in RNA replication, but when they look at the host specificity gene of the flu strain that infects waterfowl, they find that most of the variation is neutral; it doesn't affect proteins. Mutations that affect protein structure are certainly always appearing, but in the waterfowl population, it seems these mutations don't increase in frequency. We know they happen, but we don't see them. From this, we can speculate that the flu strains that are infecting waterfowl have, at least for now, gotten to be about as good as they can be.

Duck, duck, goose: How waterfowl flu spreads

Interestingly, the waterfowl flu strain doesn't usually tend to make waterfowl very sick, and it doesn't have to be spread from one duck or goose to the next. Waterfowl are really good at making bird droppings (if geese or ducks have ever taken up residence at your local park, you know what I'm talking about), and these droppings contain huge amounts of flu virus. The flu in ducks lives in the gastrointestinal tract, not the lungs, and those droppings spread the flu virus through the entire pond where it can easily infect many other ducks.

Every once in awhile, some other organism comes into contact with some of the waterfowl flu — easy to see how that can happen when you think of a pond full of influenza infected droppings. When this happens, the chance exists for the waterfowl flu to infect a different kind of animal. Scientists have evidence of several different cross-species transfers from ducks to a variety of mammals.

Which strains are evolving faster

Because of differences between neutral and non-neutral sequence variation, scientists can tell that the flu strains in waterfowl are evolving differently than the flu strains in the other groups. Although all five groups have variation at the nucleotide level, the waterfowl group doesn't have any significant variation at the protein level. The flu in waterfowl seems to be at an evolu-tionarily stable point, evidence that selection is acting differently in different groups of flu.

Take-home message: The host-specificity protein in waterfowl has evolved to some optimum level. Think of it as a peak on the adaptive landscape. (If you don't know what the adaptive landscape is, refer to Chapter 6.) Mutations that occur that change the protein are just as rapidly eliminated from the population, so we don't see them. What we do see are changes in the RNA that don't affect the protein.

Jumping from one species to another

Because different species tend to be infected with flu viruses specific to their species, scientists who collect flu strains simply have to know which organism it came from to make a pretty good prediction about which strain it is. Take a strain from a horse, for example, and analyze its genetic sequence, and chances are, it'll be recognizable as a horse flu. Sometimes, however, scientists find a sick animal, collect the specimen, do the analysis, and find an unexpected flu strain.

When an animal has been infected with a flu strain different from the one expected, scientists use phylogenetic analysis again, because in making the tree of the five groups they can spot when a particular strain is out of place. If they find that a horse is sick with the flu, for example, they only have to check the flu sequence to discover whether it's the strain they'd expect (one specific to horses) or one they didn't expect (a strain that horses don't typically get). They can also tell in what ways it's different: that it's way different from the horse flu, for example, and just like the bird flu.

Such sequence analysis tells scientists that sometimes flu from waterfowl can infect other groups. Its been shown to move to horses and pigs, as well as other animals not usually associated with the flu, such as whales, seals, and mink. Now of course we're finding these flu strains in unlucky people in Asia as well.

That waterfowl flu can be transferred to other species is significant because the waterfowl population contains the highest diversity of surface proteins. Were a change to occur in a waterfowl flu that would allow it to reproduce in human cells, for example, it could take our immune system by surprise. The strain wouldn't look like any of the other flu strains that your immune system has responded to before. Although your immune system would respond to the infection, it'd be starting from scratch. If the flu strain is particularly virulent, you might not have that much time.

Bottom line: The waterfowl flu population is a source of cross species infection, and it's implicated as a source of new variants that have infected the human population in the past, a process we may be seeing the early stages of with the H5N1 avian flu.

Learning from the Past: Flu Pandemics

The term pandemic refers to an epidemic of an infectious disease that occurs over a large area, such as a continent or even the entire world. The black plague that swept from Asia through Europe in the mid-1300s is a good example of a pandemic. Flu pandemics can occur when a flu strain that appears in the human population is so different from the circulating strains that the worldwide population is completely susceptible. As a result, this new flu is able to sweep rapidly through the entire world. Often, influenza pandemics are given names related to the regions where the virus first appeared.

In an average year, the Centers for Disease Control and Prevention estimates that influenza kills more than 20,000 Americans. The global total is harder to estimate accurately, but the World Health Organization places it at somewhere between 250,000 and 500,000 deaths per year. Pandemic years see a much, much higher death rate — sometimes into the millions.

Scientists have information on flu pandemics dating from the pandemic of 1889 to the more recent 1977 pandemic. By studying the genome sequences of the more recent strains and the antibodies of the victims in the more distant strains, researchers have been able to piece together the origin of these pandemic strains. The following sections take you on a little stroll through the pandemic flus of the last hundred years.

Pandemics of 1889 and 1900

Because sequence information isn't available, scientists have to rely on seroarchaeology — the search for antibodies in preserved tissue samples — for these pandemics. The antibodies in these tissue samples can't tell researchers about all the genes in the flu strains responsible for the pandemics, but they can identify the H and N proteins, because these proteins are the external parts of the flu virus that the body's immune system produces antibodies against. So we know that the 1889 and 1900 pandemics were caused by an H2N2 and an H3N8 strain, respectively.

The Spanish flu (1918)

This influenza pandemic strain is the first one for which researchers have complete sequence information. Sequence analysis of the genes from this strain suggests that it was the result of an H1N1 avian influenza virus entering the human population. This finding is important, particularly in light of recent concerns about the possibility that the rare but often fatal H5N1 avian flu will change to a form that can be easily spread between humans. The Spanish flu is an example of a flu strain from one host (waterfowl) infecting a new host (us).

The Asian flu (1957)

The Asian flu strain (H2N2) contained 5 gene segments from the circulating 1918 strain and 3 new ones — the different versions of the H and N genes, as well as a gene involved in RNA replication. These changes were almost certainly the result of a reassortment event with an avian strain of influenza A, and the new strain replaced the H1N1 strain from which it was derived.

The Asian flu pandemic was the result of a reassortment event between the original H1N1 strain and an avian strain that contributed the 3 new genes. These two strains ended up in the same host, the various segments got mixed up during the flu's replication, and out popped something new.

The last great pandemic

The 1918 Spanish-flu pandemic killed more people than any other disease outbreak in human history. It is estimated that between 20 million and 40 million people died of the flu during the 1918-1919 flu season. Furthermore, although influenza is usually most fatal among the very young and the very old, the 1918 flu had very high mortality across all ages.

It was not until very recently that scientists learned the complete genome sequence of the 1918 flu. No samples suitable for genome sequencing were known before that. However, in 2005, viral nucleic acids were isolated from the bodies of victims who died in Alaska during the epidemic. Their bodies had been buried in the Alaskan permafrost and had remained frozen until the present. These frozen samples were sufficiently preserved that scientists could isolate the viral genetic material (RNA) from them.

Before the rediscovery of the 1918 flu, scientists thought that the 1918 pandemic originated in a fashion similar to the 1957 and 1968 pandemics — that is, via the acquisition of several new genome segments from a reassortment event with an avian strain. Scientists now know, however, that all eight segments of the 1918 strain are very similar to avian strains, and the current hypothesis is that the 1918 pandemic was the result of an avian strain entering the human population.

The Hong Kong flu (1968)

When the Asian flu strain (H2N2; see the preceding section) went through a second reassortment with a different avian influenza strain, the H3N2 strain was born. This strain had the same N gene but had acquired 2 new genes, including a new H gene.

This is another example of a reassortment event between a human strain and an avian flu strain. This H3N2 strain replaced the H2N2 strain in the human population.

The Russian flu (1977)

In 1977, the H1N1 flu last seen in humans in 1950 reappeared. Because most people older than about 20 had been exposed to H1N1 flu, the resulting epidemic was not severe enough to be labeled a true pandemic, but it was certainly a significant event.

Scientists don't know exactly where this strain came from, but it has been suggested that it escaped from a laboratory somewhere. Support for this hypothesis is not just the strain's sudden reappearance, surprising though that is. More convincing is that the 1977 H1N1 was almost identical to the 1950 H1N1 — in all those years, this strain hadn't been evolving.

Remember that this year's flu isn't quite the same as last year's flu, and next year's flu won't be the same as the flu that follows it. The point? Twenty-seven years without change is hard to explain — hence the suggestion that maybe those 27 years were spent in a test tube in a freezer!

Since this event in 1977, two influenza A lineages have been circulating simultaneously in the human population: H3N2 and H1N1.

Fighting Back: The Art and Science of Making Flu Vaccines

Vaccines work because of immune-system memory, which essentially prepares your immune system to defend you immediately when it senses that a previous invader — or one like it — has returned for a repeat performance. (Refer to "Act 3: Building up the guard," earlier in this chapter, for details.) By combining their knowledge of how immune-system memory works and how flu strains evolve, scientists were able to come up with a strategy for vaccination.

A vaccine is a substance introduced into your body to trick it into bringing forth an immune response. This response creates antibodies, which can then protect you from the microorganism the vaccine is designed to imitate. Typically, the vaccine is developed in some way from the organism from which immunity is desired. The vaccine can be made from the whole organism or from some of the external parts, because those are the parts that your immune system recognizes. Moreover, vaccines can consist of live or dead organisms.

When you get vaccinated against the flu, you're actually receiving three different vaccines at the same time. That's because three major lineages of the flu are currently in circulation in the human population: two different types of influenza A and one influenza B strain. Because these three kinds of viruses are different enough that no single vaccine can protect against all of them, one strain from each of the three viral groups is chosen for vaccine production.

Dead vaccines

Initially, a dead vaccine appears to be the better vaccination option, if you consider that the vaccine consists of disease agents that you're going to be introducing into your body. But producing a dead flu-virus vaccine has a major disadvantage: You need to make a lot more of it than you do of a vaccine based on a living virus.

The problem doesn't sound so critical; surely manufacturers can just make more. But in fact, the production of the annual flu virus is a major undertaking. The viruses created for vaccine production are grown in chicken eggs, where the flu happily multiplies. But think about all the eggs you need to grow enough flu vaccine for all the people who need to be inoculated. In the United States alone, you need more than one quarter of a billion eggs. In addition to requiring a lot of eggs, this process takes a lot of time. It's approximately half a year between the time when flu-vaccine production starts and when the first doses are delivered for distribution.

Further complicating this long period of production is the fact that the decision about which three influenza strains should be used to make the vaccine has to be made months before the beginning of the flu season. Because the flu is always evolving, the earlier the decision is made, the less likely it is to be correct. Thus, as you may have noticed, flu shots work better in some years than in others.

Cell-based flu vaccines are newest thing in the manufacturing of flu vaccines. Rather than incubate the virus in eggs, the cell-based techniques grow vaccines in laboratory cell cultures. One of the main advantages of this technique is that, by not being dependent on a huge supply of eggs, the makers don't have the same quantity limitations or contamination challenges.

Live vaccines

The advantage of using a live flu virus in a vaccine is that a little goes a long way. The virus replicates in a person's body; the immune system ramps up to defeat the infection; and the result is immunity against further infection. Unfortunately, one of the risks is that the live virus is going to reproduce in your body before your immune system knocks it out. That situation complicates the vaccine-production process, because vaccine designers have to create a strain of the virus that won't make you sick but is still similar enough to the more virulent virus to kick your immune system into action.

Using evolution to create less harmful viruses for vaccines

The flu's ability to exchange genome segments with other influenza viruses has been nothing but trouble for humans. But scientists have turned the tables, using this mechanism to fight back. They've discovered a clever way to create new virus strains that are disguised on the outside with all the bits of the virus humans want to be immune to, but filled on the inside with a collection of different alleles (slightly different forms of the flu genes) that aren't that good at making people sick. Voilà! This "cream-filled" virus reproduces in the body well enough to cause the immune system to attack it, but not so well that the person actually gets ill.

The live-vaccine virus will begin evolving to attack you more efficiently as soon as it starts reproducing in your body. The plan, of course, is that long before enough time has gone by for the virus to develop greater efficiency at reproducing in you, your immune system will have had a chance to defeat it. That's the plan, anyway.

Using evolution to keep Vaccines safe

The fly in the live-virus vaccine ointment is evolution, of course. Even though the live virus used for vaccine production has been constructed so that it's weak and can't reproduce rapidly in the human body, it may not stay that way. As the weakened live virus reproduces, it will produce offspring with random mutations, some of which may improve their ability to live in a person. These progeny in turn will produce offspring, and some of them may be even better at beating the body's defenses. If enough time goes by, a reasonably harmless virus designed for use as a vaccine could end up as one that's much better at eating human cells for lunch.

So can scientists do anything to reduce the chance that evolving viruses will get the best of us? Short answer: Yes, by using the evolutionary process to our advantage.

The influenza that makes you sick is happiest at your normal body temperature. That shouldn't be too surprising, because it lives in the human body, and selection has had a lot of time to tune the flu for 98.6 degrees Fahrenheit. That's one of the reasons your body turns up the heat when you get sick: It's trying to burn the virus out.

Scientists have used this bit of knowledge to their (and your) advantage: They decided to turn down the heat in the laboratory. By growing the flu in colder temperatures in the laboratory, they evolved a flu strain that reproduces well at lower temperatures. These new, cold-adapted flu strains aren't very good at growing at human body temperature.

Why did scientists want this cold-adapted flu strain? Some very clever vaccine designers realized that it's colder in your nose than it is in your lungs. Your lungs are deep inside your toasty-warm body, but your nose is right up there on the edge, breathing air that's generally below 98.6 degrees — especially during the flu season.

By creating a vaccine virus that can reproduce at temperatures slightly below your body temperature but not very well at your body temperature, the designers developed a very nifty live vaccine indeed. You can spritz a little bit of it in your nose, and although it will reproduce and stimulate your immune system to respond, it can't get from your nose into your lower respiratory system, where it's just too hot. (That's why when you get the live flu virus, the nurse spritzes it into your nose.)

The power of evolution is such that, given enough time, even this virus can get out of your nose. With each round of replication, variants that are a little bit better at growing in high-temperature cells may be produced by mutation. For this reason, people with weakened immune systems, including very young children and elderly individuals, should not get the live-virus vaccine.

Predicting the future to make next year's vaccine

It's a shame that each year, designers have to redesign the flu vaccine to account for recent flu evolution. It's even more of a shame that getting the vaccine right is so hard. The health officials who make these predictions use a variety of information concerning past patterns of flu outbreaks, as well as which strains are sweeping through different parts of the world. Given the rapid evolution and incredible variation of the flu population at any given time, it's amazing that vaccine designers do as good a job as they do.

Understanding the details of flu evolution can help scientists predict what flu strain may be coming next. We don't know what's going to happen next year (that flu isn't here yet), but we do know what has happened in the past (samples of those flu viruses are in the freezer, and their genetic sequences are known). Scientists can use this information about past years' flu strains to reconstruct the history of flu evolution. Phylogenetic analysis lets them make a family tree for the past years flu strains (refer to Chapter 9 for information on how to construct these trees). This tree is scientists' and virologists' best estimate of what begat what in the never ending progression of new flu strains.

Sc-S^HS^ Robin Bush, Walter Fitch, and colleagues used the sequence of the H gene to create such a tree for the H3N2 group of influenza A currently infecting humans. They used data from 1984 through 1996, and they were able to identify which flu lineages in each year gave rise to the flu strains that made people sick the next year.

They then set out to figure out what was it about those strains that predisposed them to being the ancestors of the next big wave of infection. There are lots of little branches of the flu family tree around at any one time — what we want to know is from which of those little branches will the strain that's going to sweep through your school district arise. We want to make a vaccine against that particular sort of flu strain. The trick is figuring out which strain it is.

By figuring our how the flu is evolving in response to our immune system, we learn clues that might help us predict what next years will look like and fine tune vaccine design in advance.

Making and testing predictions

With 12 years of data in this study, Bush and the others had 11 separate pairs of data sets to test different predictions. First, they looked really hard at the flu RNA sequences for 1985 and asked what they could measure about the 1985 flu strains that could be used to predict the major 1986 flu strain. They then went back to the 1986 records to check to see whether their predictions were right. Then they did the same thing ten more times to test their ideas further. The following section explains what they discovered.

Anticipating next year's flu: The clues

Possible factors indicating that a particular flu strain is likely to give rise to next year's outbreak include the changeability of a particular strain (maybe the strain that changes the most is the one that spawns the following year's flu) and the changeability at a particular region of the virus (maybe a virus that changes in a particular way is the culprit responsible for next year's flu).

What Bush and colleagues discovered is that the first one wasn't the key, but the second one was:

The degree of general change in a particular strain.

Maybe the best candidate is the flu strain that's changing the most overall. Because this strain is changing at a faster rate, maybe it will be the one that gets around your immunity to last year's flu.

Sounds good, but this predictor didn't turn out to be an especially accurate one. Having lots of random mutations is no guarantee that those mutations result in genes that are better at fooling our immune system.

The degree of change at particular regions.

Maybe the flu strain most likely to cause next winter's problems is the strain that's changing in some particular way — changing the most at just the regions where the H protein is interacting with the human immune system, for example. This strain would be more likely to catch your immune system napping!

Turns out we don't know the exact details about which parts of the flu are the most important in interacting with our immune system. But by studying the flu's evolution we can find out! By looking at the flu stains that had been successful in the past the researches were able to identify specific sites in the H gene where changes were associated with this success.

Bush and coworkers discovered that the more mutations that occurred in a particular isolate at just these key locations, the more likely that isolate was to spawn the next year's flu. These are the flu strains that seem to be rolling with the punches of the human immune system, and they're ones we need to be prepared for in the future.

Making a more universal influenza A vaccine

Because the H and N proteins that cover the surface of the flu are so variable, designers have to reassess which particular strains they should use as models for each year's vaccine (refer to the preceding section). It's a shame that the surface of the flu is so variable from one viral strain to the next; things would be a lot more convenient if they were all the same.

Fortunately, one part of the surface of influenza A is very uniform across many isolates: the Matrix, or M, protein. When your immune system finds the flu in your body, it doesn't seem to pay much attention to the M protein, as H and N proteins are much more obvious targets. But that protein is there, and it's possible to make your immune system see it, too.

tt^HSil Walter Fiers and colleagues set out to do just that. They manufactured virus-sized particles covered with the M protein. Once Fiers and company had the little M-protein-covered particles, they tested them on mice to see whether they could stimulate an immune response that would protect the mice from influenza A. They found that even though the mice still got sick, they got much less sick than the mice that had not been vaccinated with these particles.

The fact that it is possible to make the immune system respond to the M protein raises a question: How would evolution of the influenza A M protein be affected by the presence of many such vaccinated individuals? The M protein is very uniform among different influenza A strains right now. If many people had antibodies that attached to the M protein, however, that might select for increased variation in the M protein. All the influenza A viruses make their M proteins in about the same way, but that doesn't mean they have to. Perhaps the viruses can make the M protein in many ways.

cSs^*^ Walter Gerhard and colleagues set up an experiment to see how influenza A tj/ viruses evolved in the presence of antibodies that targeted the M protein.

They took mice with weakened immune systems and gave them antibodies against the M protein. They infected the mice with an influenza A strain, and after several weeks, they examined the descendents of the original viruses to look for changes in the M protein.

Because these mice had weakened systems, they were not able to eliminate the flu virus rapidly. Because the flu virus was replicating in the mice, random mutations were occurring, and it turned out that some of these mutations did result in changes in the M protein. The scientists found only two new variants of the M protein, however, rather than the huge number of variants that would occur in the H protein in the same sort of experiment.

These experiments are encouraging, because the results suggest that the influenza particle may have only a limited number of ways to make a functional M protein. With any luck, it might be possible to vaccinate against all influenza A viruses at the same time using a vaccine based on M protein.


This type of experiment shows the potential of incorporating evolutionary information in drug design. By trying to target a part of the flu that seems more evolutionarily constrained, it may be possible to create a more universal vaccine for use in the event of a dangerous outbreak. In the case of an extremely pathogenic flu, a vaccine that offered only partial protection would be much better than no protection at all. A vaccine like this could provide one line of defense against the threat of the H5N1 avian influenza.

In fact, British and American researchers have recently tested such a vaccine (one based on the M protein), and the results are very encouraging. They indicate that the vaccine could provide universal protection against the Influenza A virus.

Part V

Was this article helpful?

0 0

Post a comment