Wrong aluminium salts used in experiments to explain adjuvant effect, scientist claims
Two recent studies examining how aluminium adjuvants boost the effectiveness of vaccines may be flawed because they did not use either of the two compounds commercially licensed for the purpose, Chemistry World has learned.
Aluminium hydroxide (in the form of crystalline aluminium oxyhydroxide , AlOOH) and aluminium phosphate (amorphous aluminium hydroxyphosphate, Al(OH)x(PO4)y) are both licensed for use as adjuvants. These aluminium salts - generically dubbed ’alum’ by doctors - have been used for decades to stimulate the immune response to the antigen in vaccines. But precisely how they help the body to produce more antibodies has remained a mystery.
Two teams of researchers recently claimed to have found the answer. A study led by Richard Flavell of the Yale University School of Medicine, and published online by Nature in May, suggested that alum activates the Nalp3 inflammasome, a receptor that prompts white blood cells to produce antibodies.  Two months earlier, work by the group of Bart Lambrecht at the Erasmus University Medical Centre in Rotterdam, Netherlands, together with colleagues at the University of Ghent, had shown that alum triggered the release of uric acid in mice. Uric acid is also known to boost the production of antibodies.
However, both groups’ experiments used Imject Alum, sold by Pierce Biotechnology of Rockford, Illinois, a division of Thermo Fisher. Imject is a roughly equal mixture of amorphous aluminium hydroxycarbonate and crystalline magnesium hydroxide which is known to be an effective adjuvant in mice, but is not used in vaccines for humans.
According to Chris Exley, who has studied the biological effects of aluminium for over 20 years, the mix of compounds in Imject more closely resembles the antacid Maalox than commercial adjuvants.
’This significant mistake in both of these high profile publications has serious implications for the interpretation of the results,’ says Exley, a reader in bioinorganic chemistry at Keele University, UK. ’While there is no dispute that Imject Alum has "adjuvant-like" properties, it will not behave like commercially available aluminium adjuvants and as such should not really be used to explain how adjuvants work.’
He points out that Imject, unlike licensed adjuvants, contains magnesium. Secondly, the aluminium in Imject is likely to be more soluble than in commercial adjuvants - in Maalox, these Al3+ ions help to buffer pH changes, contributing to its antacid properties. Both the presence of Mg2+ and the higher concentration of Al3+ in Imject may have affected the results of the two studies, Exley suggests.
Adjuvant or antacid?
Both groups have responded to the criticism by agreeing that they will repeat their experiments with a commercially licensed form of alum. ’We should certainly also test a commercially licensed form of alum, and are planning those experiments,’ says Lambrecht.
However, neither group thinks that the magnesium ions are producing the effects they observed because they tested aluminium hydroxide containing no magnesium, and got similar results.
’Imject is the most widely used form of alum adjuvant in the mouse,’ explains Lambrecht. ’We have the same results using aluminium hydroxide, but published the whole story with Imject.’
Flavell also rejects the idea that the higher solubility of aluminium in Imject is playing a role. ’We actually did experiments to assess the role of soluble aluminium and found no stimulation,’ he says. ’We will however look at the licensed adjuvant for completeness.’
Stephanie Eisenbarth, the postdoctoral fellow in Flavell’s lab who was first author on the Nature paper, says that the chemistry of alum may not be critical to its vaccine boosting properties anyway. ’There’s a whole class of compounds, including uric acid crystals, asbestos and silica, that activate an immune response through Nalp3 and may also work as adjuvants. It could be the physical nature of how these crystals interact with cells that is key to stimulating the immune response.’
Interesting? Spread the word using the ’tools’ menu on the left
1 S C Eisenbarth Et al, Nature, 2008, DOI: 10.1038/nature06939
2 M Kool et al, J. Exp. Med., 2008, DOI: 10.1084/jem.20071087