The annual flu vaccine only lasts a season because it triggers immune antibodies that specifically target a part of the flu virus that changes every year. But what if it were possible to target a part that did not change so frequently, and this part was the same in different strains so that one antibody could target many flu strains: go for breadth as opposed to specificity? It seems that one team of Howard Hughes Medical Institute (HHMI) scientists may have found such an antibody, called CH65. They write about their discovery in the 8 August issue of the Proceedings of the National Academy of Sciences.
Under study leader Dr Stephen C. Harrison, an HHMI investigator and Professor of Biological Chemistry and Molecular Pharmacology and of Pediatrics, at Harvard Medical School and at Children’s Hospital, Boston, the team turned for inspiration to the diversity of the human immune system, as Harrison explained to the press:
“Our goal is to understand how the immune system selects for antibodies and use that information to get better at making a vaccine that will take you in a direction that favors breadth over specificity.”
When the influenza virus enters our body, our immune system reacts by producing antibodies that attack antigens, mostly those on the outer protective coat of the virus. The virus has a number of different antigens, and each of our immune systems responds slightly differently, producing a diverse range of antibodies across a human population but not in one individual.
There are several strains of flu virus, and these mutate frequently, with most of the change being in the genes that code for the glycoprotein molecules on their outer coating. It is these that our immune system mostly produces antibodies for: and the hemagglutinin and neuraminidase surface proteins in particular. When the virus mutates, these “studs” on its outer coat change shape, giving it a new look which the human immune system doesn’t recognize and can’t attack until it has produced a new toolkit of antibodies, by which time the virus has invaded and started multiplying, and the individual comes down with the flu.
The flu vaccine works by giving us a head start on this process. When the vaccine enters our body, it already has (if the vaccine designers have “guesstimated” the strains that will circulate in the next flu season correctly) the new antigens, so our immune system starts producing the new antibody toolkit ahead of the anticipated virus invasion.
So developing an effective annual flu vaccine relies on anticipating specifically which antigen studs the new flu strains are going to be wearing on their coats in the coming season. Harrison and colleagues describe this as:
“Seasonal antigenic drift of circulating influenza virus leads to a requirement for frequent changes in vaccine composition, because exposure or vaccination elicits human antibodies with limited cross-neutralization of drifted strains.”
However, what if, by looking across the diverse range of human immune system responses to flu, you could find antibodies that attacked a part of the virus that did not change so frequently?
You could do this if you had genomic technology that allowed you quickly to scan the molecules in people’s immune systems. This is what Harrison and colleagues managed to obtain, with the help of collaborators at Duke University in Durham, North Carolina.
“What this allows us to do is get a snapshot of the different kinds of antibodies being made in a person in response to a vaccine,” said Harrison.
And to their surprise, and rather unexpectedly, they found an antibody that recognized multiple strains of the flu virus: the human monoclonal antibody CH65.
They were surprised because scientists had previously believed it wasn’t possible for antibodies to target the part of the flu virus that CH65 appears to reach.
CH65 targets a part of the hemagglutinin surface protein that the virus cannot mutate as readily without reducing its ability to infect human cells. The part is the “receptor pocket” that recognizes the receptors on human cells that the virus binds to in order to gain access, enter cells and start hijacking their resources. If this part were to mutate, then it would fail to recognize the human receptors and the virus would fail.
Harrison said many scientists had assumed because of the larger size of antibodies compared to their target sites, then any that targeted the receptor binder area would also target the surrounding, more changeable areas, so if they mutated, then the antibodies wouldn’t bind.
But it appears that CH65 binds so tightly to the receptor pocket that it retains this ability even when surrounding areas mutate.
To find CH65, Harrison and colleagues started with cells from a donor who had received the 2007 flu vaccine. Using the new genomic tools they generated a suite of antibodies from the donor cells to test against several flu strains. CH65 was one of these.
With the help of collaborators from the US Food and Drug Administration (FDA), Harrison’s team were able to test the antibodies against 36 flu strains that have emerged between 1988 and 2007. CH65 recognized and successfully intercepted the hemagglutinin from 30 of them.
When they compared CH65 with other antibodies from the same donor, the team was able to work out how the donor’s immune system had evolved to produce a range of antibodies with a broad immunity as a result of multiple virus exposures over time.
Harrison said:
“While it’s unusual to find such broadly effective antibodies to the flu virus, they may actually be more common than we realize.”
“What this tells us is that the human immune system can fine-tune its response to the flu and actually produce, albeit at a low frequency, antibodies that neutralize a whole series of strains,” he added.
But, here is an obvious question many might ask at this point: if we just go ahead and produce a vaccine based on CH65, won’t that just create a sort of “evolutionary pressure” that causes the flu virus to start mutating its receptor binding pocket? And then we would be back to square one, having to make a new vaccine every year.
That is why Harrison and colleagues would prefer to take this discovery into a slightly different direction, at least for now.
What they want to do now is use CH65 to investigate how an individual’s immune system chooses which antibodies to produce when confronted with a virus, because it doesn’t tackle all the antigens it meets. Also, if some people’s immune systems can produce CH65, then is there a way to coax everyone else’s to do the same?
So instead of going straight down the vaccine route, Harrison and colleagues want to take a step back and, as Harrison explained, try “to understand how the immune system selects for antibodies and use that information to get better at making a vaccine that will take you in a direction that favors breadth over specificity”.
“Developing a flu vaccine is currently a hit-or-miss enterprise,” continued Harrison. “We vaccinate with a virus or part of a virus and hope that the immune response will evolve in a useful direction.”
“But for viruses like influenza that mutate rapidly, we want to have a response that does a really good job at blocking both the strain of the virus in the vaccine and many related strains as well. These results point out what strategies we might employ to achieve that goal,” he added.
Harrison is now working with another HMMI investigator Dr Nikolaus Grigorieff, a Professor of Biochemistry at Brandeis University in Waltham, Massachusetts, to find out more about the structure of antibodies and how this changes as they evolve in response to vaccination. They hope to take snapshots of antibody structures over time, and thereby hopefully find a pattern that reveals how the immune chooses which structures to go for.
Harrison said other investigators may take CH65 down the clinical route. Some are talking about “therapeutic antibodies”, that can be given to severely ill flu patients, or patients with weakened immune systems, to help them fight the virus. He said CH65 was a “very interesting molecule to consider for that”.
Written by Catharine Paddock PhD