Wednesday, October 14, 2015

Neither apoptosis nor necrosis but necroptosis allows cross-priming of CD8 T cells

Some pathogens, especially viruses, infect only epithelial cells but not dendritic cells (DCs). This creates dilemma for immune system since activation of naive CD8 T cells require dendritic cells. So how can DCs present cytoplasmic [viral] antigens to CD8 T cells without having one?

This process is called cross-presentation. Since discovery of programmed cell death called apoptosis or silent cell death, it was proposed that apoptosis-derived antigens were cross-presented for tolerance while necrosis-derived antigens [from virus infected cells] were cross-presented for priming. However, data were not consistent and frequently contradictory results were produced. 

Now new paper in journal Science provided support for the alternative model for cross-presentation that could reconcile and explain earlier observations. It found that it is 3rd pathway, called programmed necrosis or necroptosis, rather than apoptosis or necrosis, that provides cargo proteins for cross-presentation by DCs.

To study effect of apoptosis or necroptosis, the authors transduced NIH-3T3 cell line with fusion constructs containing caspase 8 (involved in apoptosis), or RIPK3 (involved in ripoptosome-mediated necroptosis) or RIPK3ΔC (inducing cell death without ripoptosome). When exposed to dimerization reagent, these transduced cells underwent cell death according to the predicted pathway.

Next, the authors showed that dimerization of RIPK3 (referred here as acR3) or RIPK3ΔC (referred here as ac3ΔC) constructs, but not caspase 8, could induce release of damage-associate molecular patterns (DAMPs) such as HMGB1 or ATP.

More importantly, however, when these transduced cell lines [also expressing OVA antigen] were injected into mice to induce CD8 T cell response, the authors found that only wild-type RIPK3 construct-induced cell death promoted CD8 T cell expansion and effector differentiation.

In addition, the authors found that neither secondary necrosis nor mechanical [freeze-thaw] necrosis could prime CD8 T cells.

Next, the authors observed that cells undergoing RIPK3-ripoptsome mediated necroptosis selectively release IL-6 and show rapid RIPK1-mediated IκB degradation.

Finally, using CRISPR/cas9 modified CT26 tumor cell line, the authors showed that RIPK3-mediated ripoptosome assembly involving RIPK1-NF-κB pathway was crucial for immunogenic necroptosis in tumor challenge model.

In summary, this study further refined our understanding of immunogenic cell death and further defined molecular components essential to achieve it. Of note, the role of necroptosis in immunogenic cell death may finally harmonize prior data regarding necrosis or apoptosis in cross-presentation and improve our therapeutic tool box.

David Usharauli


  1. Complete misinterpretation of the paper / these results have nothing to do with immunogenic cell death as the most important marker, surface calreticulin is not properly induced and in vivo immunogenicity is less than 70 percent. Also, tthe system is completely artificial in terms of cell death induction and thus does not further our undestanding of cell death in nature, at all! Half of the graphs have no error bars and lack proper controls. Perhaps the most low quality Science paper I have seen in recent times!

  2. First, let me thank you for expressing your opinion.

    Regarding paper: I am not a big fan of Matthew Albert either (😬) but neither Zitvogel/Kroemer team deserves any special credence. Calreticulin can't possibly be the only marker for immunogenicity. I agree that system is artificial but they have used CRISPR/Cas9 system (😎) and today its all you need to get your paper in top journals (😜)


  3. I agree - Calreticulin doesn't explain everything and there is no biological reason for why immunogenicity would solely depend on one molecule. The problem I have in this paper is that - first they try to detect these markers but then without even blocking them they conclude DAMPs have no role to play; big or small journal, doesn't matter but you cannot conclude a cause-effect relationship without proper intervention experiments. I believe they have simply sensationalized the results - but the paper lacks substance. It is unfortunate that the paper did not go through proper peer-review in my opinion - better papers than this one get published in the more specialized journals like CDD, EMBO J and PNAS. Not using a single natural cell death inducer has to be considered a flaw because cell death is a very pleiotropic signalling paradigms whereas in this system cell death is rather "clean" - no natural example can parallel this - so very tough to say these results have physiological relevance (often a pre-requisite for many top journals which is neglected here). It is just an unfortunate example in my opinion - that is all.

  4. I agree that blocking ATP/HMGB1 would have provided more definite results of the role of these DAMPs in immunogenicity. It seems that the authors dismissed DAMPs as a a major contributors of immunogencity when results showed that both acR3 and RIPK3ΔC cell death could release ATP/HMGB1 but only acR3 induce CD8 T cell cross-priming.

  5. Thanks for the discussion David - enjoyed it!