T-Cell antagonism

[Ganseerpel]

<TABLE cellSpacing=0 cellPadding=0 width="100%" border=0><TBODY><TR><TD>Nature 394, 421 - 422 (1998) ? Macmillan Publishers Ltd. </TD></TR><TR><TD></TD></TR></TBODY></TABLE><!-- header ends here --><!-- text starts here -->Immunology: The original sin of killer T cells
ANDREW J. MCMICHAEL


The phrase 'original antigenic sin' was first used to describe the antibody response to influenza virus. After an initial infection, reinfection (or vaccination) with a new strain of the virus boosted the concentration of antibodies specific for the earlier infecting strain<SUP>1,2</SUP>. Although these antibodies cross-reacted with the new virus, they had higher affinity for the original infecting strain. This was immediately seen to have big implications for vaccine design -- a vaccine based on a new strain of influenza virus might be unable to prime antibodies to the intended virus in people who had already been infected with a related viral strain.
Until now, original antigenic sin has been regarded as largely an antibody phenomenon. But, on<!-- 67075 --> page 482 of this issue, Klenerman and Zinkernagel<SUP>3</SUP> describe original sin in the response of cytotoxic T lymphocytes (CTLs) to lymphocytic choriomeningitis virus (LCMV). Their work was stimulated by studies of people with the human immunodeficiency virus (HIV), who sometimes mount a CTL response to an immunodominant strain of the virus with weak or no response to the other immunogenic variants present<SUP>4,5</SUP>.
The authors attacked the problem in vivo. They infected mice with strains of LCMV that were either normal (wild type) or mutated at the immunodominant epitopes recognized by the CTL. For each mutated strain studied, they found that the CTL response was asymmetrical. Mice infected with the wild-type virus showed only a weak cross-reactive specificity when challenged with the mutant strain, reacting mainly to the wild-type virus. The low reactivity of CTLs against the mutant strain was also associated with delayed clearance of this strain by the immune system. In contrast, the CTL responses from mice infected with the mutant virus cross-reacted equally with the mutant and wild-type strains. Thus, the order in which the mice are exposed to different variants of the virus could have a significant effect on the outcome of the infection.
How does this happen? A possible explanation<SUP>6</SUP> for antibody original antigenic sin emerged when helper T cells were discovered, and it was realized that T cells react with peptide fragments of viral proteins that are often conserved between different strains. So, helper T cells, primed by the original virus and then stimulated by the new infection, might activate memory B cells that are specific for the original virus (although it is not clear why antibodies specific for the new variant are not stimulated too). In the same way, the phenomenon observed by Klenerman and Zinkernagel<SUP>3</SUP> might reflect the very strong CTL memory response to LCMV infection. Studies<SUP>7-9</SUP> of acute LCMV infection in mice showed that massive CTL responses occurred, whereby 25-50% of all CD8-positive T cells were virus specific. Even after recovery, 10% of peripheral CD8-positive T cells were specific for the immunodominant LCMV epitopes. A weakly cross-reacting new virus might, therefore, reactivate these plentiful CTL memory cells more readily than the much less abundant naive CTL precursors (<!--a1-->Fig. 1). Although this mechanism depends on very high levels of memory CTLs, similar numbers are seen in persistent HIV infection<SUP>10</SUP>, where original antigenic sin may also occur<SUP>4,5</SUP>.
<TABLE cellPadding=10 border=0><TBODY><TR vAlign=top><TD><!--/usr/users/nature/Nature/public_html/images/v394n6692/394421aa.eps.0.gif:GIF89a:280:174--></TD><TD>Figure 1 Original antigenic sin in cytotoxic T lymphocytes (CTLs). Full legend

High resolution image and legend (66k)


</TD></TR></TBODY></TABLE>A second possibility is that activated memory CTLs remove enough of the antigen-presenting cells (APCs) to abort the primary CTL response to the new viral epitope. This would be consistent with the delayed clearance of mutant compared with wild-type virus in secondary challenges. The memory CTL could simply kill the APC, although recent results<SUP>11-13</SUP> indicate another possibility. A type of APC called the dendritic cell must be activated in order to initiate CTL responses. Activation occurs through an interaction between the CD40 ligand on T helper cells and CD40 on the dendritic cells. But if this requirement for activated dendritic cells is less stringent for the memory CTLs, these CTLs could deactivate the dendritic cells, thereby inhibiting new primary CTL responses. Deactivation could occur through the interferon- and other cytokines that are released from memory CTLs within a few hours of contact with antigen<SUP>14</SUP>.
Another explanation would be that a form of T-cell-receptor antagonism occurs<SUP>15</SUP>. If the CTL clones that react with the mutant viral epitope have receptors that bind suboptimally to the wild-type epitope, the result might be partial activation, leading to anergy (a failure of the immune system to respond). Thus, the wild-type virus would antagonize the primary CTL response to the mutant strain. For this mechanism to work, antigen from a previous infection would need to persist. Low levels of integrated DNA persist in mice infected with LCMV<SUP>16</SUP>, although these levels are probably too low for antagonism to occur. Moreover, in Klenerman and Zinkernagel's study, the CTLs stimulated by the mutant virus cross-reacted well with the wild-type virus, arguing against this explanation.
Original antigenic sin in CTL responses could be very important -- it could explain some of the complexity in CTL responses to HIV and its variants<SUP>17,18</SUP>. If it were a common phenomenon, it would give viruses another means of escape from the immune system. That is, once a high-level CTL response has become fixed, immunogenic mutants might escape and become dominant in the swarm of viral quasispecies, failing to provoke an effective CTL response.
Just as in the initial description of original antigenic sin, the new work<SUP>3</SUP> has implications for vaccine design. Many groups are now trying to design CTL-inducing vaccines for the control of variable viruses such as HIV and hepatitis C virus. Original antigenic sin means that a monovalent vaccine (such as a peptide) intended to stimulate CTLs may not work if the virus varies at that epitope. Not only would the vaccine-induced CTL response fail to control infection with any variant virus, but the CTL response to the equivalent epitope in that virus might also be impaired. The vaccine might even make the infection worse. The probable solution is to choose several well-conserved epitopes for the vaccine -- but HIV is so variable, these may be in short supply.


<DD>
Andrew J. McMichael is at the Institute of Molecular Medicine, John Radcliffe Hospital, Oxford OX3 9DU, UK.
e-mail: amcmichael@hammer.imm.ox.ac.uk




References

1. Francis, T. Ann. Int. Med. 39, 203-221 (1953). Links
2. Fazekas de St Groth, S. & Webster, R. G. J. Exp. Med. 124, 331-346 (1966). Links
3. Klenerman, P. & Zinkernagel, R. M. Nature 394, 482-485 (1998). Links
4. McAdam, S. N. et al. J. Immunol. 155, 2729-2736 (1995). Links
5. Klenerman, P., Meier, U. C., Phillips, R. E. & McMichael, A. J. Eur. J. Immunol. 25, 1927-1931 (1995). Links
6. Askinas, B. A., McMichael, A. J. & Webster, R. G. in Basic and Applied Influenza Research (ed. Beare, A. S.) 157-188 (CRC Press, Boca Raton, FL, 1982). Links
7. Butz, E. & Bevan, M. Immunity 8, 167-175 (1998). Links
8. Gallimore, A. et al. J. Exp. Med. 187, 1383-1393 (1998). Links
9. Murali-Krishna, K. et al. Immunity 8, 177-187 (1998). Links
10. Ogg, G. S. et al. Science 279, 2103-2106 (1998). Links
11. Bennett, S. R. M. et al. Nature 393, 478-480 (1998). Links
12. Ridge, J. P., DiRosa, F. & Matzinger, P. Nature 393, 474-478 (1998). Links
13. Schoenberger, S. P., Toes, R. E. M., van der Voort, E. I. H. & Melief, C. J. M. Nature 393, 480-483 (1998). Links
14. Lalvani, A. et al. J. Exp. Med. 186, 859-865 (1997). Links
15. Meier, U.-C. et al. Science 270, 1360-1362 (1995). Links
16. Klenerman, P., Hengartner, H. & Zinkernagel, R. M. Nature 390, 298-301 (1997). Links
17. Haas, G. et al. J. Immunol. 157, 4212-4221 (1996). Links
18. Harrer, T. et al. J. Immunol. 156, 2616-2623 (1996). Links

</DD>

[gsgs]

Re: T-Cell antagonism

this is from 1998, I assume you have already a follow up in the pipeline...

Well, what does it mean for panflu ?

When we take the pandemic vaccine, then we migh only need one dose and we could boost it even with normal H3N2 vaccine ,
or with any other H5N1-vaccine, which could be produced right now ?

Or, we only need to stockpile one dose of prepandemic vaccine,
we could then boost it with normal, easier producable vaccine ?

Maybe it only works with real infection or with life-virus vaccine ?

[Ganseerpel]

Re: T-Cell antagonism

Well, what does it mean for panflu ?

Antigenetic sin (AS) and T-cell antagonism (TCA) are well established issues and object of investigation. The findings are likely to be of practical interest and should be considered in vaccine management, although many aspects remain open to question.

Though Maybe Lambert is right when he states:

Although such 'deviating' immunological effects are likely to occur, their impact on protection is far from established. The effect of annual influenza revaccination on mortality was recently investigated in a population-based cohort study in the Netherlands that included community-dwelling individuals aged 65 years or older<SUP>50</SUP>. Overall, during a 5-year period (1996−2002), annual revaccination was associated with a significantly reduced mortality risk of 24%, whereas the annual reduction of mortality risk was only 10% in first-time vaccinees. Thus original antigenic sin does not seem to be a major practical obstacle in the race against antigenic changes that characterizes anti-influenza vaccination and it may not require specific strategies for overcoming resulting limitations of immune responses in vaccinees.

When we take the pandemic vaccine, then we migh only need one dose and we could boost it even with normal H3N2 vaccine ,
or with any other H5N1-vaccine, which could be produced right now ?

Or, we only need to stockpile one dose of prepandemic vaccine,
we could then boost it with normal, easier producable vaccine ?

We have no evidence that this would work. One simple conclusion from these findings is that it is clearly better to give several antigens at once than to give them one after another.

We should take into account that consecutive vaccination as it is standard in common flu might have adverse effects at least in elder age groups baecause in these groups we have a higher probability of existing cross preimmunity causing an impact on the proper immune response.

It is noteworthy that the phenomenon was first observed in Influenza although present research is more focused on HIV.

Maybe it only works with real infection or with life-virus vaccine ?

AS is not linked to life virus infection. The possible feature of TCell mediated cross immunity as discussed in the other thread is mainly based on life virus infection/vaccination.

[gsgs]

Re: T-Cell antagonism

I calculate that these 26000 Dutch elderly people had an
average chance of 2.7% per year to die when unvaccinated.
Single vaccination reduced this to 2.4% per year and
repeated vaccination to 2.0% per year.

(provided they don't emigrate to Italy ;-) )

[Ganseerpel]

Re: T-Cell antagonism

Just one (inconvential) idea concerning the 1918' flu:

Could the first wave of spring 1918 have induced - at least in part - an antigenic sin to individuals who succumbed to second infection a few months later?

Taubenbergers's investigation with his first wave isolates could let us know more.

[gsgs]

Re: T-Cell antagonism

but those who had caught the first wave were reported to
have been well protected by immunity in the 2nd wave.

As I understand, in the 2nd wave, the sin would be responsable
for production of antibodies against the first wave
and not so much against the 2nd.

Also, most people died from cytokene storm(CS), an overproduction of antibodies.

Or do you mean, AS was the cause for fewer CS ?
Well, I think infection without CS was also reduced in those who contacted
the first wave.

Correct me, if wrong.

[Ganseerpel]

Re: T-Cell antagonism

but those who had caught the first wave were reported to
have been well protected by immunity in the 2nd wave.

Let me first underline that I don't think antigenic sin to be a major agens in the 1918 flu but it could have had some influence.

Also, most people died from cytokene storm(CS), an overproduction of antibodies.

Or do you mean, AS was the cause for fewer CS ?
Well, I think infection without CS was also reduced in those who contacted
the first wave.

As we don't know the exact background in which cytokine related dammage occured we can only hypothesize. A strong response to the wave I strain(s) boosted by a second infection with the fall strain (w II) would be consistent with an enhanced cytokine effect.

Under the assumption the second strain was poorly matched to w I, AS would have boosted a strong immediate response with or without enhancement of cytokine pathways but - due to poor matching - with delayed clearance and prolonged disease.

The "overshooting" cytokine response does not rule out AS effects but - on the contrary- could rather account for the phenomenon.

[gsgs]

Re: T-Cell antagonism

but, protection in WII of those who had caught WI is well documented.
(I read that in Berry's book)

By your theory, those who caught WI should have been _more_ effected.

[Ganseerpel]

Re: T-Cell antagonism

By your theory, those who caught WI should have been _more_ effected.

It's not "my theory". What I am trying to say is just that AS could explain a bit of the pathology of W II and have had an effect in a limited number of cases.

Also, most people died from cytokine storm(CS), an overproduction of antibodies.

CS is not an overproduction of Abs but an overreaction with enhanced occurence of signaling compounds causing failure of multiple organ systems. Cytokines are intercell-signaling molecules binding to a specific receptor like INF, Interleukines etc.