The cytokine family including interleukin-2 and -15 (IL-2 and IL-15) started out in a primitive vertebrate species with three related cytokines IL-2, IL-15, and IL-15-like (IL-15L), all sharing binding capacity for receptor chain IL-15Rα. IL-15L was conserved in fishes, reptiles, and some mammals but lost in human and mouse. Researchers in Japan, Germany, and the UK, have now determined the function of IL-15L using rainbow trout. Data suggest that in the jawed vertebrate ancestor IL-15 and IL-15L, forming "heterodimer cytokine" complexes with IL-15Rα, selectively induced type 1 (aka Th1) and type 2 (aka Th2) immunity, respectively, whereas free IL-2 efficiently stimulated activated and regulatory T cells by binding their surface IL-15Rα. Insights into their evolution should inspire further studies of human IL-2 and IL-15, both of which are used or tested in cancer immunotherapy.
In human and mouse, the closely related cytokines IL-2 and IL-15 both signal through IL-2Rβ heterodimer receptors, and also bind to a third receptor chain of a different molecular family, cytokine-specific IL-2Rα or IL-15Rα, which has no homologues in receptors for other human or murine cytokines. IL-2 is secreted by activated T cells, and predominantly functions as a free cytokine that binds IL-2Rαβ heterotrimer complexes on target cells. IL-2Rαβ complexes are abundant on activated T cells and, constitutively, on regulatory T cells (Tregs). Therefore, IL-2 both participates in an immuno-stimulatory positive feedback loop through activated T cells and an immuno-inhibitory negative feedback loop through Tregs. IL-15, on the other hand, predominantly uses IL-15Rα as a "heterodimer cytokine" partner at the membrane of the expressing cells or as released free IL-15+IL-15Rα complexes. IL-15 predominantly stimulates "type 1 immunity" by specifically stimulating natural killer (NK) cells and CD8+ cytotoxic T cells, and is important for the maintenance of several lymphocyte populations in intestine, lung, and skin. Recombinant IL-2 is an established anti-cancer drug (e.g. "proleukin"), but in several clinical trials IL-15 (complexed or not with IL-15Rα) is preferred as IL-15 does not have the complicating factor of specifically stimulating Tregs.
Associate Professor Johannes M. Dijkstra and co-workers at Fujita Health University, Japan, together with the groups of Professor Uwe Fischer at the Friedrich-Loeffler-Institute, Germany, and Professor Christopher J. Secombes, University of Aberdeen, UK, have been instrumental in elucidating the evolution of IL-2 and IL-15. In their recent study , for which most experiments were done by the first author Dr. Takuya Yamaguchi at the Friedrich-Loeffler-Institute, they determined that in rainbow trout all the three cytokines IL-2, IL-15, and IL-15L can bind to IL-15Rα. The tandem gene duplication that resulted in IL-15Rα plus IL-2Rα probably occurred in a primitive tetrapod species, and in fish only a single molecule of this receptor chain family is found that is very similar to mammalian IL-15Rα while quite different from mammalian IL-2Rα that acquired substantial modifications during its evolution . Dijkstra and co-workers previously determined that cytokine IL-15L already existed at the evolutionary level of sharks  and is conserved in some mammals like cattle but inactivated in human and mouse . In cattle, they found that IL-15 and IL-15L selectively bind to IL-15Rα, whereas IL-2 selectively binds to IL-2Rα [1,3], which for IL-2 and IL-15 agrees with human and mouse. However, for mammalian IL-15L a function could not be determined so far. The Yamaguchi et al. study , using rainbow trout, now is the first to identify the function of IL-15L. This function is the stimulation of the expression of cytokines IL-4 and IL-13 homologs, as shown both in vitro and in vivo, and thereby of type 2 immunity. Under similar conditions, trout IL-15 was found to selectively stimulate cytokine interferon-g expression, a marker for type 1 immunity, concluding that IL-15 and IL-15L have opposing immune functions; meanwhile, trout IL-2 was found to induce a broader cytokine palette than either IL-15 or IL-15L . All three trout cytokines IL-2, IL-15, and IL-15L, were found to induce activation of transcription factor STAT5 . IL-15 and IL-15L transcripts are abundant and also found in epithelial cells and fibroblasts, and Yamaguchi et al. speculate that the multiple AUG codons in their 5' untranslated regions prohibit translation unless the cell experiences proper signals for inducing type 1 or type 2 immunity . In mammals, such "alarmin" function of IL-15L for inducing type 2 immunity may have been taken over by newer cytokines that seem to be absent in fish, such as IL-25, IL-33, and TSLP. The trout cells most sensitive to IL-15L appear to be innate lymphocytes and they may be equivalent to mammalian ILC2 cells .
Reminiscent of IL-15 and IL-2 in mammals, trout IL-15 is more dependent on co-presentation with IL-15Rα than IL-2 , and also in fish IL-2 appears to participate in an immune-stimulatory positive feedback loop involving activated T cells  and an immuno-inhibitory negative feedback loop through the stimulation of IL-15Rα-expressing Tregs . The dependency of IL-15L on co-presentation with IL-15Rα was found to be even larger than that of IL-15, since for unbound IL-15 but not for unbound IL-15L functional activity could be observed . Retainment of IL-15+IL-15Rα complexes at the surface of expressing cells can contain cytokine activity within confined niches, which may explain why in both mammals and fish IL-15 was selected over IL-2 for stimulation of lymphocytes in mucosal tissues . However, to some extent, as known in mammals for IL-2+IL-2Rα, trout IL-2 can be presented at the cell surface together with IL-15Rα , and this type of IL-2 presentation probably deserves more attention including in mammals. Similarly, the ease with which a function of free IL-15 can be found in both mammals and trout, whereas not found for free IL-15L , underlines that the possibility of free IL-15 representing a distinct function from IL-15+IL-15Rα should be given more consideration. For trout IL-2 and IL-15L probable homodimer complexes were found , suggesting that previous observations of IL-2 and IL-15 homodimers in mammals should be studied more seriously as a potentially important trait of the IL-2/15/15L family.
Professor Dijkstra, who wrote and largely conceptualized the Yamaguchi et al. study  and is one of the two corresponding authors together with Professor Fischer, explains why understanding the evolution of this cytokine family is important: "Although IL-2 and IL-15 are very powerful and important cytokines, we only partially understand them. IL-2 and IL-2Rα are relatively simple to understand at the molecular level but complicated at the functional level because IL-2 simultaneously induces opposing functions. On the other hand, IL-15 and IL-15Rα seem relatively simple to understand at the functional level but are far from understood at the molecular level. Peculiarities such as the extremely long "leader" sequence of IL-15, multiple AUG codons in the IL-15 transcript 5'UTR, and multiple splicing forms of IL-15Rα, are poorly understood, yet seem to be important since conserved in both mammals and fish. Basically, finding similar features despite evolutionary separation for hundreds of millions of years indicates the importance of those features, and urges us to study them better not only in the non-model species but also in human and mouse."
In short, the recent Yamaguchi et al. study, Ancient cytokine interleukin 15-like (IL-15L) induces a type 2 immune response, published in Frontiers in Immunology https://doi.org/10.3389/fimmu.2020.549319 , concludes that the IL-2/15/15L cytokine family already played a major role in immune polarizations in an ancestor of jawed vertebrates, with opposing functions for IL-15 and IL-15-like.
 Yamaguchi T, Chang CJ, KargerA, Keller M, Pfaff F, Wangkahart E, Wang T, Secombes CJ, Kimoto A, Furihata M, Hashimoto K, Fischer U, Dijkstra JM (2020) Ancient cytokine interleukin 15-like (IL-15L) induces a type 2 immune response. Frontiers in Immunology 11:549319
 Dijkstra JM, Takizawa F, Fischer U, Friedrich M, Soto-Lampe V, Lefèvre C, Lenk M, Karger A, Matsui T, Hashimoto K (2014a) Identification of a gene for an ancient cytokine, interleukin 15-like, in mammals; interleukins 2 and 15 co-evolved with this third family member, all sharing binding motifs for IL-15Rα. Immunogenetics 66:93-103.
 Dijkstra JM (2014b) TH2 and Treg candidate genes in elephant shark. Nature 511:E7-9.
 Wang T, Hu Y, Wangkahart E, Liu F, Wang A, Zahran E, Maisey KR, Liu M, Xu Q, Imarai M, Secombes CJ (2018) Interleukin (IL)-2 Is a Key Regulator of T Helper 1 and T Helper 2 Cytokine Expression in Fish: Functional Characterization of Two Divergent IL2 Paralogs in Salmonids. Front Immunol 9:1683.
 Wen Y, Fang W, Xiang LX, Pan RL, Shao JZ (2011) Identification of Treg-like cells in Tetraodon: insight into the origin of regulatory T subsets during early vertebrate evolution. Cell Mol Life Sci 68:2615-2626.
Frontiers in Immunology