IL-23 --> TH17 +++ IL-23 --> TH1

By now it has become very well established that naive T-cells can and do differentiate into many many many different terminal states, although most of them will die along the way. A naive T-cell arriving in the thymus expressed both CD4 and CD8, and has to choose based on intercellular interactions with APCs (antigen-presenting cells) and the cytokine environment whether to be CD4+ or CD8+ and wind up recognizing antigen on MHCII or MHCI, respectively. CD4+ T-cells can then further differentiate, based again on cytokines, whether to become TH1, TH2, Tregs, or Th17 cells, or some memory variant of the first 3 types. TH1, TH2, and TH17 are all inflammatory phenotypes; Tregs are anti-inflammatory in that they restrain the adaptive immune response from going beserk and killing off the rest of the organism (e.g., septic shock). There're several cans, maybe even a tanker truck, of alphanumeric soup for all the various cytokines, CDs (cellular domains), transcription factors, etc. that regulate T-cell differentiaton. But today we're going to talk mostly (because there'll probably be others that necessarily slip in) about TGFb (transforming growth factor beta), IL-6 (interleukin-6), IL-12, IL-17, IL-23, and its receptor: IL-23R. It should be noted that IL-23R is a homodimer of the IL-12Rp40 (one of 2 subunits of the Il-12R) and a unique IL-23 heterodimerizing domain (p19).

So if proliferating, un-fully-differentiated T-cells are exposed to a mileau of TGFb, IL-6, and IL-21, they generally start to look more like TH17 cells based on their surface molecule expression and cytokine secretion. TH17 cells become TH17 cells when they start to secrete IL-17 (oddly, though, neither TH1 nor TH2 cells are characterized by their secretion of IL-1 [which is a generally pro-inflammatory cytokine] or IL-2 [which is a paracrine {usually} T-cell pro-proliferative signaling cytokine]). But TH17 cells aren't very good at being TH17 cells unless they can detect IL-23.

The group discussed in the paper generated IL-23R knockout mice and tested their TH17-like cells for various parameters. It turns out that without IL-23R (effectively the same thing as having absolutely no IL-23 as far as we know), TH17-like cells don't proliferate very well in the lymph nodes where they'd normally do so. This failure to proliferate as massively as they otherwise would means that there are fewer TH17 cells in circulation and thus fewer are available to infiltrate target inflammed tissue. However, no IL-23 signaling apparently doesn't affect the molecular markers that effect tissue migration, so it seems that IL-23 is primarily involved in driving lymph node expansion of activated TH17 cells. Without IL-23, TH17 cells are also much weaker in that they go apoptotic much more readily. Oddly, the group also reported that in vivo activation without IL-23 didn't affect Foxp3 expression, which I don't understand the relevance of because Foxp3 is primarily a marker for the development of Tregs.

Backing out to a broader picture: it has been found that inflammed tissues have high IL-23 expression. It has also been experimentally determined that injected IL-23 into the skin results in IL-22 mediated dermal inflammation. TH17 cells produce IL-22. So it looks like IL-23 in inflammed tissues may be attracting TH17 cells to come and check it out, then expand if it's necessary and come back to cause some more IL-22-mediated inflammation to try to help get the problem resolved all happy like. But it has also been shown, by the group discussed in the paper, that this local inflammation can happen even without antigen-capable T-cells in a model of dermal delayed-type hypersensitivity.

And, wait!, if that isn't enough: it gets weird, too. In vitro studies have shown that restimulating TH17 cells with IL-23 can cause them to morph into TH1-like cells as characterized by IFNg (interferon gamma) production. Does this mean that TH17 cells themselves aren't a stable phenotype? Are they an evolutionary leftover that can segue into TH1 when necessary? Or are there unique TH17 inflammatory pathways that don't overlap much with the TH1 and TH2 pathways?

Either I don't know enough (very plausible), or we need more experiments!

Note: This paper was kind of odd in that it was a broad summary of current knowledge focusing on the work in 1 specific paper by authors not involved in that paper. Maybe it's a Nature thing. Also, note that I've come to the conclusion that "Not-So-Mad Science Blah Blah Blah" in unnecessary in research blogging. The icon is enough.

Yeonseok Chung, & Chen Dong (2009). Don't leave home without it: the IL-23 visa to TH-17 cells Nature Immunology, 10 (3), 236-238