In this issue of The Journal, investigators from the University of Oviedo, Spain, report novel insights on the function of antimalarials and expand their data on the influence of polymorphisms regulating tumor necrosis factor (TNF) and interleukin 10 (IL-10) expression¹. The authors show that among patients with systemic lupus erythematosus (SLE) treated with antimalarials, those who express the genotype associated with low TNF and high IL-10 levels had significantly higher serum interferon-α (IFN-α) versus those with other genotypes, while those with the genotype associated with high TNF and low IL-10 levels had increased regulatory T cells. TNF and IL-10 promoter polymorphisms did not differentiate patients based on levels of IFN-α or regulatory T cells among those not treated with antimalarials. These data stimulate consideration of the mechanisms by which chloroquine and hydroxychloroquine modulate cytokine expression in SLE, and suggest approaches that may eventually allow prediction of who will respond to antimalarials, a mainstay of SLE therapy.

In addition, however, this report may provide one pillar of a bridge to span a perceived divide in lupus cytokine studies: the one between IFN-α and TNF. Both cytokines play essential roles in host defense against microbes, but their roles in autoimmune and inflammatory diseases are complex and not yet understood. One proposal addressing this complexity suggested that immunity can be seen as a dynamic system driven by opposite vectors, with SLE developing when IFN-α dominates and rheumatoid arthritis developing when TNF is the dominant cytokine in a pathologic process². In reality, however, there is solid evidence that both IFN-α and TNF are overexpressed in many patients with SLE, and that, indeed, both may be highly relevant in the disease process. Since our groups have worked on either side of the divide, we appreciate the opportunity to discuss the mechanisms that account for contributions of both IFN-a and TNF and to reframe SLE as a disease that calls for a more nuanced view that incorporates complex regulatory circuits and potential roles for both cytokines as potential therapeutic targets. It is clear that this bridge cannot be completed with today's limited understanding of these circuits. However, the data presented by Patricia Lopez and colleagues may help us to sketch its structure.

Although somewhat hampered by a lack of simple detection methods of sufficient specificity, and thus based on labor intensive biological assays, IFN-α serum levels have long been known to be increased in SLE, and to correlate with lupus disease activity^3-5. More recently, microarray and real-time polymerase chain reaction analysis of SLE peripheral blood cells showed increased expression of many IFN-inducible genes, demonstrating a "signature" consistent with in vivo stimulation by IFN-α, again in association with disease activity^6-8. The source of IFN-α in SLE has been elucidated in that plasmacytoid dendritic cells of SLE patients were found to produce increased amounts of IFN-α⁹, and increased IFN-α expression in the skin has been demonstrated¹⁰. IFN-α is known to have diverse effects on immune function, and many of those activities, including maturation of dendritic cells and induction of proinflammatory chemokines, are consistent with observations of altered immune function in patients with SLE^11,12.

Further support for the hypothesis that IFN-α plays a pathophysiologic role in SLE comes from observations that recombinant human IFN-α, when administered for hepatitis or other diseases, can induce SLE flares and drug-induced lupus¹³. Finally, preliminary data from a first clinical trial of a monoclonal anti-IFN-α antibody in SLE showed ablation of the IFN signature and suggested clinical improvement in at least some study subjects¹⁴.

TNF is similarly highly increased in SLE and correlates well with SLE disease activity^15,16. Although soluble TNF receptors are also significantly increased in patient serum, TNF bioactivity is not inhibited by availability of those receptors¹⁷. TNF is also overexpressed in active lupus nephritis, where again it correlates with disease activity¹⁶, and has been found in lupus skin lesions¹⁸. Monocytes/macrophages are presumed to be the major cellular source of TNF in lupus and can produce TNF when stimulated with immune complexes¹⁶. Indeed, immune complexes are important stimuli for both IFN-α and TNF. In the case of IFN-α, immune complexes containing nucleic acids can access intracellular Toll-like receptors after binding to Fc receptors and induce IFN-α. Fc receptors are also likely to contribute to TNF production induced by lupus immune complexes.

Consideration of the relative roles of IFN-α and TNF in the pathogenesis of SLE has been complicated by the occasional development of lupus autoantibodies when those cytokine pathways are manipulated. Lupus autoantibodies and occasionally clinical disease have developed in patients who receive therapeutic recombinant IFN-α, demonstrating the capacity of that cytokine to promote autoimmunity in a susceptible host¹⁹. In the case of TNF, however, it is the blockade of this cytokine that is associated with induction of antichromatin (and antiphospholipid) autoantibodies and the occasional induction of transient drug-induced lupus-like syndromes²⁰. The immunologic basis of these clinical observations has not been fully elucidated. The data regarding IFN-α are generally felt to be consistent with the concept that the cytokine promotes autoimmunity through multiple mechanisms, while several hypotheses have been raised to explain the capacity of TNF blockade to promote autoantibody production. These include induction of apoptosis, particularly by infliximab, as well as modulation of other immunoregulatory properties of TNF¹⁶. The idea that loss of an inhibitory effect of TNF on IFN-α production might contribute to the production of lupus autoantibodies has also been considered, based on studies demonstrating the suppression of IFN-α production by TNF²¹. However, while serum IFN-α levels can increase under some conditions of TNF blockade²², there is no experimental support for the concept that those increases account for autoantibody production.

While the hypothesis that there are IFN-α-driven diseases such as SLE and TNF-driven diseases such as RA¹¹ presents a model that serves to simplify understanding of rheumatic diseases, a consideration of the stages of disease pathogenesis as well as levels of disease activity suggests a more complex picture. We propose a role for both these cytokines in the pathogenesis of SLE as the scope of the immune complex triggers of cytokine induction broaden and the extent of inflammation and tissue damage grow over time (Figure 1). The first clinical studies with the combination of azathioprine with the anti-TNF antibody infliximab suggested potential benefits on organ inflammation, whereas no flares were observed, despite a transient increase in antichromatin and antiphospholipid autoantibodies^23,24.

[click, then close, image]

Figure 1. Overview of the role of TNF and IFN-α and their interplay in SLE.

What does the new report add to this complex picture? On the one hand, the authors have investigated both IFN-α and TNF serum levels of SLE patients, a combination that had so far not been published. Although one could criticize the assays used, their results are probably correct, since the increased levels of IFN-α and TNF are consistent with the previous literature. In contrast to what some may have expected, serum levels of IFN-α and TNF in both SLE patients and healthy individuals showed a weak positive correlation (r = 0.275). Additional new data measuring IFN-α functional activity and TNF by ELISA in SLE define patterns of cytokine expression based on a genetic variant in PTPN22, a phosphatase that regulates lymphocyte activation²⁵. Patients with the SLE-risk allele tend to show either a correlation of IFN-α and TNF values or an IFN-α high pattern. Unfortunately, neither the study of Lopez, et al nor the new study of PTPN22 variants provides data on the disease activity or organ involvement of the SLE patients. Preliminary data from one of our groups indicate that SLE patients can be characterized based on peripheral blood gene expression profiles into those expressing the IFN signature, those expressing activation of an inflammation pathway (including TNF, IL-1, and IL-8 mRNA), and those with both or neither pathway activated (Kirou KA, Crow MK, unpublished observations). Consistent with a concomitant role for both IFN-α and TNF in lupus nephritis, those patients with gene expression profiles showing activation of both pathways had a higher prevalence of renal disease than those with either pathway activated. It is tempting to speculate that another novel finding, that TNF induces low level IFN-b production in monocytes/macrophages²⁶, may play a role in the dual activation of these cytokine pathways, since both of these type I interferons use the same receptor.

The other novel piece of evidence lies in the discrepancy between patients genetically prone to produce high levels of TNF (TNFhi) and those not carrying such TNFhi alleles, but producing high levels of IL-10 (TNFlo IL-10hi)¹. Serum IFN-α levels of the latter, TNFlo IL-10hi patients, were almost twice as high taking antimalarials versus non-users of these drugs. In contrast, this was not true for the TNFhi patient population, where the IFN-α levels of those taking antimalarials were no different than those taking other agents.

To our eyes, these findings look like a threshold effect of TNF on IFN-α (Figure 2). Under conditions in which both cytokines are present at low levels, each would help to keep the other in check. But in the setting of immune system activation, most likely triggered by immune complexes in SLE, but potentially as well by activation of Toll-like receptors by pathogens, this reciprocal downmodulation is overridden, and production of both cytokines is increased. If TNF is reduced below a certain threshold, however, as effected by TNF blockade²², by antimalarials in TNFlo individuals¹, or when TNF is genetically absent in lupus-prone New Zealand White x New Zealand Black mice²⁷, IFN-α production is unchecked and therefore increases, with the potential to influence mechanisms of tolerance and autoimmunity.

[click, then close, image]

Figure 2. Putative relationship between TNF (x-axis) and IFN-α (y-axis) serum levels in SLE. Vertical stripes denote pathologically increased TNF, horizontal stripes increased IFN-α.

In line with the data presented by Lopez, et al¹, evidence is accumulating that genetic factors are significantly influencing the pathophysiology of SLE, probably including the involved cytokine patterns. As noted above, a PTPN22 variant associated with SLE is associated with higher IFN-a activity, and patients genetically prone to produce more TNF are at an increased risk of developing the disease. It will take a truly comprehensive overview to explain the interplay between various genetic background motifs, environmental factors, and immunoregulatory circuits that result in distinct patterns of cytokine production, organ involvement, and levels of disease activity. While the picture describing these complex scenarios is not yet clear, we may already have most of the puzzle pieces we will need to bring clarity to the role of cytokines in SLE.

For the moment, it appears that in well controlled lupus and in health, both IFN-α and TNF will control each other. On the other hand, when TNF is diminished below a certain threshold, such as by TNF blockers²⁴ or by antimalarials in TNFlo SLE patients¹, IFN-α production may be unleashed, increasing the propensity to develop autoantibodies. Once those autoantibodies are available, augmented innate immune system activation by autoantibody-containing immune complexes can perturb cytokine networks to favor IFN-α, TNF, or both. Immune complexes can induce TNF production, particularly in individuals carrying TNFhi alleles. Similarly, nucleic acid-containing immune complexes, including DNA-anti-dsDNA and RNA-anti-RNA-binding protein immune complexes, will induce plasmacytoid dendritic cells and probably other cell types to produce IFN-α. Accordingly, as autoimmunity broadens in a lupus patient, the level of cytokines will increase, along with SLE disease activity, in line with the finding of Lopez, et al that serum levels of TNF and IFN-α correlate¹. In addition, TNF may directly stimulate the interferon pathway by inducing IFN-b ²⁶. New understanding of the interplay of multiple gene variants, together with improved knowledge of the distinct events involved in initiation of autoimmunity, amplification of immune dysregulation, and activation of effector mechanisms that implement target organ damage will be needed to complete the full picture.

REFERENCES

Search PubMed for:

2. Banchereau J, Pascual V, Palucka AK. Autoimmunity through cytokine-induced dendritic cell activation. Immunity 2004; 20:539-50. [MEDLINE]

3. Hooks JJ, Moutsopoulos HM, Geis SA, Stahl NI, Decker JL, Notkins AL. Immune interferon in the circulation of patients with autoimmune disease. N Engl J Med 1979;301:5-8.

4. Ytterberg SR, Schnitzer TJ. Serum interferon levels in patients with systemic lupus erythematosus. Arthritis Rheum 1982;25:401-6. [MEDLINE]

5. Dall'era MC, Cardarelli PM, Preston BT, Witte A, Davis JC. Type I interferon correlates with clinical and serological manifestations of systemic lupus erythematosus. Ann Rheum Dis 2005;64:1692-7. [MEDLINE]

6. Bennett L, Palucka AK, Arce E, et al. Interferon and granulopoiesis signatures in systemic lupus erythematosus blood. J Exp Med 2003;197:711-23. [MEDLINE]

7. Kirou KA, Lee C, George S, et al. Coordinate overexpression of interferon-alpha-induced genes in systemic lupus erythematosus. Arthritis Rheum 2004;50:3958-67. [MEDLINE]

8. Baechler EC, Batliwalla FM, Karypis G, et al. Interferon-inducible gene expression signature in peripheral blood cells of patients with severe lupus. Proc Natl Acad Sci USA 2003;100:2610-5. [MEDLINE]

9. Blanco P, Palucka AK, Gill M, Pascual V, Banchereau J. Induction of dendritic cell differentiation by IFN-alpha in systemic lupus erythematosus. Science 2001;294:1540-3. [MEDLINE]

10. Reefman E, Kuiper H, Limburg PC, Kallenberg CG, Bijl M. Type I interferons are involved in the development of UVB-induced inflammatory skin lesions in systemic lupus erythematosus (SLE) patients. Ann Rheum Dis 2008;67:11-8. [MEDLINE]

11. Banchereau J, Pascual V. Type I interferon in systemic lupus erythematosus and other autoimmune diseases. Immunity 2006;25:383-92. [MEDLINE]

12. Crow MK. Type I interferon in systemic lupus erythematosus. Curr Top Microbiol Immunol 2007;316:359-86. [MEDLINE]

13. Ioannou Y, Isenberg DA. Current evidence for the induction of autoimmune rheumatic manifestations by cytokine therapy. Arthritis Rheum 2000;43:1431-42. [MEDLINE]

14. Wallace DJ, Petri M, Olsen N, et al. MEDI-545, an anti-interferon alpha monoclonal antibody, shows evidence of clinical activity in systemic lupus erythematosus [abstract]. Arthritis Rheum 2007;56:S526-S527.

15. Studnicka-Benke A, Steiner G, Petera P, Smolen JS. Tumour necrosis factor alpha and its soluble receptors parallel clinical disease and autoimmune activity in systemic lupus erythematosus. Br J Rheumatol 1996;35:1067-74. [MEDLINE]

16. Aringer M, Smolen JS. The role of tumor necrosis factor-alpha in systemic lupus erythematosus. Arthritis Res Ther 2008;10:202. [MEDLINE]

17. Aringer M, Feierl E, Steiner G, et al. Increased bioactive TNF in human systemic lupus erythematosus: associations with cell death. Lupus 2002;11:102-8. [MEDLINE]

18. Zampieri S, Alaibac M, Iaccarino L, et al. Tumour necrosis factor alpha is expressed in refractory skin lesions from patients with subacute cutaneous lupus erythematosus. Ann Rheum Dis 2006;65:545-8. [MEDLINE]

19. Gota C, Calabrese L. Induction of clinical autoimmune disease by therapeutic interferon-alpha. Autoimmunity 2003;36:511-8. [MEDLINE]

20. Charles PJ, Smeenk RJ, De Jong J, Feldmann M, Maini RN. Assessment of antibodies to double-stranded DNA induced in rheumatoid arthritis patients following treatment with infliximab, a monoclonal antibody to tumor necrosis factor alpha: findings in open-label and randomized placebo-controlled trials. Arthritis Rheum 2000;43:2383-90. [MEDLINE]

21. Palucka AK, Blanck JP, Bennett L, Pascual V, Banchereau J. Cross-regulation of TNF and IFN-alpha in autoimmune diseases. Proc Natl Acad Sci USA 2005;102:3372-7. [MEDLINE]

22. Mavragani CP, Niewold TB, Moutsopoulos NM, Pillemer SR, Wahl SM, Crow MK. Augmented interferon-alpha pathway activation in patients with Sjogren's syndrome treated with etanercept. Arthritis Rheum 2007;56:3995-4004. [MEDLINE]

23. Aringer M, Graninger WB, Steiner G, Smolen JS. Safety and efficacy of TNF blockade in systemic lupus erythematosus — an open label study. Arthritis Rheum 2004;50:3161-9. [MEDLINE]

24. Aringer M, Steiner G, Graninger WB, Hofler E, Steiner CW, Smolen JS. Effects of short-term infliximab therapy on autoantibodies in systemic lupus erythematosus. Arthritis Rheum 2007;56:274-9. [MEDLINE]

25. Kariuki SN, Crow MK, Niewold TB. PTPN22 C1858T polymorphism is associated with skewing of cytokine profiles toward high IFN-activity and low TNF- in lupus patients. Arthritis Rheum 2008; In press.

26. Yarilina A, Park-Min KH, Antoniv T, Hu X, Ivashkiv LB. TNF activates an IRF1-dependent autocrine loop leading to sustained expression of chemokines and STAT1-dependent type I interferon-response genes. Nat Immunol 2008;9:378-87. [MEDLINE]

27. Jacob CO, McDevitt HO. Tumour necrosis factor-alpha in murine autoimmune 'lupus' nephritis. Nature 1988;331:356-8. [MEDLINE]