Licence to kill

When the body's own cells become infected, the immune system can take the extreme step of killing them to preserve the rest of the organism. It is no surprise that this potentially dangerous mechanism has to be kept under strict control, and on pages 4741, 4782 and 4803 of this issue, three groups suggest a new mechanism by which this control operates.

The immune system's T-killer cells exist as inactive precursors that must be activated to develop killing capacity. Activation entails recognition of antigenic peptides bound to class I molecules of the major histocompatibility complex (MHC) on the surface of antigen-presenting cells (APCs). But this precursor-APC interaction is often not enough to stimulate the T-killer cell (Fig. 1a, overleaf) and, in these cases, a T-helper cell that can recognize an antigen on the same APC is also required4.

Figure 1: Antigen-presenting cells (APCs) need a licence to help T-killer cells.

a, T-killer precursors recognize antigen on resting APCs, but do not receive enough signal to be activated. b, According to the traditional model, T helpers and T killers recognize antigen on the same APC. The helper is activated and produces interleukin-2 which contributes to activation of the killer. c, In the new model, supported by the experiments of Ridge et al.1, Bennett et al.2 and Schoenberger et al.3, the APCs are licensed to activate T-killer cells by T helpers or by other stimuli.

The need for helpers and the nature of this help have intrigued immunologists for years. The first model (Fig. 1b) suggested that T-helper and T-killer cells recognize their specific antigens simultaneously on the same APC, and that cytokines (such as interleukin-2) produced by the activated T-helper cell would act on the T killer and facilitate its response4,5. Although interleukin-2 can substitute for helper cells in some systems, there are two main problems with this model. First, it requires that two rare antigen-specific cells meet on the same antigen-bearing APC — an event that has an extremely low probability. Second, and more importantly, some killer responses can be elicited in the absence of helper cells6, implying that the need for help is conditional rather than absolute.

The alternative theory, proposed by Polly Matzinger on theoretical grounds, is a ‘licensing’ model. Matzinger suggested that by recognizing antigen on APCs, T-helper cells deliver a signal that activates the APCs7. The licensed APC can then directly stimulate T-killer cells (Fig. 1c). The key to this model is a specialized ‘professional’ APC, which has been identified as the dendritic cell. Dendritic cells sample antigens in all body tissues and migrate to lymph nodes where they present antigens to T cells8. The molecules responsible for the interaction between the T-helper and dendritic cells are called CD40L and CD40. CD40L is a membrane molecule expressed by antigen-stimulated T-helper cells. It interacts with and triggers CD40, a surface receptor that can activate dendritic cells9,10,11.

The three new papers support this model of T-killer-cell activation. Bennett et al.2 and Schoenberger et al.3 show that mice lacking T-helper cells cannot mount a killer response when they are injected with APCs that display an antigen recognized by T-killer cells. Remarkably, however, the authors could induce a killer response by injecting the mice with an antibody against CD40 — by binding CD40 and triggering the APCs, the anti-CD40 antibody acts as a surrogate of T-helper cells. Moreover, Schoenberger et al. noticed that, in normal mice, they could block the induction of a helper-dependent T-killer response using antibodies to CD40L. This inhibition was reversed by adding the stimulatory anti-CD40 antibody. Bennett and colleagues also showed that mice lacking either CD40 or CD40L cannot mount helper-dependent T-killer responses.

To define the stimuli that enable the APC to activate T-killer cells, Ridge et al.1 manipulated dendritic cells in vitro. They show that dendritic cells carrying specific antigen cannot stimulate T-killer responses in vitro or in vivo unless they are first stimulated by anti-CD40 antibodies or specific T-helper cells. The interaction between T-helper cells and APCs can be dissociated in time from the interaction between APCs and T killers. So, the rare antigen-specific T cells do not need to interact with APCs simultaneously — they can do it sequentially.

These results1,2,3 clearly illustrate that the interaction between CD40L and CD40 can be crucial in bringing the APCs to a state in which they can autonomously trigger T-killer responses. Interestingly, however, they also suggest that there may be alternative ways to activate APCs. Bennett et al. show that a T-killer response to the same antigen may or may not depend on the T-helper cells, according to whether the antigen is administered on cells or in association with an adjuvant (a substance that induces inflammation). In addition, Ridge et al. show that dendritic cells can also be activated after infection with influenza virus.

What happens to dendritic cells during this activation, and why do different pathways lead to the same effect? It is well known that dendritic cells need to be activated to perform optimally. In their resting state, they possess low levels of MHC and co-stimulatory molecules, so they are poor at stimulating T-killer cells. Resting dendritic cells can be activated by inflammatory cytokines and bacterial products that upregulate MHC and the co-stimulatory molecules, thereby increasing the ability to stimulate T cells9. Furthermore, activation of dendritic cells by CD40 results in increased production of cytokines, especially interleukin-12 (refs 10,11). The latest findings1,2,3 can thus be explained by the fact that co-stimulatory molecules and interleukin-12, which are well-known requirements for activation of T-killer cells, need to be induced in dendritic cells by T-helper cells and inflammatory stimuli.

The licensing model reconciles, in a simple concept, many apparently contrasting observations. For instance, it explains how the response to the same antigen can be T-helper dependent or independent. In the presence of adjuvant (which Charlie Janeway calls “the little dirty secret of immunologists”12), the inflammatory cytokines generated may be enough to activate dendritic cells to a state where they can autonomously stimulate a T-killer response, even in the absence of T-helper cells. On the other hand, if antigen is presented by resting dendritic cells, the T-helper-mediated CD40 triggering would be absolutely required. But this model is still compatible with the possibility that particularly strong antigens might be able to stimulate T-killer responses, even in the absence of help or inflammation.

The licensing concept has further implications for the generation of polarized immune responses. Activation of the APCs and production of interleukin-12 are required for the T-helper responses that protect from intracellular pathogens. Conversely, suppressive T cells might inhibit or modulate the function of the dendritic cells, and in this way interfere with the immune response. Thus, rather than being chaperones that bring T helpers and T killers together, the dendritic cells have now become active protagonists that are licensed to activate killer cells. Stimuli from pathogens, inflammatory cytokines and T-helper cells are integrated in dendritic cells, which ultimately tune the capacity to stimulate T-cell responses. It makes sense to have different ways to activate dendritic cells so that an appropriate response to any offending agent can be mounted effectively.

The possibility of manipulating the activation of dendritic cells will open new avenues for vaccination and therapy. By stimulating dendritic cells, the response to weak antigens such as tumour-specific antigens (which are normally encountered outside an inflammatory context) may be boosted. This could be done using anti-CD40 antibodies, as proposed by Ridge et al.1, Bennett et al.2 and Schoenberger et al.3. Alternatively, as others have suggested11, dendritic cells could be pulsed with an antigen that is recognized by the T-helper memory cells, thereby causing them to be activated in a timely fashion in lymphoid organs.


  1. 1

    Ridge, J. P., Di Rosa, F. & Matzinger, P. Nature 393, 474–478 (1998).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Bennett, S. R. M. et al. Nature 393, 478–480 (1998).

    ADS  CAS  Article  Google Scholar 

  3. 3

    Schoenberger, S. P., Toes, R. E. M., van der Voort, E. I. H., Offringa, R. & Melief, C. J. M. Nature 393, 480–483 (1998).

    ADS  CAS  Article  Google Scholar 

  4. 4

    Keene, J. A. & Forman, J. J. Exp. Med. 155, 768–782 (1982).

    Google Scholar 

  5. 5

    Mitchison, N. A. & O'Malley, C. Eur. J. Immunol. 17, 1579–1583 (1987).

    Google Scholar 

  6. 6

    Buller, R. M., Holmes, K. L., Hugin, A., Frederickson, T. N. & Morse, H. C. Nature 328, 77–79 (1987).

    ADS  CAS  Article  Google Scholar 

  7. 7

    Guerder, S. & Matzinger, P. J. J. Exp. Med. 176, 553–564 (1992).

    Google Scholar 

  8. 8

    Banchereau, J. & Steinman, R. M. Nature 392, 245–252 (1998).

    ADS  CAS  Article  Google Scholar 

  9. 9

    Sallusto, F. & Lanzavecchia, A. J. Exp. Med. 179, 1109–1118 (1994).

    Google Scholar 

  10. 10

    Caux, C. et al. J. Exp. Med. 180, 1263–1272 (1994).

    Google Scholar 

  11. 11

    Cella, M. et al. J. Exp. Med. 184, 747–752 (1996).

    Google Scholar 

  12. 12

    Janeway, C. A. J Cold Spring Harb. Symp. Quant. Biol. 54, 1–13 (1989).

    Google Scholar 

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Lanzavecchia, A. Licence to kill. Nature 393, 413–414 (1998).

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.