Type I and type III interferon-induced immune response: It's a matter of kinetics and magnitude

Authors

  • David Olagnier Ph.D.,

    1. Viral Pathogenesis Program, Vaccine and Gene Therapy Institute of Florida, Port St. Lucie, FL
    Search for more papers by this author
  • John Hiscott Ph.D.

    Corresponding author
    1. Viral Pathogenesis Program, Vaccine and Gene Therapy Institute of Florida, Port St. Lucie, FL
    • Address reprint requests to: John Hiscott, Ph.D., Viral Pathogenesis Program, Vaccine and Gene Therapy Institute of Florida, 9801 SW Discovery Way, Port St. Lucie, FL 34987. E-mail: jhiscott@vgtifl.org; fax: 772-345-0625.

    Search for more papers by this author

  • See Articles on Pages 1250 and 1262

  • Potential conflict of interest: Nothing to report.

Abbreviations
GAS

gamma-activated sequences

HCV

hepatitis C virus

IFN

interferon

IFNAR

IFN-α receptor

IFNLR1

interferon λ receptor 1

IL

interleukin

IRF9

IFN regulatory factor 9

ISG

IFN-stimulated gene

ISGF3

IFN-stimulated gene factor 3

ISRE

IFN-stimulated response element

JAK

Janus kinase

PHHs

primary human hepatocytes

STAT

signal transducer and activator of transcription

TYK

tyrosine kinase

The interferon (IFN) family of cytokines, including type I and III IFNs, represent a first line of defense that “interferes” with a vast range of viral infections.[1] IFN cytokines are commonly classified into distinct families, based on their structural properties and on the receptor complex through which they signal.[1] Type I IFNs, including IFN-β and -α, exert their biological activities through a heterodimeric receptor complex known as IFN-α receptor (IFNAR) that consists of IFNAR1 and IFNAR2 chains. Discovered in 2003, type III IFNs, including interleukin (IL)−28A (IFN-λ2), IL-28B (IFN-λ3), and IL-29 (IFN-λ1), are distinct from type I IFNs and act through a different receptor system that is preferentially expressed on the surface of epithelial cells, including hepatocytes.[2-4] Type III IFNs bind to a heterodimeric receptor consisting of the IL-10 receptor 2 chain and the interferon λ receptor 1 (IFNLR1) chain that is structurally different from the IFNAR. Recently, a new coding region upstream of IFNL3 on chromosome 19q13.13, designated as IFNL4, was identified that encodes a new member of the type III IFN family, called IFN-λ4.[5]

Signaling cascades of both type I and III IFNs converge on the activation of Janus kinase (JAK)1 and tyrosine kinase (TYK)2 kinases, leading to the subsequent phosphorylation and activation of latent signal transducer and activator of transcription (STAT) transcription factors. Phosphorylated STAT1 and STAT2 heterodimerize and interact with IFN regulatory factor 9 (IRF9) to form an IFN-stimulated gene factor 3 (ISGF3) transcription complex that binds to the IFN-stimulated response element (ISRE); STAT1 can also homodimerize and bind the gamma-activated sequences (GAS) element. Altogether, binding of these STAT complexes to specific promoter sequences leads to the establishment of an antiviral response through up-regulation of hundreds of IFN-stimulated genes (ISGs; Fig. 1), many of which are poorly understood biologically with regard to their antiviral activity.[6] Despite what appear to be common signaling cascades, type I and III IFNs have begun to reveal divergent biological activities.[7-10] However, it is still unclear whether these related cytokine family members induce common or distinct patterns of ISGs that result in the programming of divergent antiviral and/or immune modulatory states. In this issue of Hepatology, Bolen et al. and Jilg et al. have approached this question using time-resolved, microarray-based gene expression profiling in human hepatocytes. Both studies reach a similar, but somewhat unpredicted, conclusion that type I and III IFNs stimulate similar sets of ISGs, albeit with very distinct kinetics of response and magnitude of stimulation.[11, 12]

Figure 1.

Type I and III IFN-signaling pathways lead to different kinetics and strengths of stimulation of ISG response. Both type I and III IFNs induce STAT phosphorylation through the JAK-TYK kinases associated with the respective receptor subunits. Receptor engagement results in activation of STAT1 and STAT2, which complex with IRF9 to form the transcription factor, ISGF3. STAT1 can also homodimerize after phosphorylation. These complexes induce expression of genes with ISRE or GAS in their promoters, leading to different kinetics and magnitude of ISG stimulation. The graph below indicates the relative abundance, kinetics, and magnitude of the resulting ISG response to type I and III IFN.

These two parallel studies from Bolen et al. and Jilg et al. converge on an answer that contrasts with studies demonstrating that type I and III IFNs induce both overlapping and distinct sets of genes.[8, 13, 14] In the current experiments, transcriptome analysis from human hepatoma Huh7 cells or primary human hepatocytes (PHHs) stimulated with comparable concentrations of IFN-α, -β, -λ1, -λ2, or -λ3 revealed a hierarchy of gene expression, with IFN-β identified as the most biologically active cytokine, followed by IFN-α and IFN-λs (Fig. 1). However, the researchers did not discover any distinct sets of genes modulated by IFN-λs, and all IFN-λ-induced genes were also induced by type I IFN,[11, 12] as previously suggested.[15] Although the pattern of gene expression induced by either type I or III IFNs was very similar, the magnitude of the response triggered by IFN-β was more than 10-fold higher than the response promoted by IFN-λs. Bolen et al. also concluded that, among the IFN-λ family, IFN-λ3 was the most potent ISG inducer[11] and corroborates another study demonstrating that IFN-λ3 possesses a higher level of antiviral activity than either IFN-λ1 or IFN-λ2.[16]

The studies also intriguingly reported on a unique kinetic profile of ISG expression in IFN-α-treated Huh7 or PHHs.[11, 12] Indeed, both IFN-β and IFN-λ triggered a sustained ISG response; on the other hand, IFN-α produced a kinetic profile of genes that peaked at early times of treatment and then rapidly decreased (Fig. 1). In an attempt to investigate the molecular factors and the signaling elements contributing to the peculiar kinetics of ISG response in IFN-α-stimulated hepatocytes, Bolen et al. focused on the phosphorylation status of different residues within the ISG response-driven master regulator, STAT1. No difference was observed in STAT1 tyrosine phosphorylation between IFN-β and -α, but serine phosphorylation at position 727 was more strongly stimulated by IFN-β, compared to IFN-α. However, blocking protein kinase C delta—the predominant kinase involved in S727 phosphorylation—did not affect the expression of ISGs, thus suggesting that type I IFNs elicit a greater ISG response independently of serine 727 phosphorylation status. Interestingly, Jilg et al. finally reported that this early, transient up-regulation of ISGs observed after IFN-α treatment was altered by hepatitis C virus (HCV) infection.[12] Taken together with a previous study reporting that combinations of IL-29 with type I IFN induce greater antiviral activity against HCV than individual cytokines alone,[17] the observation that HCV dampens early IFN-α-induced ISG response reveals an obvious complementary role for IFN-α and IL-28B in mounting an appropriate antiviral response against this flavivirus in hepatocytes.

The observation that the difference between type I and III IFNs is more quantitative than qualitative is made all the more puzzling because biological divergence between these families of cytokines was previously observed both in vitro and in vivo.[7, 10] In one study, IFN-λ, in contrast to IFN-α, was shown to prevent herpes simplex virus 2 infection at the vaginal mucosa.[7] The specific restriction of type III IFN receptor expression to epithelia may explain some of the divergent antiviral effects observed between type I and III IFNs. This observation also suggests that IFN-λs may exert biological efficacy through stimulation of the immune system, rather than through induction of an antiviral state in vivo. Taken together, these observations indicate that, even though the gene expression profiles induced by type I and III IFNs are the same, the biological functions associated with these cytokines differ in vivo and support the notion that IFN-λ plays a local, rather than systemic, role in antiviral immunity.

In both studies, the researchers demonstrate that the main difference observed in vitro between type I and III IFNs is purely quantitative. Part of this observation may be explained by a significant difference in relative expression levels of these receptors on the hepatocyte cell membrane, an aspect the researchers did not take into account in their analysis. Finally, type I IFNs trigger a stronger ISG response than IFN-λs in hepatocytes; however, other cell types coexpress both type I and III IFN receptors, and it will be important to determine whether the difference in magnitude of ISG stimulation (and the absence of divergent ISG response) is reproduced in other tissue- or cell-type–specific contexts. The difference in the relative strength of IFN-λ family members may be important when considering the development of new IFN therapies for HCV infection or for cancer. Indeed, the standard HCV therapeutic protocol using pegylated IFN-α is expensive, poorly tolerated, and only partially effective and highlights the need for alternative options. The observed poor clinical efficacy of IFN-α is consistent with studies in hepatocytes demonstrating that IFN-α generates a strong, but transient, ISG response, sufficient to generate adverse effects without necessarily providing a long-lasting protective immune response. In contrast, IFN-λs provide a less-powerful, but more-sustained ISG response in vitro, which would better fit with the expectation required for a good therapeutic candidate. Recombinant IFN-λ1 is currently being tested in its pegylated form in a phase III clinical trial in comparison with IFN-α. The present observation from Bolen et al. demonstrates that IFN-λ3 is a stronger inducer of gene expression than IFN-λ1 in vitro. Combined with the recent genome-wide association studies that identified an IL28B genetic polymorphism as a major predictor of HCV spontaneous clearance,[18-20] these studies further emphasize the potential use of the IFN-λ family, particularly IFN-λ3, in the clinic.

Overall, these two studies reveal functional differences between type I and III IFNs and show the distinct strengths and kinetics of ISG induction in hepatocytes. Given that the set of genes induced by these different cytokines is overlapping, one may ask why evolution maintains two apparently redundant antiviral systems. One answer is that IFN-λ stimulation results in an IFN response that differs both spatially and temporally, from that of type I IFN, activating a more defined subset of cells. This possibility highlights the importance of determining the distribution of the type III IFN receptor complex in vivo to better understand its biological functions. Another consideration is that both type I and III IFNs rely on mitogen-activated protein kinases to recruit auxiliary transcription factors that cooperate with the ISGF3 complex to mediate ISG transcription. The weaker activation of the ISG response observed with IFN-λs could be dependent upon these auxiliary transcription factors, and further studies will be required to determine the molecular basis of the differential ISG response generated by type I and III IFNs. A final—more philosophical—answer to the above question is that the IFN systems are not redundant at all, and we are still simply observing this particular aspect of nature, without fully understanding it.

  • David Olagnier, Ph.D.

  • John Hiscott, Ph.D.

  • Viral Pathogenesis Program

  • Vaccine and Gene Therapy Institute of Florida

  • Port St. Lucie, FL

Ancillary