Vitamins prime immunity

The finding that derivatives of vitamin B can bind to an antigen-presenting protein that stimulates specialized immune cells suggests a novel mechanism by which the immune system detects microbial infections. See Article p.717

In addition to their vital functions in development and metabolism, many vitamins have crucial roles in the immune system. The functions of two lipid-soluble vitamins, vitamin D (ref. 1) and vitamin A (ref. 2), in modulating immune responses are already known. But on page 717 of this issue, Kjer-Nielsen et al.3 suggest a very different immune function for vitamins B2 (riboflavin) and B9 (folic acid), which are water-soluble vitamins. The authors provide evidence that molecules produced when bacteria metabolize certain B vitamins can activate a class of immune T cell called mucosa-associated invariant T (MAIT) cells. The proposal that MAIT cells detect infected cells through vitamin metabolites attached to the cells' surface is the first suggestion that vitamins can act as antigens (substances that activate T and B cells of the immune system), and boosts our understanding of this novel arm of the immune system.

T cells are major players in immunity, providing protection against infection. The most common ones are CD4+ and CD8+ T cells, which are found throughout the body and carry a broad repertoire of antigen receptors on their surface. CD4+ and CD8+ T-cell receptors bind to peptide antigens (fragments of proteins) that are 'displayed' on the surface of other cells by a cell-membrane protein belonging to the major histocompatibility complex (MHC) family. Development of these conventional CD4+ and CD8+ T cells depends on the presence of MHC molecules.

By contrast, MAIT cells are a type of unconventional T cell that are mostly found in the intestine, liver and lung4 and that have only limited antigen-receptor diversity. MAIT-cell development depends on an MHC-related protein called MR1, which has been highly conserved over the course of mammalian evolution. Because the amino-acid sequence of MR1 is very similar to that of MHC molecules, it has been speculated that MR1 binds to specific antigens that lead to MAIT-cell activation. Indeed, genetic and biochemical studies suggest that MR1 presents antigen for MAIT-cell activation, but the chemical nature of the antigen was unknown5.

Intriguingly, and uniquely among known T-cell populations, MAIT-cell survival also depends on the commensal microbiota — the non-pathogenic microorganisms that live on and in the body. Furthermore, MAIT cells are activated by the presence of cells infected with a diverse range of bacteria and yeast strains (although not viruses)6,7. Together, these findings hinted that MR1 is likely to bind microbial antigens, which are then presented to MAIT cells.

Kjer-Nielsen et al. have taken a key step in identifying exactly which antigens MR1 presents by defining the crystal structure of MR1 bound by a metabolite of folic acid called 6-formyl pterin (6-FP). Their study stemmed from the serendipitous observation that using media containing folic acid enhanced the folding of denatured MR1 protein.

The authors' crystal structure shows that the MR1 antigen-binding groove specifically accommodates pterin rings, which are scaffold structures contained in some B vitamins and their metabolites. They also found that, although the 6-FP–MR1 complex does not activate MAIT cells in vitro, related riboflavin derivatives, when bound to MR1, do. This is the first demonstration that MR1 binds vitamin-B metabolites and that some of these metabolites can activate MAIT cells, and it therefore defines a new model of antigen presentation to immune cells. It was already known that MHC molecules present peptides to CD4+ and CD8+ T cells, and that another MHC-like protein called CD1d presents lipid molecules to a class of T cell called NKT cells; now we have evidence that MR1 presents vitamin-B metabolites to MAIT cells (Fig. 1).

Figure 1: Modes of antigen presentation.

a, Conventional CD4+ and CD8+ T cells bind to antigens presented by MHC molecules on the surface of other cells. These antigens are typically short chains of amino acids (peptides) derived from proteins. The small balls extending from the peptide are amino-acid side chains that anchor the peptide in the MHC or are detected by the T-cell receptor. An example of a peptide derived from influenza virus that stimulates CD8+ T cells is depicted below the cells. b, Another class of T cell, called NKT cells, recognizes antigens derived from lipid molecules that are presented by cells expressing a molecule called CD1d, which has deep grooves that can accommodate the lipid chains of the antigen. An example of a glycolipid that stimulates NKT cells is depicted. c, Kjer-Nielsen and colleagues3 demonstrate that a third type of antigen-presenting molecule, called MR1, presents metabolites of B vitamins to T cells called MAIT cells. The authors present a crystal structure of the MR1 molecule bound to a derivative of vitamin B9, which shows that the antigen-binding groove accommodates the pterin-ring structures characteristic of B vitamins and their metabolites. An example of a vitamin-B2 metabolite that stimulates MAIT cells is shown.

However, exactly how this antigen-presentation process is linked to immunity to microbes is still not fully clear. Kjer-Nielsen and colleagues suggest that a mechanism by which MAIT cells detect and control infection is the display of vitamin-B metabolites on the surface of infected host cells. In support of this proposal, they cite the previous finding that the metabolic pathway that generates the antigenic molecules seems to be present only in microbes previously found to activate MAIT cells in vitro6,7. This correlative observation will need to be tested empirically to determine the importance of vitamin-B-metabolite presentation in controlling infection.

It will also be interesting to identify the cellular location and mechanism by which vitamin-B metabolites bind to MR1 proteins, and the role of infection in this process. In germ-free mice, which do not contain any commensal bacteria and therefore also lack MAIT cells, the addition of certain commensal bacterial strains allows MAIT cells to develop7. But, curiously, not all of these commensal strains have the molecular pathway that makes the metabolites studied by Kjer-Nielsen and colleagues, which implies that other MR1 ligands might be involved in microbial detection by MAIT cells. In addition, some cell-signalling molecules, such as interleukin-12 and interleukin-23, are known to activate MAIT cells8,9, and this might mitigate the importance of activating signals derived from vitamin presentation by MR1.

Another pertinent question is that of the roles that MAIT cells have in the gut. Vitamins help to orchestrate the relationships between mammalian host immunity, commensal gut microbiota and pathogenic microorganisms10. For example, vitamin B9 and its derivatives can serve as coenzymes in essential metabolic pathways11, and this vitamin is also required for the survival of a type of T cell called regulatory T cells12. Kjer-Nielsen and colleagues' findings suggest that interactions between the host and gut microbiota might also be influenced by MR1-dependent presentation of microbial antigens to MAIT cells.

A model that emerges from this idea is that, during early mammalian development, the colonization of the host by commensal bacteria13 provides vitamin metabolites that act as ligands for MR1, allowing MAIT cells to develop in the thymus. These cells then migrate to other organs, in particular the lungs, liver and gut, where they help to guard against bacterial infections. Although further work is required, it is attractive to speculate that MAIT-cell-dependent protection against pathogens could be augmented by dietary provision of vitamins or by pterin-based therapies. Such augmentation might enhance immunity to microbes, or even help to treat immunodeficiencies.


  1. 1

    Chun, R. F., Adams, J. S. & Hewison, M. Expert Rev. Clin. Pharmacol. 4, 583–591 (2011).

  2. 2

    Hall, J. A., Grainger, J. R., Spencer, S. P. & Belkaid, Y. Immunity 35, 13–22 (2011).

  3. 3

    Kjer-Nielsen, L. et al. Nature 491, 717–723 (2012).

  4. 4

    Treiner, E. et al. Nature 422, 164–169 (2003).

  5. 5

    Huang, S. et al. Proc. Natl Acad. Sci. USA 106, 8290–8295 (2009).

  6. 6

    Gold, M. C. et al. PLoS Biol. 8, e1000407 (2010).

  7. 7

    Le Bourhis, L. et al. Nature Immunol. 11, 701–708 (2010).

  8. 8

    Chua, W.-J. et al. Infect. Immun. 80, 3256–3267 (2012).

  9. 9

    Chiba, A. Arthritis Rheum. 64, 153–161 (2012).

  10. 10

    Nicholson, J. K. et al. Science 336, 1262–1267 (2012).

  11. 11

    Said, H. M. Biochem. J. 437, 357–372 (2011).

  12. 12

    Kunisawa, J., Hashimoto, E., Ishikawa, I. & Kiyono, H. PLoS ONE 7, e32094 (2012).

  13. 13

    Koenig, J. E. et al. Proc. Natl Acad. Sci. USA 108 (suppl. 1), 4578–4585 (2011).

Download references

Author information

Correspondence to Wei-Jen Chua or Ted H. Hansen.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Chua, W., Hansen, T. Vitamins prime immunity. Nature 491, 680–681 (2012).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.