HIV aids deadly pathogen
Salmonella can wreak havoc in (or kill) people infected with HIV -- and not for the reason scientists have long assumed.
Salmonella typhimuriumImage: Wikimedia commons, V. BrinkmannMax Planck Institute for Infection BiologyInstead, a new study in Science shows that Salmonella's ability to cause disease in HIV patients does not appear to stem from a weakened or ineffective immune system, but an overactive one that actively protects the bacteria. The findings may help direct research on developi
can wreak havoc in (or kill) people infected with HIV -- and not for the reason scientists have long assumed.
Image: Wikimedia commons, V. BrinkmannMax Planck Institute for Infection Biology
Instead, a new study in Science
shows that Salmonella
's ability to cause disease in HIV patients does not appear to stem from a weakened or ineffective immune system, but an overactive one that actively protects the bacteria. The findings may help direct research on developing effective vaccines against the pathogen.
"In an HIV-infected person, you would expect that if you're not seeing clearance of a pathogen, it's because [the person is] not making any antibodies against that specific pathogen," said immunologist linkurl:Susan Moir;https://ugsp.nih.gov/scholars_mentors/mentors_d.asp?m=07&id=1522 of the National Institute of Allergy and Infectious Diseases's Laboratory of Immunoregulation. "But they found the opposite -- they found a lot of antibody, but it was directed against the wrong thing."
(NTS) are bacteria that infect humans who eat improperly prepared food products, which cause stomach problems in healthy people, but can be fatal in those with HIV -- particularly patients living in Africa with limited access to antiretroviral therapy.
To understand the mechanism of infection, immunologist linkurl:Calman MacLennan;http://medweb4.bham.ac.uk/ssp/Printable.aspx?id=1260 and his colleagues at the University of Birmingham in the UK and the University of Malawi in Africa exposed blood samples from healthy and HIV-infected African adults to invasive Salmonella
strains. While the blood from healthy individuals successfully killed the invading bacteria, HIV-infected blood was much less effective.
Suspecting that the HIV blood lacked the NTS-specific antibodies needed to kill the pathogen, the researchers measured the levels of immunoglobulin G (IgG) that specifically bind to the bacteria. Surprisingly, the team found that blood samples unable to eliminate the infection had higher levels of IgG than blood samples that could kill Salmonella
"That's the paradox," MacLennan said -- blood from HIV patients was less effective at killing the Salmonella
, but "these patients have lots of antibodies."
Mixing blood from healthy and HIV-infected individuals also impaired Salmonella
-killing ability, indicating that there was some sort of inhibitory molecule present in the blood of HIV patients. "Instead of [an immune] deficiency, there was something that was inhibiting the normal killing action of the blood," MacLennan said.
Curiosity piqued, the team did a series of analyses to isolate the inhibitory factor -- what turned out to be an antibody specific to lipopolysaccharide (LPS), a membrane protein common to most gram negative bacteria, including Salmonella
. But when these antibodies successfully bound to Salmonella
, they did not kill it. Furthermore, they prevented the effective antibodies -- those targeted towards Salmonella
outer membrane proteins -- from reaching their targets and clearing the infection.
"Some antibodies were protective, and others were on the contrary facilitating infection," said immunologist linkurl:Jean-Laurent Casanova;http://www.rockefeller.edu/research/faculty/abstract.php?id=323 of The Rockefeller University. It's a "rather unusual finding," Moir added.
Whether or not this mechanism may be the cause of other opportunistic infections in HIV-infected individuals -- particularly other gram negative bacteria with the same LPS molecule on their surface -- is up for debate, Moir said. "We don't know whether this is a very unique situation where all the stars align," she said, or if it could be more generally applicable in terms of immunity against other infections. MacLennan and his colleagues are currently investigating this question.
Both healthy and HIV-infected individuals had both inhibitory and protective antibodies. The difference seemed to stem from the ratios of the two, with HIV patients having much higher levels of the inhibitory kind relative to the protective ones. Why this is the case, however, is a bit unclear.
One possibility is that, in HIV-infected individuals, gut bacteria leak into the plasma, increasing the amount of circulating LPS, which is common to many different bacteria, and causing the body to generate more antibodies specific to that bacterial protein. "The immune response in HIV [patients] does seem to alter the permeability of the gut wall and the ease with which bacteria can come across the gut and get into the blood," MacLennan said.
The results may have important implications for the development of a vaccine against Salmonella
, Moir said. "If a vaccine can be developed, this study helps in directing the research towards an antibody that's directed towards the outer membrane proteins, not LPS," she said. "Stay away from LPS [or] you'll be making inhibitory proteins."
Furthermore, "it may have general implications beyond Salmonellosis and HIV as well," Casanova added. "Beyond HIV-infected individuals, it suggests that if you alter the antibody response, that may precipitate Salmonellosis and perhaps other infectious diseases."
C.A. MacLennan, et al., "Dysregulated Humoral Immunity to Nontyphoidal Salmonella in HIV-Infected African Adults," Science, 328:508-12, 2010.
**__Related stories:__***linkurl:Q&A: The future of HIV vaccines;http://www.the-scientist.com/blog/display/56189/
[30th November 2009]*linkurl:HIV antibody duds explained;http://www.the-scientist.com/blog/display/56156/
[19th November 2009]*linkurl:Salmonella vaccine lift-off;http://www.the-scientist.com/blog/display/54721/
[2nd June 2008]