October 05, 2005
1918 Killer Pandemic Was An Avian Flu

A complete recreation of the 1918 H1N1 pandemic flu strain shows that it was an avian flu.

Scientists have re-created the “Spanish flu” virus that killed up to 50 million people in 1918-19 and shown that it shared traits with the H5N1 strain of avian flu.

An analysis of the re-created pathogen has shown that, like its modern cousin, it began as a bird virus and jumped species into humans with mutations that made it peculiarly virulent and lethal.

Dr. Jeffrey Taubenberger of the US Armed Forces Institute of Pathology and other scientists have just published papers in Science and Nature demonstrating that the 1918 virus was an avian virus in origins.

"We now think that the best interpretation of the data available to us is that the 1918 virus was an entirely avian-like virus that adapted to humans," Taubenberger told reporters in a telephone briefing.

"It suggests that pandemics can form in more than one way."

The more deadly 1918 pandemic virus is unlike the 1957 and 1968 flus in that the 1918 flu did not recombine with human influenza strains. That the 1918 strain did not recombine with human influenza strains and at the same time that it was orders of magnitude more lethal is probably not a coincidence.

"We now think that the 1918 virus was an entirely avian-like virus that adapted to humans," said Mr. Taubenberger. This is a different situation than the last two pandemics we had, the Asian flu in 1957 and the Hong Kong flu in 1968, which are mixtures in which a human-adapted influenza virus acquired two or three new genes from an avian influenza source. So it suggests that pandemics can form in more than one way, and this is a very important point."

He says it also suggests that the current Asian bird flu, known by its scientific designation H5N1, could evolve into a human killer with just a few more mutations that allow it to jump more efficiently among people.

"It suggests to us the possibility that these H5 viruses are actually being exposed to some human adaptive pressures and that they might be acquiring some of these same changes," he added. "In a sense, they might be going down a similar path that ultimately led to 1918."

This is the most important fact here: The 1918 H1N1 influenza virus did not need to co-infect a human and swap genes with a human influenza strain in order to gain the mutations needed to cause a highly lethal human influenza pandemic. That ups the probability that H5N1 could start a human pandemic.

The scientists took the recreated 1918 virus and performed experiments on it in a CDC Atlanta lab to identify proteins and genes key to its virulence.

"We felt that we had to re-create the virus and run these experiments to understand the biological properties that made the 1918 virus so exceptionally deadly," said Terrance Tumpey, a flu researcher at the Centers for Disease Control and Prevention and lead author of the Science study. "We wanted to identify the specific genes responsible for virulence, which we feel will advance our ability to prepare vaccines and make antiviral medicines that are effective against future pandemic strains."

Although the genetic data has been made part of a public database, the 10 or so vials of the virus itself - grown in human kidney cells - are contained under tight security guidelines set for potential biological weapons at the CDC's lab in Atlanta.

Fox News has a good article reporting that some of the mutations which H5N1 has already picked up are similar to 1918 H1N1 mutations and probably are moving H5N1 closer to human transmissibility and a human pandemic.

The good news is that the H5N1 flu bug still has a long way to go. The 1918 bug seemed to need several changes in every one of its eight genes. The H5N1 virus is making similar changes but isn't very far along.

"So, for example, in the nuclear protein gene we speculate there are six genes crucial [for human adaptation]," Taubenberger says. "Of those six, three are present in one or another H5N1 strain. But usually there is only one of these changes per virus isolate. That is true of other genes as well. You see four, five, or six changes per gene in the 1918 virus, whereas H5N1 viruses only have one change or so. It shows they are subjected to similar [evolutionary] pressures, but the H5 viruses are early on in this process."

Which H5N1 strains is Taubenberger comparing the 1918 H1N1 strain to? How old are the strains he is comparing to? Has he compared to any H5N1 strains isolated from recent Indonesians who have recently died from bird flu?

H5N1 appears to have picked up important mutations on the road toward human adaptation including one that has made it much more lethal in mice.

There is one ominous sign. It's in a flu gene protein called PB2. A single change in this gene makes H5N1 extremely deadly to mice. The same single change helps bird flu to adapt to mammals.

For example, the change in PB2 was seen in six of the seven H5N1 viruses spreading among captive tigers in Thailand.

Robert Webster of St. Jude Children's Research Hospital showed in a PNAS paper in the summer of 2004 that H5N1 has become much more lethal in mice and this is an indicator that H5N1 is becoming better adapted to mammals. Webster also points to an expanded range for H5N1 including tigers and domestic cats. The expanded range gives the virus more ecological niches in which it can further adapt to mammals and pick up more mutations that would help it become transmissible in humans.

Also check out a CDC report on possible H5N1 transmission between tigers. I realize some of my readers think I'm being excessively alarmist by writing posts about avian flu. But I just do not see humans as so different from all the other mammals as to think that a virus that is hopping between a bunch of species is going to draw the line and avoid humans. These new reports about the 1918 H1N1 strain which hopped from birds to humans and the parallels with H5N1 strike me as a strong reason not dismiss the threat this virus poses.

The New Scientist has a good article on the latest findings. The 1918 H1N1 virus was less dependent on cellular machinery to replicate.

Meanwhile, Terrence Tumpey at the US Centers for Disease Control in Atlanta and colleagues used the sequences to rebuild the virus itself, and infect mice with it. They report this week that unlike other flu viruses, 1918 does not need a protein-splitting enzyme from its surroundings to replicate, instead using some hitherto-unknown mechanism. And as in 1918, it rapidly destroys lungs (Science, vol 310, p 77).

Pathogens that jump species are a lot more lethal that pathogens that have been transmitted primarily between members of a species for a long time. The knowledge that the 1918 influenza was from birds and had no recombination with human influenza strains should give pause to anyone wondering whether H5N1 poses a serious threat.

Update: It probably took only a couple dozen mutation to make the 1918 influenza into a massive killer.

The bird flu viruses now prevalent share some of the crucial genetic changes that occurred in the 1918 flu, scientists said, but not all. The scientists suspect that with the 1918 flu, changes in just 25 to 30 out of about 4,400 amino acids in the viral proteins turned the virus into a killer. The new work also reveals that 1918 virus acts much differently from ordinary human flu viruses. It infects cells deep in the lungs of mice and infects lung cells, like the cells lining air sacs, that would normally be impervious to flu. And while other human flu viruses do not kill mice, this one, like today's bird flus, does.

As these researchers advance along in their work they are going to come up with much better metrics for measuring how far the H5N1 avian flu is from being capable of creating a pandemic in humans.

Update II: Henry Niman claims the 1918 flu was the product of a human flu and swine flu recombination.

Share |      Randall Parker, 2005 October 05 09:56 PM  Pandemic Signs

Mr. Econotarian said at October 6, 2005 8:30 AM:

If we can determine specific protein mutation sequences to go after, perhaps the use of monoclonal antibodies developed specifically for them could be of value.

James Bowery said at October 6, 2005 8:58 AM:

Is there a name for parasites that inhabit multiple hosts -- not serially during different parts of their lifecycle -- but in parallel, so they aren't dependent on any particular host species?

Isn't it obvious why parasites of this character would be most lethal and why they should be of particular concern?

There really should be a name for this kind of parasite.

jerryofva said at October 6, 2005 11:17 AM:

It looks like we will not have an H5N5 outbreak in human-to-human form until the end of 2006 at the earliest. For those of us old enough to remember the swine flu panic in 1976 when a fatal case of H1N1 was contracted by a soldier at Fort Dix, a year lead will be more then enough time to prepare for a mass vaccination campaign.

Karen said at October 27, 2005 4:16 PM:

Q: What "mutation" of the virus would have to happen for it to change to person-to-person, and what factors would cause that mutation? -- Sarah (Last name not provided), Denver, Colorado

A: That's a question that a lot of virologists are trying to answer. And what we're trying to do is compare the genetic sequences of H5N1 with other known human pathogenic viruses. What we know, for example, is that mutations on the hemagglutinin gene -- that's the "H" in the H5N1 -- can allow the virus to bind better to cells in the human respiratory tract. There are also mutations in a gene called PB2 that seems to be important for infections into humans.

How quickly can it mutate? That's a very difficult question to answer. There are a number of ways in which the virus can mutate. It can mutate gradually over time, or it can recombine in a co-infection in a person or in an animal, and when recombination happens, that can introduce a lot of mutations very rapidly.

Q: Wouldn't we assume that this virus would be potentially mutating in infinitely different ways in different instances? -- NPR's Robert Siegel

A: That is quite correct, the virus is mutating randomly and it's only when a particular mutation is advantageous for it to infect a particular new host that the mutation becomes selected and takes over.

Mary T said at November 28, 2005 7:39 PM:

If you go to the Fox news website, there is an article that just came out today (Nov. 28th) that states China says the bird flu virus in humans is mutating.

In other words it's genetic order has changed in the strain they see in China. As I understand that is what has global leaders and scientists so afraid; in that it could change in a way that would start "the first match into the dried forest." I imagine they are living "on pins and needles, hoping it does not happen; at least not this year so that maybe by next year we will be more preparared for a pandemic. The thing that concerns me is that China is known for covering up their misfourtunes from the world. That is the last thing the CDC and other official need, as they need so very importantly to be able to jump on the situation as soon as there is a person to person virus. Not just China, but poor countries may not be able to monitor all of what is going on with their people letting this virus loose into the world and out of it's pandora's box.

The article as mentioned above reads:


Chinese labs have found that the genetic order of the H5N1 virus seen in humans infected in China is different from that found in humans in Vietnam, Xinhua news agency reported Monday.

In China's human cases, the virus has mutated "to a certain degree," health ministry spokesman Mao Qun'an was quoted as saying.

"But the mutation cannot cause human-to-human transmission of the avian flu," he noted.

Brian Schwartz said at June 22, 2007 1:33 PM:

were there any reports of sick or dead birds prior to the 1918 pandemic?

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright ©