In this issue of Blood, Guerra et al1 investigate the role of tumor necrosis factor alpha (TNF-α) and its cross talk with interferon gamma (IFN-γ) in regulating and restoring steady-state hematopoiesis in a murine model of anemia of inflammation.
Anemia of inflammation (AI) is the second most prevalent form of anemia worldwide, following iron deficiency anemia. It is a hallmark of various conditions, including infections, autoimmune diseases, and malignancies.
AI is characterized by the following:
Impaired iron utilization, primarily due to inflammation-induced upregulation of the iron-regulatory hormone hepcidin, which leads to low serum iron levels despite adequate or increased total body iron stores.
Reduced red blood cell (RBC) production, resulting from impaired erythropoietin synthesis, inhibition of erythroid precursor maturation, and a shift in progenitor cell commitment toward the myeloid rather than erythroid lineage, all processes driven by elevated levels of proinflammatory cytokines.
Shortened RBC life span due to increased erythrophagocytosis by macrophages.2,3
Several inflammatory cytokines have been implicated in the pathophysiology of AI. Although the role of interleukin-6 (IL-6) is well established, being the primary cytokine responsible for hepcidin upregulation,3 the contributions of other proinflammatory cytokines, such as TNF-α and IFN-γ, which are also overproduced in AI, remain less clearly defined.
To investigate this, the authors explored the role of TNF-α and IFN-γ in modulating erythropoiesis during both acute and chronic inflammation, using a well-established murine model of AI based on heat-killed Brucella abortus (HKBA) injection.4 They first examined how AI develops in the absence of TNF-α. Interestingly, although wild-type mice recovered from anemia within 4 weeks (see figure panel A), Tnf-α knockout (KO) mice failed to restore RBC counts and hemoglobin levels, exhibited persistently elevated white blood cell counts, and died within 10 weeks (see figure panel B).
TNF-α and IFN-γ cross talk dictates inflammation outcome in AI. (A) In wild-type mice, HKBA injection triggers an acute upregulation of TNF-α and IFN-γ, which resolves over time. Anemia typically recovers within 4 to 8 weeks. (B) In Tnf-α KO mice, HKBA injection causes a sustained increase in IFN-γ levels, leading to a cascade of detrimental effects including severe anemia, bone marrow failure, and ultimately death. (C) Combined treatment with rTNF-α and aIFN-γ fully rescues the lethality induced by HKBA injection in Tnf-α KO mice. aIFN-γ, anti–IFN-γ antibody. The figure was prepared with the help of Francesca Carlomagno (Federico II University, Naples, Italy) using BioRender.com.
TNF-α and IFN-γ cross talk dictates inflammation outcome in AI. (A) In wild-type mice, HKBA injection triggers an acute upregulation of TNF-α and IFN-γ, which resolves over time. Anemia typically recovers within 4 to 8 weeks. (B) In Tnf-α KO mice, HKBA injection causes a sustained increase in IFN-γ levels, leading to a cascade of detrimental effects including severe anemia, bone marrow failure, and ultimately death. (C) Combined treatment with rTNF-α and aIFN-γ fully rescues the lethality induced by HKBA injection in Tnf-α KO mice. aIFN-γ, anti–IFN-γ antibody. The figure was prepared with the help of Francesca Carlomagno (Federico II University, Naples, Italy) using BioRender.com.
Notably, the repression of erythropoiesis in TNF-α–deficient mice was not associated with altered hepcidin levels or iron dysregulation, nor with inadequate erythropoietin production. These findings suggest that alternative mechanisms may be contributing to the observed defects.
HKBA primarily elicits a T helper 1 (Th1)-type immune response.5 In wild-type mice, HKBA injection induces a transient upregulation (within 4-6 hours) of Th1 cytokines, including TNF-α, IL-12p40, and IFN-γ (see figure panel A), as well as the Th2 cytokine IL-6, likely as a downstream effect of elevated TNF-α and IFN-γ levels.6 This inflammatory response resolves by 4 weeks. In contrast, Tnf-α KO mice exhibit impaired IL-6 induction during the acute phase, but display persistent elevation of Th1-associated cytokines (IFN-γ and IL-12p40) in the long term, with IFN-γ abnormally elevated at 4 weeks (see figure panel B).
Persistent IFN-γ upregulation shifts the immune response from a protective, acute phase to a harmful, chronic inflammatory state.7 Interestingly, in HKBA-induced inflammation, IFN-γ has a negative effect on the survival of Tnf-α KO mice: its deletion prevents IL-12p40 upregulation during chronic infection and completely rescues the increased mortality observed in HKBA-treated Tnf-α KO mice. However, treatment with an anti–IFN-γ antibody during the late phase of inflammation is ineffective in preventing mortality in Tnf-α KO mice, suggesting that acute IFN-γ upregulation contributes to the HKBA-induced phenotype in this model.
The authors then further investigated the mechanisms underlying impaired survival in HKBA-challenged mice lacking the Th1 cytokine TNF-α. In addition to defective erythropoiesis, these mice exhibited elevated white blood cell and lymphocyte counts. Specifically, both CD4+ and CD8+ T cells were increased in the bone marrow and spleen of Tnf-α KO mice. However, only CD8+ T cells showed increased IFN-γ production, suggesting that they are the primary source of IFN-γ overproduction in this setting.
Regarding B cells, the absence of TNF-α resulted in a persistent reduction of the B-cell population compared to wild-type controls. Macrophages were also affected, as shown by their increased numbers in the spleen of HKBA-treated Tnf-α KO mice.
This persistent state of chronic inflammation and sustained IFN-γ production triggers a cascade of detrimental effects, including reduced RBC life span and increased clearance, hematopoietic stem cell exhaustion, bone marrow failure, and ultimately death.
Tnf-α KO mice challenged with inflammatory stimuli exhibit a complex phenotype characterized by both impaired early inflammatory responses and defective resolution and repair mechanisms, resulting in prolonged and dysregulated inflammation. This reflects a failure of the anti-inflammatory and tissue-repair functions normally mediated by TNF-α during the resolution phase of inflammation.8
In the HKBA model, IFN-γ plays a central role in impairing the resolution of inflammation. But what drives its persistent upregulation in the absence of TNF-α?
TNF-α normally constrains IFN-γ activity through indirect pathways, such as induction of anti-inflammatory cytokines (eg, IL-10) and suppression of STAT1 signaling. In TNF-α–deficient conditions, these regulatory mechanisms are lost, resulting in unchecked IFN-γ production. Furthermore, TNF-α negatively regulates IFN-γ–dependent inflammation by limiting the duration of IL-12 production by macrophages, a proinflammatory cytokine that enhances T-cell responses, further supporting its role in tempering IFN-γ–driven inflammatory cascades.9
Excessive levels of both TNF-α and IFN-γ can be detrimental.10 Treatment of Tnf-α KO mice with recombinant TNF-α (rTNF-α) at 6 to 8 weeks after HKBA challenge, when IFN-γ levels are already elevated, results in rapid mortality within 1 to 2 days. The failure of rTNF-α to improve survival suggests that TNF-α may only be beneficial when IFN-γ is also downregulated. Consistent with this hypothesis, Tnf-α KO mice can only be rescued from HKBA-induced lethality when rTNF-α is administered in combination with an anti–IFN-γ antibody (see figure panel C).
Although the precise mechanism underlying this accelerated death is not fully understood, it is possible that the combined overexpression of TNF-α and IFN-γ sensitizes cells to PANoptosis, a form of inflammatory cell death involving apoptosis, necroptosis, and pyroptosis, implicated in severe immune conditions such as cytokine storm syndrome and severe acute respiratory syndrome coronavirus 2 infection.10
Through the characterization of TNF-α and its functional interplay with IFN-γ, Guerra et al advance our understanding of the role these cytokines play in the pathogenesis of AI. Their findings highlight the critical importance of a balanced TNF-α/IFN-γ axis in controlling inflammation. Moreover, these insights could clarify certain unexplained hematologic side effects, mainly infections and immune-related complications, observed in patients treated with TNF-α inhibitors, which are widely used for autoimmune and inflammatory diseases such as rheumatoid arthritis, ankylosing spondylitis, psoriasis, inflammatory bowel disease, and juvenile idiopathic arthritis.
Conflict-of-interest disclosure: L.S. declares no competing financial interests.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal