IL-30† (IL-27A): a familiar stranger in immunity, inflammation, and cancer

Over the years, interleukin (IL)-27 has received much attention because of its highly divergent, sometimes even opposing, functions in immunity. IL-30, the p28 subunit that forms IL-27 together with Ebi3 and is also known as IL-27p28 or IL-27A, has been considered a surrogate to represent IL-27. However, it was later discovered that IL-30 can form complexes with other protein subunits, potentially leading to overlapping or discrete functions. Furthermore, there is emerging evidence that IL-30 itself may perform immunomodulatory functions independent of Ebi3 or other binding partners and that IL-30 production is strongly associated with certain cancers in humans. In this review, we will discuss the biology of IL-30 and other IL-30-associated cytokines and their functions in inflammation and cancer.


Introduction
Adaptive T cell responses are shaped by multiple factors, and soluble mediators, especially cytokines produced by 'innate' immune cells, play an instructive role in the process 1 . Heterodimeric cytokines that belong to the IL-12 family are instrumental to the generation of inflammatory Th1 and Th17 immunity. The p35 and p19 subunits covalently linked to the common p40 subunit form bioactive IL-12 and IL-23, respectively, and are produced by activated antigen-presenting cells, such as macrophages and dendritic cells, following stimulation with innate stimuli via pattern recognition receptors, including TLRs 2 . These cytokines trigger differentiation processes in responding T cells through which the T cells acquire distinct effector properties conferring the ability to eliminate antigenic organisms. IL-12 activates the Stat4/T-bet pathway and generates IFNγproducing Th1 effector cells capable of clearing intracellular pathogens, whereas IL-23 amplifies Th17 programs initiated through the IL-6-TGFβ-Rorγt axis, clearing extracellular bacterial or fungal infections 3,4 . The essential protective immunity mediated by these cytokines requires tight regulation, and inadequate control results in autoimmune inflammation. Targeting or utilizing these key cytokines and/or their signaling pathways is being exploited in therapeutic approaches to intervene in inflammatory disorders, including inflammatory bowel disease, psoriasis, and multiple sclerosis, and to develop cancer immunotherapies 2 .
IL-30, also known as IL-27p28 or IL-27A, was discovered in a search for novel helical cytokines 5 . It forms a noncovalent bond with the Ebi3 subunit, which also associates with the IL-12p35 subunit to form IL-35 6 . The IL-30/Ebi3 complex, named IL-27, exhibits antiinflammatory (and regulatory) functions. It was initially thought to be an IL-12-like cytokine promoting Th1 immunity because of its ability to induce T-bet and IL-12Rβ2 expression through STAT1 activation during Th1 differentiation 7 . It was subsequently discovered that IL-27 also exerts a potent inhibitory role during Th17 differentiation 8 . It prevents activated T cells from acquiring RORγt and IL-17 expression via a STAT1-dependent mechanism 9 . IL-27 also suppresses Th2 immunity by inhibiting Th2 differentiation and cytokine production or by directly inhibiting group 2 innate lymphoid cells (ILC2s) 10,11 . IL-27 is a potent inducer of IL-10 in activated T cells but does not induce Foxp3 expression 12 . IL-27 also suppresses TGFβ-induced Foxp3 + regulatory T cell differentiation 13 , although it tends to support functions of already differentiated regulatory T cells 14,15 . Hence, IL-27 mediates an extremely broad array of functions involved in controlling T-cell immunity by targeting multiple pathways. Further extending the combinatorial complexity of the IL-12 family, there is increasing evidence that IL-30 can be secreted independently of Ebi3 and mediates distinct immunoregulatory functions 16 . In this review, we will discuss the immunobiology and functions of IL-30 and IL-30-associated cytokines (primarily IL-27) in the settings of inflammation and cancer (Table 1).

IL-30 and IL-27
The main source of IL-30 is cells of myeloid origin, namely, monocytes, macrophages, and dendritic cells 17 . IL-30 can also be produced by microglial cells and astrocytes in the central nervous system, alveolar macrophages and interstitial macrophages in lung tissue, and Kupffer cells in the liver [18][19][20] . Extensive efforts have been made to identify the cellular mechanisms inducing IL-30 expression. Molle et al. reported that TLR3 and TLR4 ligation in murine bone marrow-derived dendritic cells or human monocyte-derived dendritic cells triggers Il30 (Il27p28) mRNA synthesis via IRF3-dependent mechanisms 21 . ELISA experiments measuring IL-27 production in this study utilized the IL-27p28/IL-30 Quantikine kit from the R&D Systems, which primarily measures IL-30 with modest cross-reactivity with recombinant IL-27 (Ebi3/p28 fusion protein) (https://resources.rndsystems. com/pdfs/datasheets/m2728.pdf). Thus, the cytokine measurements mostly represented IL-30 but also included the IL-27 heterodimer. Liu et al. 22 reported that Il30 (Il27p28) mRNA expression and IL-30 (IL-27p28) protein secretion (IL-27p28/IL-30 Quantikine ELISA kit from R&D Systems) in macrophages are induced through a MyD88-mediated activation pathway and that combined stimulation with IFNγ dramatically enhances the production. McNab et al. showed that type I IFN enhances IL-27 secretion in M. tuberculosis-infected macrophages 23 . The IL-27 ELISA kit from eBioscience (Thermo Fisher), which detects both forms of the cytokine and does not distinguish between the IL-27 heterodimer and IL-30, was used in this study. Therefore, stimulation via innate receptors or cytokines is the primary signal responsible for the generation of IL-30 and/or IL-27 in myeloid cells.
Kimura et al. 24 unexpectedly discovered that a subset of murine CD4 T cells is also capable of secreting IL-30 and the IL-27 heterodimer. In this study, IL-30 secretion was measured using the Quantikine ELISA kit described above, while IL-27 heterodimer production was determined using the IL-27 LEGEND MAX TM ELISA kit from BioLegend, which utilizes anti-IL-30 and anti-Ebi3 mAbs as the capture and detection Abs, respectively. The authors found that IL-27 + CD4 T cells exert a regulatory function, inhibiting the IL-2 production and proliferation of other T cells during malaria infection. These regulatory T cells are Foxp3 − CD11a + CD49d + antigen-specific cells and are distinct from Foxp3 + regulatory or IL-10producing Tr1 T cells 24 . B cells also produce the IL-27 heterodimer in response to TLR and CD40 signaling, which was measured by sandwich ELISA using anti-IL-30 and anti-Ebi3 polyclonal Ab pairs (BioRXiv, doi.org/  B cells are not  antibody-secreting cells but cooperate with IFNγ to support proliferation, survival, class switching, and plasma cell differentiation, functioning as 'helper' B cells that link innate and adaptive signals for optimal antibody responses. In another study, B cell-derived IL-30 was shown to be critical in promoting CD8 T cell reconstitution following antibody-mediated lymphoablation 25 . In a study using an antibody-mediated lymphoablation model in solid organ transplantation, it was discovered that CD8 T cell reconstitution requires CD4 T cells and B cell Il30 mRNA expression 25 . In vivo neutralization using an anti-IL-30 mAb reversed CD8 T cell reconstitution 25 28 . Importantly, neutrophil-derived IL-30 can act directly on neutrophils, modulating bacteria-induced oxidative burst and cytokine production, i.e., antibacterial immunity 28 . The MaxDiscovery TM IL-30 ELISA test kit from Bio Scientific was used to specifically measure IL-30 secretion in the neutrophil study 28 . Whether neutrophils produce the IL-27 complex or IL-30 during sepsis remains to be determined. Overall, IL-30 and IL-27 can be secreted by multiple sources, and the mechanisms inducing their production could be highly diverse. It is important to distinguish IL-30 from IL-27, as the former can exert biological functions in the absence of Ebi3 (discussed below). It is also worth noting that different reagents for IL-30 and or IL-27 detection may provide different readouts that are difficult to adequately interpret.

Measuring IL-27 heterodimer production versus IL-30 production
While measuring the mRNA expression of p28 or p28associated subunits is widely used to assess expression, it is equally important to assess protein complexes from different sources. Commercially available human IL-27 ELISA reagents measure IL-30/Ebi3 heterodimers, as human IL-30 is not freely secreted without Ebi3 29 . In the case of murine IL-27, however, extra precautions must be made when the reagents are chosen. Early studies reported IL-27 production by ELISA, which in fact measured IL-30 secretion as well. Murine IL-27 ELISA kits available from R&D Systems or Thermo Fisher will detect not only IL-27 complexes but also IL-30. For example, the DuoSet IL-30 ELISA kit from R&D Systems has 4.6% crossreactivity with recombinant IL-30/Ebi3 proteins (the kit reads a sample containing 6.3 ng/ml recombinant mouse IL-30/Ebi3 as 286 pg/ml, https://resources.rndsystems. com/pdfs/datasheets/dy1834.pdf). Although the crossreactivity of other IL-30 ELISA reagents is unknown, substantial cross-reactivity is highly suspected. There are separate reagents capable of detecting IL-30/Ebi3 only. An ELISA kit from BioLegend was designed to detect the IL-27 heterodimer. However, it needs to be determined whether free IL-30 is also detected. Alternatively, Pascual and colleagues successfully utilized a combination of anti-Ebi3 and anti-IL-30 mAbs to detect murine IL-27 complexes in culture supernatant 30 . We also utilized anti-Ebi3 (R&D Systems, cat# MAB18341) and biotinylated anti-IL-27 Abs (detection Ab included in the Thermo Fisher IL-27 ELISA kit, cat# 88-7274-88) as capture and detection Abs, respectively, to detect the IL-27 complex but not IL-30 (our unpublished result).

Other IL-30 associated subunits
Another layer of complexity in IL-30 biology comes from its ability to form complexes with multiple subunits. In addition to the Ebi3 subunit, cytokine-like factor 1 (CLF1) can associate with IL-30. Crabe et al. reported that CLF1 encoded by the cytokine receptor-like factor 1 gene binds IL-30 and that the IL30/CLF1 heterodimer can be secreted by dendritic cells following activation 31 . The IL-30/CLF1 complex enhances NK cell activity, inhibits CD4 T cell proliferation, and induces IL-17 and IL-10 production, suggesting dendritic cell regulation of NK and T cell functions via this complex 31 . Tormo et al. 32 demonstrated that the IL-30/CLF1 complex also acts on B cells. In particular, the complex induces enhanced antibody production and plasma cell differentiation. Interestingly, IL-27Rα is not necessary for the IL-30/CLF1 complex to control B cell functions. Murine IL-30 has also been shown to form a complex with the IL-12p40 subunit. Wang et al. 33 utilized a bicistronic vector encoding both the p28 and p40 subunits and found that the two subunits could form a complex. The complex was shown to suppress T cell proliferation and differentiation into Th1/ Th17 phenotype cells but support Treg expansion 33 . Flores et al. independently engineered IL-30/p40 complexes using a 15 amino acid linker and named them "IL-Y" 34 . IL-Y expresses potent anti-inflammatory properties; however, the effect appears to be independent of Tregs, as it reduces the proportion of Tregs 34 .
Receptors and signaling of IL-27, IL-30, and IL-30associated cytokines As IL-30 is able to form distinct complexes with various partnering subunits, the receptors utilized by such divergent complexes are also variable. The receptor for IL-27 is a heterodimeric transmembrane protein composed of WSX-1 and gp130 5 . WSX-1, an IL-6/IL-12 family receptor, structurally resembles gp130 35 and was initially thought to be necessary for the development of Th1-type immunity 36 . gp130 is shared by receptors for IL-6, IL-35, ciliary neurotrophic factor, and cardiotrophinlike cytokine, while WSX-1 expression determines IL-27 responsiveness and is thus named IL-27Rα. WSX-1 is highly expressed on lymphocytes, but other cell types including dendritic cells, macrophages, and intestinal epithelial cells also express the receptor [37][38][39] . IL-27 signaling utilizes the Jak/Stat and p38MAPK pathways 17,40 , and more than one Stat molecule is activated by IL-27 stimulation. Among the STATs, Stat1 and Stat3 are the primary mediators of IL-27 signaling. The highly diverse functions of IL-27 appear to utilize different Stat pathways. For example, Stat1 is essential for IL-27-induced Tbet expression, whereas IL-27-induced proliferation utilizes the Stat3 pathway 41,42 . The IL-27-Stat3 axis also plays a critical role in IL-27-induced PD-L1 or CD39 expression in macrophages and DCs 43,44 . On the other hand, the IL-27-Stat1 axis seems to function in IL-27induced CD39 expression in regulatory T cells 45 . Therefore, IL-27-derived signaling pathways could differ depending on the target molecules and cell types.
The receptors utilized by IL-30/CLF1 complexes are somewhat distinct from the IL-30/Ebi3 pair. Utilizing Ba/ F3 transfectants, Crabe et al. 31 demonstrated that IL-30/ CLF1 activates cells expressing IL-6Rα in addition to WSX-1, utilizing a tripartite receptor (gp130/WSX-1/IL-6Ra). In support of this idea, blocking IL-6R with an anti-IL-6Rα Ab is sufficient to inhibit IL-30/CLF1-induced proliferation 31 . Analogous to IL-27, IL-30/CLF1 activates both the Stat1 and Stat3 pathways. IL-30 signaling via IL-6Rα raises the possibility that IL-30 may associate with the soluble IL-6Rα protein to 'trans-signal' in target cells through gp130-expressing target cells 31 . Indeed, an IL-30/ sIL-6Rα fusion protein was able to stimulate Stat3 pathways and to induce the proliferation of gp130 transfectants. However, whether this "trans-signaling" mechanism occurs during immune responses remains to be investigated.
There is emerging evidence that IL-30 is capable of exerting immunoregulatory functions independent of other subunits. Although the efficacy may be lower, IL-30 inhibits IL-12-induced liver toxicity as effectively as IL-27, independent of Ebi3 or IL-27Rα 46 . During in vitro T cell differentiation, IL-30 is capable of inhibiting Th17 differentiation in the absence of Ebi3; however, unlike IL-27, IL-30 is unable to induce IL-10 expression 16 16 . However, these findings were observed in conditions where highly variable concentrations of recombinant IL-30 were used. For example, 50-100 ng of IL-30 is needed to antagonize IL-6driven Stat3 phosphorylation or IL-27-driven Stat1 phosphorylation 47 . However, a study using transfection of gp130 and IL-6Rα showed that IL-30 induces phosphorylation of both Stat1 and Stat3 and that IL-30 can even activate signal transduction through gp130 without IL-6Rα at a higher concentration 47 . Likewise, a high concentration of human IL-30 can induce the phosphorylation of both Stat1 and Stat3 29 . Therefore, the precise ability of IL-30 to induce signaling demands further investigation.

IL-30 transgenic (Tg) mouse models
As an effort to understand the role of IL-30 in vivo, Stumhofer et al. generated IL-30 Tg mice in which IL-30 transgene expression is controlled by the lck proximal promoter and the immunoglobulin intronic heavy chain enhancer, driving IL-30 expression in B and T cells 16 . The IL-30 level in the serum of this animal model is 1-1.5 ng/ml. IL-30 overexpression in lymphocytes does not affect the number of mature B cells or the ratio of CD4 and CD8 T cells, although the total number of T cells with an activated phenotype is elevated. However, the production of inflammatory cytokines following in vitro stimulation was found to be comparable. No overt signs of tissue inflammation have been observed, and regulatory T cell development also remains intact. However, IL-30 overexpression results in the defective generation of antigen-specific IgG responses and germinal center formation. IL-30 Tg mice are resistant to autoimmune inflammation, EAE, and EAU, and the tissue infiltration of antigen-specific Th1 and Th17 cells is significantly reduced in this model 48 .

IL-27 Tg mouse models
Nakanishi and colleagues generated IL-27 Tg mice overexpressing single-chain IL-27, in which Ebi3 is flexibly linked to IL-30 with the (Gly4Ser)3 linker, under the liver-specific human serum amyloid P component promoter and the rabbit b-globin gene 11 . The resulting Tg mice have a circulating serum IL-27 complex level of 0.5-1 ng and exhibit no obvious morphological abnormalities in the liver or immunological defects in the primary and secondary lymphoid tissues. Following induction of intestinal parasitic infection or allergic airway inflammation, IL-27 Tg mice display inhibited Th2 immune responses. However, these mice display dysregulated hematopoiesis characterized by increased megakaryocyte numbers and shortened survival 49 . A separate IL-27 Tg mouse model was generated by breeding mice expressing IL-30 and Ebi3 transgenes 50 . Transgene expression was limited to T and B cells. These IL-27 Tg mice spontaneously develop and succumb to systemic inflammatory disease, with extensive immune infiltrates found in multiple tissues. Peripheral immune cells display highly activated phenotypes and elevated inflammatory cytokine production. Importantly, Foxp3 + regulatory T cell development is severely defective in these IL-27 Tg mice, which is attributed to reduced IL-2 production.

IL-27Rα-deficient mouse models
Germline IL-27Rα-deficient mice were developed by Mak and colleagues by targeting an exon encoding a part of the second fibronectin type III domain 51 . IL-27Rα deficiency does not affect hematopoietic or lymphocyte development. In vitro T cell proliferation is slightly increased, while IFNγ production is impaired. IL-27Rα −/− mice display high susceptibility to Leishmania major infection. In support of this susceptibility, IFNγ production by KO T cells is greatly reduced. Separate IL-27Rα −/− mice were also developed by Chen et al. 36 , in which exons 2-8 were targeted. Analogously, Th1-type cytokine production is greatly reduced in these KO mice. Defects in Th1 immunity were further tested in a Listeria monocytogenes infection model that relies on protective Th1 responses. Consistent with the in vitro findings, these KO mice were unable to clear the pathogens. We generated IL-27Rα conditional knockout animals to test the cell type-specific functions of IL-27 signaling 15 . When Foxp3 Cre transgenic mice were used in the cross to generate Treg-specific IL-27Rα −/− mice, we found that IL-27 signaling in Tregs is indispensable for Treg-mediated inhibition of autoimmune and allergic inflammation 15,52 , suggesting that Tregs may be more critical than conventional T cells in mediating IL-27-dependent control of chronic inflammatory responses.

IL-30-deficient mouse models
Germline IL-30 −/− mice were developed by Neurath and colleagues 53 . IL-30 −/− mice develop normally and do not display gross or histologic abnormalities. By using a T cell-dependent colitis model, IL-30 −/− mice were found to develop intestinal inflammation similar to that induced by wild-type T cells. On the other hand, IL-30 deficiency was associated with increased disease severity demonstrated by greater weight loss during intranasal Sendai virus infection 54 . This susceptibility was not due to defects in viral clearance but was associated with more severe pathology, suggesting that IL-30 or IL-27 is critical in limiting pathology during viral infection. An independent IL-30 −/− mouse model was also developed by Li and colleagues 55 . IL-30-deficient mice succumb to death following sublethal LPS injection. IL-30 conditional knockout mice were developed by Zhang et al. 56 . Dendritic cellspecific IL-30 −/− mice are highly susceptible to concanavalin A-induced hepatitis, which is attributed to uncontrolled production of IFNγ by CD4 T cells.

IL-27 and IL-30 in inflammation
The ability of IL-27 to suppress inflammatory responses, particularly in autoimmunity, has extensively been investigated. The initial observation that IL-27Rα-deficient mice are highly susceptible to EAE strongly suggests that IL-27 plays a key role in regulating self-reactive Th17-type encephalitogenic T cells 57 . In support of this hypothesis, we and others have demonstrated that systemically administered IL-27 inhibits the development of EAE 15,58 . The primary target cells of IL-27 appear to be CD4 T cells, as Th17 differentiation is directly suppressed by IL-27 57 . However, we showed that IL-27 acting on Foxp3 + Treg cells might be more critical for the in vivo actions of IL-27. This conclusion was drawn by utilizing a Treg cell-specific IL-27Rα −/− mouse model that develops severe EAE following induction. Given that the IL-27 responsiveness of conventional CD4 T cells remains intact in this model, it is intriguing to note that Treg stimulation by IL-27 is indispensable for IL-27mediated regulation of autoimmune inflammatory responses. This notion is further supported by results from adoptive transfer experiments where only wild-type but not IL-27Rα −/− Treg cells protected mice from lethal EAE when transferred into recipients whose endogenous Treg cells were depleted 59 .
Similar cellular mechanisms of IL-27 regulating inflammatory responses have also been observed in allergic inflammation. Intranasal administration of IL-27 reduces eosinophil infiltration in the airways and airway hyperresponsiveness as well as allergic rhinitis 60 . The IL-27-induced effects on T cell cytokine production and Treg cell expansion are most pronounced, suggesting that IL-27 may directly act on T cells 60 . Interestingly, Chen et al. reported that preventive administration of IL-27 was able to attenuate allergic inflammation but that therapeutic administration had little effect 61 . We also reported that the primary target cells for the anti-inflammatory functions of IL-27 are Foxp3 + Treg cells, since intranasal administration of IL-27 loses its effect in the absence of Treg cells or in Treg cell-specific IL-27Rα −/− mice 52 .

Human bronchial epithelial cells are known to express the IL-27 receptor complex 62 . Indeed, IL-27 enhances ICAM-1 expression on bronchial epithelial cells and augments
TNFα-induced IL-6 production, which is partly attributed to the ability of IL-27 to increase TNFα receptor expression 62 .
It is worth noting that IL-27 may play divergent roles in different inflammatory diseases. The role of IL-27 in diabetes is different from that in other autoimmune diseases or allergic inflammation. Ciecko et al. reported that IL-27 is essential for type 1 diabetes development 63 . Utilizing NOD mice deficient in IL-30 or IL-27Rα, they found that these mice are completely resistant to diabetes development. The primary targets of IL-27 underlying this resistance seem to be T cells, as IL-27R −/− splenic T cell transfer into NOD.Rag1 −/− recipients recapitulated resistance 63 . Mechanistically, IL-27 directly regulates the balance of Treg and Th1-type CD4 T cells, while it also modulates the pathogenic functions of CD8 T cells. Spontaneous development of autoimmune inflammation in the lacrimal and salivary glands in NOD mice is similarly affected by IL-27 63 . Interestingly, Qi et al. 64 recently reported contrary findings for IL-27 in experimental Sjögren's syndrome. Utilizing the same IL-30 −/− NOD mouse model, it was reported that IL-30 deficiency results in aggravated disease manifestation and that severe disease is associated with a reduction in IL-10 + CD4 T cell subsets. Exogenous IL-27 administration attenuates inflammation and increases IL-10 + CD4 cells. In support of this, the levels of plasma IL-27 and IL-10 + CD4 T cells are decreased in patients with Sjögren's syndrome 64 .
IL-27 induces the inhibitory receptor T cell immunoglobulin and mucin domain-3 (Tim-3) and IL-10 via a nuclear factor, NFIL3 65 . NFIL3-overexpressing CD4 T cells attenuate T cell-induced colitis, and elevated IL-10/Tim-3 and reduced IFNγ expression are found in these T cells. Furthermore, IL-27 treatment fails to suppress inflammation in NFIL3-deficient T cells.
A potent anti-inflammatory function of IL-30 was first reported in a cytokine-induced liver injury model. Li and colleagues reported that IL-30 expression was induced by inflammatory cytokines such as IL-12 and IFNγ and that IL-30 administered via hydrodynamic delivery of a plasmid encoding IL-30 attenuated liver fibrosis induced by CCl 4 or hepatotoxicity induced by IL-12 independent of IL-27 or IL-27Rα 46,66 . Mechanistically, IL-30 inhibits hepatotoxicity by reducing the IFNγ level 46 . Moreover, the IL-30-induced reduction in liver fibrosis occurs via the removal of activated hepatic stellate cells by NK cells recruited to the liver 66 . Another approach for systemic IL-30 delivery using adenoassociated viral vectors (AAVs) found that systemic IL-30 delivery only slightly inhibited EAE development 67 . Of note, the serum IL-30 level measured following AAV-IL-30 administration was maintained at 200 ng/ml 67 , whereas the level following hydrodynamic plasmid delivery was only~1 ng/ml 66 . Whether different serum IL-30 levels underlie the discrepant findings remains to be examined. IL-30 has also been shown to attenuate experimental sepsis by modulating NKT cell cytokine profiles 55 . Caspi and colleagues utilized IL-30 Tg mice and reported that IL-30overexpressing mice are highly resistant to autoimmune inflammation, EAE, and EAU (experimental autoimmune uveitis), primarily by antagonizing Th1 and Th17 responses 48 . Whether this is due to IL-30 or to the elevated formation of the IL-27 complex remains unclear.

IL-27 and IL-30 in infection
The role of IL-27 was first tested in a Toxoplasma gondii infection model, in which protective immunity is dependent on the development of Th1-type immunity. Villarino et al. reported that infected IL-27Rα −/− mice succumbed to infection due to uncontrolled T cell-mediated inflammatory responses 68 . The parasite burden was effectively controlled; however, T cells in this condition displayed a highly activated phenotype with increased proliferative activity 68 . Similar IL-27-dependent regulation of CD4 T cell immunity was also noted in an African trypanosome infection model 69 . Montes de Oca et al. recently reported a similar finding in a Leishmania infection model, in which IL-27R deficiency significantly expedited parasite clearance at the expense of enhanced tissue damage caused by an imbalance in antigen-specific Th1-and Tr1-type CD4 T cells 70 . Interestingly, IL-27 exerted these effects by regulating the metabolic profiles of Th1-type CD4 T cells 70 . IL-27Rα −/− CD4 T cells were more glycolytic, suggesting that IL-27 seems to limit Th1 cell glycolysis to further protect against tissue pathology during infection.
The role of IL-27 was further explored in viral infections. Zuniga and colleagues reported a T cell-intrinsic role for IL-27 in chronic viral infection 71 . IL-27R deficiency in T cells impairs the accumulation of virusspecific CD4 T cells and the control of the viral load during chronic LCMV infection. Moreover, IL-27 signaling has been shown to control key innate immune cells during early LCMV infection, as type I IFN expression by DCs and NK cell function indicated by granzyme and IFNγ production seem to be affected by IL-27R deficiency 72 . IL-27 signaling is also important for the generation of IFNγ-producing CD8 T cells in infection. Mohrs and colleagues reported that in both T. gondii and influenza virus infection models, IL-27Rα −/− CD8 T cells are defective in IFNγ production 73 .
In addition to CD4 and CD8 T cells, different target cells of IL-27 have been identified in other infection models. Moro et al. reported that IL-27 antagonizes the function of ILC2s using an Alternaria alternata-induced lung inflammation model 74 . A. alternata is a major fungus associated with ILC2-mediated asthmatic inflammation. Coadministration of A. alternata extract with IL-27 results in significantly reduced eosinophilia and type 2 cytokine production in the lungs. IL-27 reduces the production of eotaxin and type 2 cytokines by ILC2s but not by Th2 CD4 T cells. In support of this finding, IL-27Rα expression in ILC2s is greater than that in CD4 T cells, suggesting that IL-27 may have different impacts on ILC2s versus T cells 74  Whether IL-30 plays a regulatory role in the setting of infection has not been formally explored.

IL-27 and IL-30 in cancer
Divergent roles for IL-27 and IL-30 in anticancer immunity are being increasingly appreciated. Because of the abilities of IL-27 to induce IFNγ and activate NK cell activity, the antitumor function of IL-27 has been examined. Utilizing single-chain IL-27-secreting colon carcinoma and murine neuroblastoma models, it was shown that these tumor cells induce robust antitumor immune responses 76,77 . IFNγ neutralization or CD8 T cell depletion abrogates these antitumor effects, suggesting a link between IL-27 and adaptive cytotoxic immunity 76,77 . In NK cells, IL-27 directly activates T-bet and granzyme B expression and enhances both viability and cytolytic activity 78 . IL-27 gene transfer results in rejection of NK cell-resistant head and neck squamous cell carcinoma by NK cells 78 . Alternatively, IL-27 can support hematopoietic stem cell differentiation into M1type macrophages and enhance their antitumor effects. Yoshimoto and colleagues reported that engineering B16F10 melanoma cells to express IL-27 substantially increased the number of tumor-infiltrating CD11b + myeloid cells with expression of M1-type genes including iNOS, IRF8, and Il12p40 and decreased expression of M2-associated markers such as Arg-1, Ym1, and Fizz1 49,79 . The importance of myeloid cells was further corroborated by anti-Gr1 Ab treatment, which abrogated the antitumor effects of IL-27 79 . IL-27 is also able to directly act on tumor cells to perform antitumor functions. IL-27 inhibits the proliferation of B16 melanoma transfectants expressing IL-27R by augmenting the expression of IRF-1 80 . It was further demonstrated that IL-27 has broad impacts on the expression of genes involved in angiogenesis and invasion. Zorzoli et al. 81 reported that IL-27 inhibits acute myeloid leukemia (AML) cell growth in an NSG mouse model by upregulating antiangiogenic genes such as IFNγ, CXCL10, and tissue inhibitors of metalloproteinase (TIMP)-2 and by downregulating genes involved in tumor spreading such as cadherin-6 (CDH6). Similar antiproliferative properties of IL-27 were also observed in prostate cancer cells. Di Carlo et al. reported that IL-27 inhibits prostate cancer cell proliferation and modulates the expression of genes involved in the angiogenic process such as fms-related tyrosine kinase, fibroblast growth factor receptor, CXCL10, and TIMP3 82 . Different IL-27 target genes were also identified in primary human multiple myeloma cells 83 . Incubation of cancer cells with IL-27 results in downregulation of proangiogenic genes, including AKT, angiopoietin, MMP9, VEGF, and laminin, and in upregulation of antiangiogenic genes, such as CXCL9 and CXCL10 83 . IL-27 also downregulates multiple stemnessrelated genes in lung adenocarcinoma and squamous cell carcinoma cell lines and inhibits tumor cell growth in xenograft models 84 . Last, IL-27 can inhibit human melanoma cell growth by inducing the expression of TNF-related apoptosis-inducing ligand 85 . Overall, IL-27 appears to exert antitumor effects by targeting multiple genes and pathways. It is worth noting that IL-27R expression (i.e., an indicator of IL-27 responsiveness) is lost in high-grade and advancedstage prostate cancer cells 82 . Since tumor-infiltrating immune cells express functional IL-27R, harnessing the ability of IL-27 to stimulate antitumor activity in immune cells could be an effective approach.
However, the protumorigenic functions of IL-27 have also been observed. As discussed above, IL-27 is a potent inducer of IL-10 in T cells, and IL-10 is capable of inhibiting antitumor immunity 86 . Karakhanova et al. reported that IL-27 induces PD-L1 expression on DCs and that IL-27-treated DCs display a reduced capacity to stimulate T cell proliferation and cytokine production 87 . IL-27 acting on DCs has an additional impact, negatively regulating T cell immunity. It was shown that IL-27 induces CD39 expression in DCs and that IL-39 reduces extracellular ATP and diminishes nucleotide-dependent activation of the inflammasome, resulting in T cell inhibition 44 . The protumor activity of IL-27 may also be mediated by Treg cells. We previously reported that IL-27 stimulation of Treg cells is necessary for Treg cells to properly control T cell responses 15 . Park et al. 45 recently reported that IL-27 stimulation induces CD39 expression on tumorinfiltrating Treg cells, leading to inhibition of CD8 T cell activation. IL-27R deficiency in Treg cells is thus sufficient to rescue antitumor CD8 T cell activity from Treg cellmediated inhibition 45 . In addition, IL-27 may induce immune inhibitory molecules in human epithelial ovarian cancer cells, including the IL-18 binding protein, an inhibitor of IL-18 proinflammatory activity 88 ; indoleamine 2,3-dioxygenase (IDO), which depletes tryptophan and causes T cell dysfunction and death 89 ; and PD-L1, which inactivates T cell function by engaging PD-1 on activated T cells 90 . Finally, IL-27 also directly promotes the proliferation and survival of AML cells 91 . This pro-proliferative/ survival effect of IL-27 seems to operate via the MAPK/ERK signaling pathway. IL-27 is also capable of interfering with the responsiveness of cancer cells to the chemotherapeutic drugs cytarabine and daunorubicin 91 .
In contrast, the roles of IL-30 are mostly protumorigenic. Di Meo et al. reported for the first time that IL-30 may support prostate cancer development 92 . From immunohistochemistry and real-time RT-PCR analyses, elevated IL-30 expression was found in prostate cancer epithelial cells and seemed to be correlated with an advanced grade and stage 92 . IL-30 expression was most pronounced in tumor-infiltrating leukocytes, particularly CD68 + macrophages, and was found even in infiltrating leukocytes in the prostate-draining lymph node. The mechanism underlying the action of IL-30 on prostate cancer progression may be linked to the ability of IL-30 to support prostate cancer stem-like cell (PCSLC) survival 93 . PCSLCs are able to produce IL-30 and promote tumor onset by enhancing proliferation and vascularization. Airoldi et al. 94 reported that IL-30 expression is similarly elevated in breast cancer patients, particularly those with triple-negative HER2 + subtypes, and that analogous to prostate cancers, IL-30 is found in inflammatory leukocytes infiltrating tumors and the draining lymph nodes, and its expression correlates with the stage of the disease. Mechanistically, IL-30 alters the PCSLC expression of genes involved in immune suppression, stemness, and metastasis. IL-30 also increases the expression of the proto-oncogene Myc and Muc1, along with that of the genes encoding the growth factors EGF and VEGF-A 94 . In a xenograft model system, local IL-30 injection into mice implanted with breast cancer cells substantially upregulated genes involved in tumor growth and angiogenesis 94 .

Disparities in IL-30 biology between mice and men
Most of our insights into IL-30 biology stem from mouse models, but analysis of human and mouse IL-30 reveals an important disparity in the secretion pattern of this cytokine. Mouse IL-30 is readily secreted by various cell types, immune cells, and transfected model cell lines into the culture supernatant in the absence of Ebi3. Coexpression of Ebi3 further enhances the secretion of IL-30 as part of the heterodimeric IL-27 complex 29,95 . In contrast, human IL-30 is not secreted in isolation but instead retained in the endoplasmic reticulum of cells, where it becomes a target of ER-associated degradation, showing that human IL-30 does not fold properly in isolation 29 (Fig. 1). Coexpression of Ebi3 induces secretion of human IL-30 95 , and it has recently been shown that human IL-30 and Ebi3 indeed form a stable complex, IL-27, so that cellular retention of Ebi3 coretains IL-30 in cells 29 . This discrepancy was attributed to a single amino acid difference between the species. Mouse IL-30 has two cysteines, which can form a disulfide bond to stabilize the protein. The first cysteine residue is located in helix B of the four-helix bundle fold of IL-30, and the second cysteine residue is located in a large flexible loop connecting the C and D helices 96,97 . Human IL-30 has a leucine at the position corresponding to the second cysteine 96,97 . Importantly, the substitution of this leucine with a cysteine (L162C) in human IL-30 is sufficient to render it secretion competent in isolation regardless of Ebi3 expression, similar to murine IL-30 29 . Along the Fig. 1 Species differences in IL-30 and IL-27 biogenesis. a The biogenesis of human (h) IL-27 is shown. Human IL-30 depends on assembly with Ebi3 to fold, be released from the endoplasmic reticulum (ER) chaperone Bip and function once outside the cell. If it remains unpaired, it is degraded by ER-associated degradation (ERAD). b In contrast, murine (m) IL-30 has two structurally adjacent cysteines c that can form a disulfide bond, so murine IL-30 can fold and be secreted in isolation or as a pair with Ebi3. same line, mutation of the second cysteine in murine IL-30 (C158L) renders its secretion dependent on the coexpression of Ebi3 29 . Detailed computational and biochemical analyses have provided insights into the underlying mechanism: disulfide bond formation in IL-30 leads to the shielding of hydrophobic residues, enabling escape from the ER Hsp70 chaperone BiP for autonomous secretion 29 . A species comparison revealed that this mechanism is strictly conserved in evolution: whenever IL-30 forms a disulfide bond, it can be secreted autonomously, becoming capable of performing biological functions; if it fails to form a disulfide bond, e.g., due to having only a single cysteine, its secretion depends on Ebi3 96 . Mice and humans are thus only representatives of a larger number of species in regard to IL-30 secretion. This finding is of particular interest since cysteines are among the most highly evolutionarily conserved amino acids 98 , implying functional selection criteria during evolution that have favored the presence or absence of secretion-competent IL-30. IL-12 heterodimers, including IL-27, have likely evolved from cytokine:receptor pairs 99 . During evolution, in certain species, including humans, IL-27 has evolved not only into a cytokine but also into an obligate heterodimeric cytokine by losing one of the cysteines in the IL-30 subunit. Although humans thus seem to have generally lost secretion of IL-30 during evolution, human cells remain responsive to this cytokine 29 , which potentially opens up doors for using engineered human IL-30 for immune modulation. Furthermore, the multiple pairings of IL-30 and roles for IL-30 in human immunity may suggest that we do not appreciate the full picture of human IL-30 secretion and functions yet. As an example, polymorphisms in the IL-30 gene have been reported in numerous disease settings, including premature coronary artery disease 100 , bladder cancer 101 , osteosarcoma 102 , asthma 103 , autoimmune thyroid disease 104 , and IBD 105 . Whether these mutations alter the folding and secretion pattern of IL-30 or one of its pairings remains to be determined.

Conclusion and outstanding questions
Given the emerging evidence for the ability of IL-30 to regulate immune responses by itself, it is evident that IL-30 is more than a partnering subunit that forms IL-27 complexes. Here, outstanding questions demanding future investigations are discussed. First, the formation of various IL-30-associated complexes, their precise sources and tissue sites, and the mechanisms involved in their production need to be examined. It is possible that different sources of myeloid cell origin may be the primary sources of IL-30-associated cytokine complexes and that the nature of inflammatory cues and tissue factors may play a role. Second, reagents specific for IL-30 and IL-30associated complexes need to be developed. As discussed above, one cannot distinguish the production of IL-30 from that of complexes containing IL-30. ELISA reagents that specifically detect IL-30/Ebi3 complexes but not IL-30 are now available and are beginning to be used. Because of considerable cross-reactivity, however, detecting IL-30 free of other binding partners remains challenging. Third, based on the disparity between murine and human IL-30 secretory behavior, it is important to develop an animal model where IL-30 behaves like human IL-30, i.e., lack of secretion of IL-30 in the absence of binding proteins. This model could be useful for gaining insights into the functions of IL-30 in vivo. Fourth, elevated IL-30 secretion is strongly associated with numerous types of cancer, raising the possibility that dysregulated IL-30 secretion may be linked to cancer development. Therefore, it is important to understand the causal relations among IL-30 polymorphisms; IL-30 folding, secretion, and pairing patterns; and the progression of diseases, including tumors and other inflammatory conditions. Note † IL27 (also known as IL30, IL-27A, p28, or IL27p28) is an approved name for the interleukin-27 alpha subunit (HUGO Gene Nomenclature Committee). In this review paper, we will use IL30 (Il30) and IL-30 for the gene and protein product, respectively.