Interleukin-9 as a T helper type 17 cytokine

Abstract

The production of interleukin-9 (IL-9) by CD4 T cells has gathered renewed interest as the result of the observation that its expression is broader than originally thought. This includes the production of IL-9 by a recently characterized subset of CD4 helper T (Th) cells that are termed Th9 as well as production by additional T-cell subsets including Th17 cells. There is an incomplete understanding as to which IL-9-producing T-cell subsets develop under physiological conditions. We describe the conditions used to generate IL-9 in Th17 cells in vitro. We also summarize conditions where both IL-9 and IL-17 are found in vivo and propose that Th17 cells producing IL-9 may co-exist and interact with Th9 cells during conditions of autoimmunity, allergy and infection.

Introduction

The capacity for CD4 T cells to differentiate into different subsets has been of interest since the discovery of the first two helper T-cell subsets by Mosmann et al.¹ These were Th1 cells, which produce interferon-γ (IFN-γ) and help to clear intracellular bacteria and viruses, and Th2 cells, which produce interleukin-4 (IL-4), IL-5 and IL-13 and aid in the clearance of extracellular pathogens. Several additional subsets have emerged, some of which include: regulatory T cells, which express the transcription factor FoxP3 and help to suppress immune responses; Th17 cells, which express IL-17 and are involved in the clearance of extracellular bacteria and fungi; and Th22 cells, which express IL-22 but not IL-17 and are associated with pathogenic responses in the skin.^2–6 However, even as these subsets are created to simplify immune responses there is increasing evidence that T cells that initially acquire one phenotype do not necessarily maintain it. This is particularly true of the Th17 subset.^7,8

Interleukin-9 is a member of the common γ-chain family of cytokines, using this receptor in combination with the cytokine-specific receptor IL-9Rα. It has primarily been studied as a mediator of immunity to nematode infection as well as a pathogenic factor in atopic responses.^9–11 There is renewed interest in IL-9 because of the identification of a subset of T cells that produce this cytokine. However, there are conflicting reports over which T cells can produce this cytokine. They include, but are not limited to, Th2, Th9, Th17 and regulatory T (Treg) cells.¹² Here we overview the production of IL-9 by some of these T-cell subsets with an emphasis on the capacity of Th17 cells to produce this cytokine.

IL-9 production in T cells: Th9, Th2 and Treg

The most consistent production of IL-9 production in T cells is from those generated with the cytokines transforming growth factor-β (TGF-β) and IL-4, which have been coined Th9. First described by Schmitt et al.,¹³ this subset requires IL-2 for their generation and can be inhibited by IFN-γ. These studies were repeated and further expanded to show that TGF-β responsiveness is also required for their generation.¹⁴ In addition, this subset does not express the transcription factors T-bet, GATA-3, RORγt, and FoxP3 and do express the transcription factor PU.1 in both human and mouse T cells.^14–16 In addition, there are reports that in mice Th9 cells largely co-express IL-10.^14,16 However, further studies demonstrated that while Th9 cells may express both IL-9 and IL-10, IL-10 production in Th9 cells is not regulated by the transcription factor PU.1.¹⁵ Data using humans T cells also show that simultaneous production of IL-9 and IL-10 does not occur under Th9-cell conditions.

In contrast to Th9 cells, the production of IL-9 by Th2 cells and Treg cells is more controversial. In the case of Th2 cells, which are generated in the presence of IL-4, it has been suggested that low levels of TGF-β present in media could cause the production of a low number of Th9 cells in these cultures. Differentiation of T cells in vitro also shows that Th9 cells can be generated from Th2 cells,¹⁴ and IL-9 production seems to correlate with low IL-4 expression.¹⁵ This suggests that co-expression of IL-4 and IL-9 may represent T cells that are in transition between the Th2 and Th9 subsets or that are T cells that have not committed to either lineage.¹⁵ In the case of Treg cells, it appears that IL-9 production varies greatly in vitro. Using intracellular staining of human T cells from healthy individuals, one group reported co-expression of IL-9 and FoxP3 whereas another did not.^17,18 Additional data generated in mice, including the sorting of FoxP3⁺ Treg cells and re-stimulation in vitro as well as the immunofluorescent staining of FoxP3 and IL-9 in sections from tolerant skin grafts, suggest that some Treg cells can produce IL-9.^12,19,20 There is also one report that IL-9 production in Treg cells can be reduced by retinoic acid.²¹ This suggests that one explanation of the conflicting reports could reflect the relative amount of retinoic acid present in the serum used in T-cell cultures in vitro as well as the amount present locally at different sites in vivo.

IL-9 production in Th17 cells

The Th17 cells have emerged as a T-cell subset that has the ability to produce IL-9 both in vitro and in vivo. During primary Th17 differentiation cultures of murine naive T cells using TGF-β and IL-6, some co-production of IL-17A and IL-9 has been observed by intracellular staining. This production seems to require low levels of IL-2 in the culture.²² In human T cells, similar observations have been made. However, unlike the situation with mouse T cells, primary T-cell cultures produced very small amounts of IL-9. However, with further skewing using in vitro methods a portion of Th17 cells can acquire the expression of IL-9.¹⁷ Consistent with these results, sorted IL-17F⁺ and RORγt⁺ CD4 T cells, respectively, from mice immunized with myelin oligodendrocyte peptide (MOG)_35–55 emulsified in complete Freund's adjuvant can produce IL-9 upon re-stimulation.¹² In addition, conditions used to stimulate IL-17A production from human memory CD4 T cells can also promote IL-9 expression. This co-production of IL-17A and IL-9 by CD4 T cells can be further enhanced either by IL-1β or IL-21 alone as well as their use in a Th17 cocktail of cytokines.^17,18,23

Functionally, the production of IL-9 by Th17 cells may have some autocrine effect on Th17 expansion because in vitro there is reduced accumulation of these during their initial differentiation.²⁴ These results are also consistent with the observation by another group that under standard immunization conditions used to induce experimental autoimmune encephalomyelitis (EAE), a mouse model of multiple sclerosis, IL-9R-deficient mice have a decreased accumulation of Th17 cells.¹²

Unlike the case in vitro where Th17 cells can be studies in isolation of other IL-9-producing T cells, in vivo it is possible that in certain circumstances they may co-exist and interact with each other. Indeed, IL-21 is one of the cytokines shown to enhance Th9 differentiation in vitro.²³ This cytokine is of particular interest because Th17 cells are one potential source of it and it also helps to promote their own expansion.^25–27 One possible model (shown in Fig. 1), is that Th17-derived IL-9 can have autocrine effects to promote their own expansion and paracrine effects to enhance mast cell accumulation. In addition, Th17-derived IL-21 could be furthering the differentiation of Th9 T cells, which in turn may also be acting on Th17 cells and mast cells, as well as promoting their own expansion. Supporting this idea, when T cells from IL-17F reporter mice immunized with MOG_35–55 emulsified in complete Freund's adjuvant were sorted into IL-17F⁺ CD44^hi and IL-17F− CD44^hi subsets, IL-9 production was observed in both groups after re-stimulation. These observations suggest that IL-9 production in T cells was likely to be from both Th17 and Th9 T cells during EAE.¹² It has also been observed in vitro that many of the IL-17 promoting cytokines in humans including TGF-β, IL-1β, IL-6, IL-21 and IL-23 do not inhibit and in some cases actually promote the generation of IL-9 producing Th9 and/or Th17 cells.^17,18,23

Figure 1.

Proposed model of T helper type 17 (Th17), Th9, and mast cell interactions. Th17 derived interleukin-9 (IL-9) and IL-21 can act in a paracrine manner to further their own expansion or in an autocrine manner to further the expansion of Th9 cells. Interleukin-9 derived from both of these subsets can also act on mast cells. TGF-β, transforming growth factor-β.

There is some evidence to suggest that IL-9 production in Th17 cells is pathogenic during autoimmunity. Memory T cells from the peripheral blood of patients with type 1 diabetes have a significantly increased frequency of IL-17⁺ IL-9⁺ T cells and total IL-9⁺ T cells after re-stimulation under Th17 differentiation (TGF-β, IL-1β, IL-6, IL-21 and IL-23) in comparison with those of healthy control subjects. This difference was not observed if the memory T cells were re-stimulated with a mixture of IL-1β, IL-6, IL-21 and IL-23 or with TGF-β by itself. There was also strong linear correlation between the total IL-17⁺ T cells to the total IL-9⁺ IL-17⁺ T cells but not correlation between total IL-9⁺ T cells and these double-positive T cells.¹⁷ This suggests that the Th17 population in patients with type 1 diabetes have an increased propensity to secrete IL-9; however, it falls short of definitively proving that these cells are there to increase disease severity.

Mouse studies in EAE also suggest that IL-9 is pathogenic in autoimmunity. Using a standard immunization for disease, in which IL-9 production by T cells is likely to be from both Th17 and Th9 cells, it was observed that IL-9Rα deficiency was associated with attenuated disease, decreased Th17 accumulation, and decreased production of IL-6 by macrophages in the central nervous system.¹² Additional work using an adoptive transfer model of EAE disease with myelin-specific T-cell receptor transgenic T cells that were skewed under Th17 conditions with low dose IL-2 (which could produce IL-9) as well as Th17 conditions with IL-23 (which results in a loss of IL-9 production in mice²⁴) could cause disease with similar severity after adoptive transfer. However, each produced a different pathological phenotype. It was also observed that Th9-skewed T cells could also cause disease symptoms with another distinct pathological phenotype, although it was less efficient than that produced by Th17 cells.²² This heterogeneity in phenotype may prove to be relevant in treating patients with different kinds of myelitis as it may prove to reflect the relative contribution of Th1, Th17 and Th9 cells. Indeed, it was observed that the polyclonal transfer of Th9 cells into mice deficient for T cells and B cells resulted in a peripheral neuropathy in a subset of recipient mice.¹⁶ In humans, it was also observed that IL-9 is found in the cerebrospinal cord fluid of patients with both atopic myelitis and opticospinal multiple sclerosis; however, IL-9 is only correlative with IL-17 in the latter group.²⁸ Further research may show that IL-9-producing Th17 cells may be having a pathogenic effect in opticospinal multiple sclerosis while Th9 cells are having a principle effect in atopic myelitis patients.

Allergic responses represent another area where IL-9 production by Th17 and Th9 cells may be important, particularly because of their association with the accumulation and degranulation of mast cells. One study of patients with bronchial asthma showed that these patients had significantly elevated numbers of IL-17⁺ IL-4⁺ T cells in comparison with healthy controls; however, there was no overt difference in the total percentages of CD4 T cells expressing IL-4, IL-17 or IFN-γ. To further study and characterize this effect they generated T-cell clones that they subdivided into the following groups: Th1, Th2, Th17 and ‘Th17/2’ (IL-17⁺ IL-4⁺ clones), and using enzyme-linked immunosorbent assay found that the ‘Th17/2’ clones also expressed IL-9, IL-5, IL-8, IL-13, IL-21 and IL-22.²⁹ This suggests that these cells may represent the same IL-17⁺ IL-9⁺ T cells that are also elevated in patients with type 1 diabetes. Further studies in this area, as well as in models of oral and systemic allergies to antigen, may be necessary to understand the heterogeneity of relevant antigen-specific T cells.

During some types of infections, both IL-9 and IL-17 may play a role in modulating immune responses. A study of H1N1 influenza compared the expression of multiple cytokines in the sera of healthy individuals, of those with mild disease, of hospitalized non-critical patients and of hospitalized critical patients and found elevated levels of IL-9, IL-17, tumour necrosis factor-α (TNF-α), IFN-γ, IL-10 and IL-13 in both groups of hospitalized patients in comparison with controls.³⁰ Further work would be needed to analyse the source of these cytokines and evaluate if they are elevated because they are beneficial in clearing infection or if they are actually contributing to disease pathology. However, one study of IL-17 suggests that IL-17 from CD4 and CD8 T cells is beneficial in clearing influenza infection in mice because its blockade resulted in increased weight loss and decreased survival.³¹

Studies of infection by Pseudomonas, a Gram-negative bacterium that is associated with disease in both hospitalized and immune-compromised patients, also suggest that both IL-9 and IL-17 are important for survival. However, as in the influenza studies, it is unclear if the production of these cytokines is from Th17 cells or a combination of many other cell types. After vaccination and pulmonary challenge, it was observed that IL-17 deficiency resulted in decreased survival of mice.³² This and similar studies looking at this bacteria in the context of pulmonary challenge observed decreased neutrophil recruitment with low or absent IL-17 levels.^32–34 Separate studies examining IL-9 production in the context of pulmonary challenge and septic shock with this bacterium suggest that IL-9 is also important for the survival of these mice. This appeared to be through an effect on alveolar macrophages, monocytes and dendritic cells. More specifically IL-9 decreased the oxidative burst of antigen-presenting cells after challenge as well as decreased their production of TNF-α, IL-12 and IFN-γ, resulting in an attenuation of the Th1 response in these mice.^35–37 Further work is needed to study both IL-9 and IL-17 in this context to see if IL-9 is influencing the ratio of Th1 and Th17 cells as well as the activity of CD8 T cells present.

IL-9 production by mast cells

Although new interest has arisen in the production of IL-9 by CD4 T cells, it should not be overlooked that IL-9 production can also occur in mast cells. In vitro this can occur in part through stimuli such as histamine, IL-1, antigen-specific immunoglobulin E and antigen, and IL-9 itself and can help to facilitate their growth and expansion.^38,39 Further work would be needed to demonstrate the extent to which mast-cell-derived IL-9 can influence immune responses in comparison to that from various T-cell subsets.

Targets of IL-9

There are multiple cell types responsive to IL-9 including mast cells, T cells, antigen presenting cells, and epithelial cells of the lung and the gut. In mast cells, IL-9 appears to promote their recruitment and expansion. This is illustrated by the observation that IL-9- and IL-9R-deficient mice fail to accumulate and expand mast cells during disease models such as EAE and pulmonary granuloma induced by Schistosoma mansoni eggs.^11,12 Interleukin-9 also appears to directly cause the production of TGF-β in antigen-presenting cells. This leads to a decrease in lipopolysaccharide-induced oxidative burst and to TNF-α release.^36,37 In addition to indirect effects on T cells through antigen-presenting cells, IL-9 appears to increase the suppressive effect of Treg cells as well as enhance the proliferation and/or accumulation of Th17 cells.²⁴ Interleukin-9 can also have downstream effects on epithelial cells of the airway and intestinal mucosa to regulate these barriers, and its over-expression at these sites leads to airway hyper-responsiveness and increased intestinal permeability, respectively.^10,40

Concluding remarks

Overall, IL-9 production seems to occur in several T-cell subsets in vivo including Th17 cells. Future studies of IL-9 and Th17 cells will be needed to delineate the extent to which this phenotype is maintained over time. It would also be of interest if, like Th2 cells, Th17 cells can become Th9 cells under the right conditions and if the converse can also occur. In vivo studies must also be performed using multiple models of autoimmunity, allergy and infection to determine the extent to which IL-9 production occurs in these T-cell subsets and the extent to which these individual T-cell subsets contribute to immune responses.

Glossary

Abbreviations:

EAE

experimental autoimmune encephalomyelitis

IFN

interferon

IL

interleukin

TGF-β

transforming growth factor-β

Th

helper T

TNF

tumour necrosis factor

Treg

regulatory T

Disclosures

The authors have no conflicts of interest to disclose.