Los Alamos National Laboratory

Can We Understand Influenza?

Hippocrates of Cos, the Father of Medicine, was writing about flu in 412 BC, but people still don’t fully understand the virus that causes the disease. Los Alamos is marshaling its capabilities in detection, analysis, modeling, and genomic sequencing to learn how to preempt one of the most infectious human diseases.

On April 1, 2009, a 10-year-old boy was admitted to an urgent care clinic in San Diego County, California, with flu-like symptoms: fever, cough, and vomiting. By chance, the clinic was participating in trials aimed at developing a new influenza diagnostic, so once a sample had been taken, it was immediately analyzed.

image of magazine graphic

Sure enough, the boy tested positive for influenza A, the virus that gives rise to the flu, but the test could not identify the virus’s subtype. Influenza A is actually a large family of viruses, with a family tree consisting of dozens of major branches (the viral subtypes) and each branch splitting into thousands of twigs (genetically distinct viral strains). Knowing the subtype would have given clinicians some insight into what to expect from the virus, much the way that knowing an apple is a granny smith conjures up expectations for the apple’s taste, color, and texture.

After a local laboratory also failed to identify the virus, a sample was sent to the Center for Disease Control and Prevention (CDC) in Atlanta, Georgia, arriving on April 15. The CDC quickly concluded that the boy had contracted a new virus that had never circulated through the human population before. Two days later, the CDC received a sample taken on March 30 from a 9-year-old girl from Imperial County, right next door to San Diego County. The girl had caught essentially the same virus.

Those first two cases of what is now called the swine flu immediately raised red flags within the CDC. The new virus was identified as a unique variant of an H1N1 subtype, a “novel H1N1 influenza A virus.” Because humans had not been exposed to this H1N1 strain, it would go unrecognized by our immune systems. Thus, it was reminiscent of another H1N1 virus, the infamous Spanish flu virus, which from 1918 to 1920 infected approximately one-third of the planet’s 1.6 billion people and killed as many as 50 to 100 million, according to modern estimates.

flu table

Fortunately, the 2009 H1N1 strain appears to be much kinder than the 1918 version. This is apparent from the case-fatality rate (CFR), loosely defined as the number of people who die from a disease (and not from secondary causes) divided by the number of people who contract the disease. The Spanish flu’s CFR was about 2.5 percent, which is at least 50 times greater than what is observed for the swine flu.

Ruy Ribeiro, an influenza expert with Los Alamos National Laboratory’s Theoretical Division emphasizes how huge the difference is. “It’s the difference between 50 deaths versus 2,500 deaths per 100,000 cases.”

Ribeiro points out that the situation could be far worse. Influenza A is nothing short of remarkable in its ability to infect different species, including humans, chickens, pigs, bats and cats, whales and quails, ferrets, seals, horses, and ducks. Each species typically is susceptible to a small number of viral subtypes, but mutations will always produce new strains that can cross over to other species. The H5N1 avian flu virus (bird flu) jumped from wild birds to humans in 2003. The virus doesn’t pass from one person to another, but heaven help us if it ever mutates into one that does—its CFR is greater than 60 percent.

Can Los Alamos Find Answers?

Seasonal influenza A viruses closeup of Influenza A description

Both the swine flu and Spanish flu viruses spread easily among people and are virulent (able to cause disease), but the former is not very deadly, while the latter was. The bird flu virus is both virulent and very deadly but doesn’t spread from person to person. Structurally and genetically, the three viruses are nearly identical. Why do they affect people so differently?

“There’s no simple answer,” says Ribeiro, “other than to say it’s in the details of how the virus and host organism interact with each other. Unfortunately, those host-pathogen interactions are not well understood.”

Ribeiro and Murray Wolinsky from the Laboratory’s Bioscience Division head a 16-person cross-disciplinary team composed of scientists from the Theoretical and Bioscience Divisions, along with the Computer, Computation, and Statistical Sciences Division. The team is focusing on the host side of those interactions, trying to find genes within a cell that respond one way when the cell is infected with a high-virulence influenza virus, but a different way with a low-virulence virus. If successful, the team could have a way to assess the virulence of new viruses.

More important, however, are the experimental and analytical techniques that team members are developing, techniques that could help them unravel the cascade of molecular interactions that result when a cell responds to a stimulus. (See “Unraveling the Host-Pathogen Interaction”) If that goal were to be achieved, doctors and researchers alike would gain tremendous insight into many areas of biology, medicine, and health.

branches H5, H3, and H1 closeup of data tree description

Another large team of Los Alamos researchers, now headed by Chris Detter of Bioscience Division, is attacking the pathogen side. In a joint effort with the University of California, Los Angeles (UCLA), Detter’s team is helping to establish a global network of organizations that will continually monitor influenza and other infectious agents by gathering samples from critical sources. (For influenza, those sources are birds, pigs, and flu-struck humans). The samples will be sent to any of several automated, high-throughput sequencing laboratories, where researchers will obtain the genetic sequences of the pathogens in the samples. (See “The High-Throughput Laboratory Network”.) Estimates are that the entire genome of any influenza virus can be sequenced in less than half a day.

Having sequence data is equivalent to having the keys to the city because those data open so many research doors. Scientists have access to nearly every influenza sequence through the Influenza Sequence Database, developed and maintained by Los Alamos and managed for over a decade by scientist Catherine Macken. Los Alamos scientists use such data, for example, to understand the structure of influenza’s proteins, to follow the course of a pandemic, to develop new influenza detectors, and increasingly, to understand why some strains are more virulent than others.

Influenza Basics

Influenza A is a severely stripped-down biological entity about a million times smaller in volume than a cell. It consists of a core of genetic material (single stranded RNA instead of double-stranded DNA) packed tightly together with proteins. A protein matrix protects the core, while a protein-studded lipid membrane surrounds and protects everything. The virus is atypical in that its RNA comes in eight separate segments rather then the one long strand that is common for RNA viruses. Each segment contains a single gene that codes for 1 of 11 different proteins, with three of the genes each coding for two proteins.

Influenza’s sole purpose is to make copies of itself, but it lacks almost all of the resources to do so. The virus must infect a cell and use the cell’s resources to make proteins and help copy its genome. In addition, the virus must circumvent cellular defenses and avoid alerting the host’s immune system.

Two H5N1 avian flu viruses closeup of two H5N1 avian flu viruses. description

The latter task is complicated by two viral proteins, hemagglutinin (HA) and neuraminidase (NA), both of which protrude from the surface of the virus. HA anchors the virus to the host cell by binding to sialic acid, a type of sugar that graces the surface of cells in the upper respiratory system of mammals (or the intestines of birds). NA is important for helping newly made viruses exit the cell. Both proteins are antigens, meaning they can trigger the immune system to produce antibodies that will stick to the proteins and prevent them from functioning.

The virus counters this vulnerability by relying on antigenic drift—random mutations of the HA and NA genes that make the corresponding proteins unrecognizable to the immune system. How does that happen? Packed into the virus’s core is a protein complex that makes the complementary strand to a single-stranded RNA segment, which can be used to manufacture the protein that’s encoded within the segment’s gene. But the complex can also make a complement of the complement, that is, a copy of the original RNA segment.

The protein complex is error prone, however, and makes, on average, one mistake (mutation) every time the virus’s genome gets copied. The upshot is that the strain that infects a cell is often not the strain that leaves it. A new virus with, say, a mutated HA gene can have the mutated HA antigen already expressed on its surface by the time it leaves the cell and therefore go at least partially unrecognized by the host’s immune system. The seasonal flu viruses that plague us each winter typically are new strains that have antigenically drifted away from strains already circulating within the human population.

As an aside, over thousands of years, antigenic drift helped HA evolve into 16 separate varieties (H1 through H16) and NA into 9 (N1 through N9). A virus’s subtype is a particular combination of HA and NA, for example, H1N1.

Another of the virus’s survival strategies takes advantage of the segmented genome. Two viruses from different species, say duck and human, infect the same host, typically a pig, and produce a duck/human hybrid virus by exchanging RNA segments. The hybrid can gain the ability to cross species, say from pig to human, as was the case with the swine flu virus. (See “Creating a Hybrid”)

A Glimpse of Virulence

While every influenza A virus uses the same tactics to survive, logic dictates that high-virulence strains with the potential to cause severe illness must interact with cells differently than low-virulence strains do. “Our research shows that there is a clear distinction in the body’s response to high- versus low-virulence strains,” says Ribeiro.

For example, one of the better-known factors that influence a strain’s virulence is the length of a chain of amino acids that runs between two fragments of the HA molecule. This chain must be cut if the virus is to get its RNA into a host cell.

What happens is that the virus, bound to the cell surface by HA proteins, enters the cell by endocytosis: the cell membrane binding the virus craters and then deepens into a pocket with the virus attached to its inside. The pocket pinches off from the membrane, so that the virus is in the cell but trapped within the enclosure (called an endosome).

The cell begins to make the interior of the endosome acidic in an effort to break down whatever is inside. Under acidic conditions, however, HA changes its shape, which causes the virus’s outer membrane to fuse with the endosome’s. A pore then opens in the fused region, establishing a channel through which the viral RNA enters the cell.

RNA cannot enter the cell unless HA changes shape, and HA can’t change its shape unless the amino-acid chain gets cut. Researchers speculate that if the chain is “long,” it will protrude a bit outside the body of the protein. Shortly after HA is made, the chain can be cut by a wide variety of enzymes found in most cells. If the chain is “short,” however, it will run closer to the protein, and the cellular enzymes can’t cut it. The protein gets attached to the virus intact, and is cut by only a few types of small enzymes found outside cells in the nose, throat, and upper part of the lungs. Viruses with short-chain HA proteins therefore tend to be less virulent than those with long-chain HA proteins because the latter can infect many more types of cells.

“The length of the amino-acid chain is a strong virulence factor, but there are dozens of others,” says Ribeiro. “We hope our research will help the influenza community understand the complex interactions.”

The Big Picture

Sometime during the 2008–2009 flu season, the novel H1N1 virus gained the ability to jump ship from pigs to humans. It circulated in Mexico for several months before it encountered a little American boy, then spread across every continent in less than two months. It was a remarkable evolutionary accomplishment for the new strain on the block.

Any influenza pandemic, however, is but one battle in an epic conflict between man and microbe, a battle in which the evolutionary power of a short generation time, a mere 20–30 minutes in some bacteria, gives the microbe a distinct advantage. While it’s naïve to think of winning that war, humanity hopes to achieve a détente that will allow civilization to prosper. The steps taken by Los Alamos and researchers around the world to understand influenza and its interactions will have an effect. Not this flu season, and maybe not the next, but soon we may understand the enemy well enough to reach and sustain that détente.

—Jay Schecker

This article is dedicated to Tony Beugelsdijk, former leader of the Los Alamos High-Throughput Laboratory project, who passed away August 23, 2009.

In this issue...

+ More on this article