Myeloid-Derived Suppressor Cells in Sepsis: Pathophysiology, Duality, and Therapeutic Frontiers.

INTRODUCTION

Sepsis remains a formidable challenge in intensive care units worldwide, where its high incidence and mortality impose a substantial burden on public health systems ( 1 , 2 ). Historically, clinical interventions have concentrated on infection source control and mitigating the early, uncontrolled “cytokine storm” ( 3 , 4 ). However, extensive clinical evidence reveals that many sepsis patients, upon surviving the acute hyperinflammatory phase, do not progress towards recovery. Instead, they transition into a protracted state of persistent immune dysfunction ( 5 ). This state is clinically characterized by a significantly increased susceptibility to nosocomial secondary infections, reactivation of latent viruses (such as cytomegalovirus and herpes simplex virus), and ultimately, progressive organ dysfunction and heightened long-term mortality ( 6 , 7 ).

This paradigm shift has identified sepsis-associated immunosuppression as a critical, unresolved challenge in current research and treatment. Within this complex immunosuppressive network, a diverse array of regulatory cells and molecular pathways is activated, collectively compromising the host’s immune surveillance and pathogen clearance capabilities. Among these, myeloid-derived suppressor cells (MDSCs) have emerged as a pivotal cell population implicated in this process ( 8 ).

MDSCs are not a terminally differentiated population but rather a heterogeneous collection of immature myeloid cells (IMCs), encompassing myeloid progenitors and precursors of granulocytes, monocytes, and dendritic cells (DCs). While maintained at negligible levels in the bone marrow under healthy conditions, their generation, expansion, and activation are profoundly reprogrammed during pathological states such as sepsis, malignancy, and severe trauma ( 9 , 10 ). These peripherally mobilized MDSCs possess a potent repertoire of immunosuppressive mechanisms, enabling them to efficiently suppress key functions of both adaptive immunity (e.g., T cells) and innate immunity (e.g., natural killer [NK] cells) ( 11 , 12 ). Clinically, the abundance of circulating MDSCs in sepsis patients correlates significantly with disease severity, sequential organ failure scores, and 28-day mortality, highlighting them as a promising prognostic biomarker and therapeutic target ( 13 ).

Growing evidence suggests that the role of MDSCs in sepsis may be biphasic, although this concept is primarily derived from murine models and observational clinical data. In the early stages, preclinical studies indicate that a moderate expansion of MDSCs might confer a protective effect by dampening excessive inflammatory responses, thereby potentially limiting tissue damage. However, as the disease progresses, clinical observations correlate persistent MDSC expansion with detrimental outcomes. Several single-center cohort studies have linked this persistent immunosuppressive phenotype to an increased risk of secondary infections and mortality, posing a potential obstacle to restoring immune homeostasis ( 14 , 15 ). Therefore, an in-depth exploration of the dynamic changes of MDSCs during the course of sepsis, and the regulatory mechanisms controlling their functional switch, is of profound significance for elucidating the nature of sepsis-induced immunosuppression and developing precise immunomodulatory strategies.

BIOLOGICAL CHARACTERISTICS OF MDSCS

Cellular origin and differentiation trajectory

MDSCs represent a heterogeneous population of IMCs whose cardinal feature is a potent immunosuppressive function. Under pathological conditions, such as cancer, infection, and severe trauma, these cells undergo pathological expansion and activation ( 9 ). MDSCs originate from the same hematopoietic differentiation pathway as normal myeloid cells. In physiological homeostasis, hematopoietic stem cells differentiate through common myeloid progenitor and granulocyte-macrophage progenitor stages to generate mature, functional cells like neutrophils and monocytes ( 16 , 17 ). This process is stringently regulated to maintain immune equilibrium.

However, in pathological states like sepsis or malignancy, this normal myeloid differentiation is profoundly disrupted by persistent, potent inflammatory signals ( 18 ). Pathogen-associated molecular patterns, damage-associated molecular patterns, and associated cytokine storms (e.g., granulocyte-macrophage colony-stimulating factor [GM-CSF], IL-6, IL-1β, and VEGF) drive this disruption. These signals initiate a state of “emergency” or “pathological” myelopoiesis. By persistently activating key transcription factors, notably STAT3 and NF-κB, these stimuli exert a dual effect: they arrest the differentiation of IMCs into mature antigen-presenting cells like DCs or macrophages (e.g., by repressing critical differentiation factors like IRF8), while simultaneously inducing and reinforcing a potent immunosuppressive phenotype ( 19 , 20 ). This pathogenic process culminates in the massive accumulation of these dysfunctional immature cells within the bone marrow, spleen, peripheral blood, and inflammatory sites, collectively identified as MDSCs.

Classification and phenotypic characteristics

MDSCs are broadly categorized into two major subsets in both mice and humans: granulocytic or polymorphonuclear MDSCs (PMN-MDSCs) and monocytic MDSCs (M-MDSCs). A smaller population of early-stage MDSCs has also been identified in humans ( 21 ).

PMN-MDSCs: In humans, this subset is defined as CD11b + CD14 − (CD15 + or CD66b + ). A key distinguishing feature is their lower density compared with normal neutrophils, causing them to enrich in the 1.077 g/mL layer during density gradient separation. Lectin-like oxidized LDL receptor-1 (LOX-1) has been identified as a specific marker for this population ( 21 ). Morphologically, they display abnormal nuclear segmentation and loose chromatin. Their suppressive capacity is primarily driven by high expression of reactive oxygen species (ROS) and PGE 2 , which function by nitrating the T-cell receptor (TCR) and inhibiting T-cell migration ( 22 ). The corresponding murine PMN-MDSC population is phenotyped as CD11b + Ly6G + Ly6C lo . These cells, morphologically similar to neutrophils but with less segmented nuclei, exert suppression largely through high expression of ROS and arginase 1 (ARG1), leading to arginine depletion and ROS/peroxynitrite (ONOO − )-mediated inhibition of TCR signaling ( 23 ).

M-MDSCs: The human M-MDSC phenotype is CD11b + CD14 + HLA-DR lo/− , characterized by the lack of high HLA-DR expression seen on mature monocytes. They also highly express CXCR1 and CD84. Their immunosuppressive functions rely heavily on arginine metabolism and the secretion of inhibitory cytokines. In mice, M-MDSCs are identified as CD11b + Ly6G − Ly6C hi . These cells, which resemble inflammatory monocytes, primarily utilize high expression of inducible nitric oxide synthase (iNOS) (leading to NO production) and IL-10 to suppress T-cell signaling and promote Treg differentiation ( 24 , 25 ).

Early-stage MDSCs: This minor population, representing <5% of MDSCs in humans, is defined as Lin − HLA-DR − CD33 + . Comprising myeloid progenitors, these cells are reported to promote disease progression via the secretion of factors such as VEGF ( 26 , 27 ).

Shared biological hallmarks: Across subsets and species, MDSCs exhibit several common functional characteristics. They universally express high levels of S100A8/A9 proteins, which can amplify their immunosuppressive function by activating NF-κB ( 28 ). Furthermore, MDSCs undergo profound metabolic reprogramming, characterized by enhanced glycolysis and upregulated fatty acid oxidation (FAO); notably, FAO inhibitors can reverse their suppressive function ( 29 ). Finally, endoplasmic reticulum stress is a common feature, with high expression of CHOP and sXBP1 regulating cell survival and differentiation. CHOP deficiency, for instance, has been shown to reduce MDSC suppressive capacity ( 30 ). These collective characteristics form the biological basis for the potent immune regulation exerted by MDSCs in pathological states.

THE DOUBLE-EDGED SWORD: TEMPORAL DYNAMICS OF MDSCS IN SEPSIS

The role of MDSCs in sepsis is not static; it evolves dynamically with the progression of the syndrome. Their impact shifts from a potentially protective mechanism in the initial hours to a pathological driver of immunosuppression in the later stages ( 31 ) (Fig. 1 A). Key observational clinical studies investigating MDSCs in human sepsis are summarized in Table 1 .

The early acute phase

In the initial hyperinflammatory phase (“cytokine storm”), the rapid mobilization of MDSCs appears to serve a physiological purpose: to limit collateral tissue damage caused by excessive immune activation.

In multiple murine models of polymicrobial sepsis (CLP), the adoptive transfer of MDSCs during the early phase improved survival by dampening systemic levels of TNF-α, IL-6, and IFN-γ. Conversely, depletion of MDSCs (e.g., via anti-Gr-1) during this window exacerbated organ injury and increased mortality. This protective effect is mediated by the release of anti-inflammatory mediators (e.g., IL-10 and TGF-β) and the direct scavenging of ROS ( 31 ).

It is crucial to note that evidence for this “beneficial” phase is largely derived from controlled animal models. Human data is sparse, primarily because patients rarely present during this ultra-early window, making validation difficult.

The late/chronic phase

As sepsis progresses into a subacute or chronic phase, the persistent expansion of MDSCs becomes maladaptive, fueling the state of immunoparalysis ( 32 ). Clinical studies have shown that MDSC levels are markedly elevated in patients during the late phase of sepsis and are significantly associated with increased mortality ( 33 ). The failure of MDSC counts to return to baseline is a hallmark of chronic critical illness (CCI).

This sustained immunosuppression leaves patients vulnerable to opportunistic pathogens. High MDSC burdens have been directly correlated with the development of secondary sepsis ( 34 ). Depletion of MDSCs alleviates immunosuppression and reduces susceptibility to secondary infections in sepsis models ( 35 ).

In this phase, MDSCs actively suppress T-cell proliferation and cytokine production via ARG1 and iNOS, effectively crippling the adaptive immune response required to clear residual infection ( 36 ).

METABOLIC REPROGRAMMING AND CELLULAR INTERACTIONS

In the sepsis microenvironment, MDSCs undergo distinct metabolic changes that fuel their immunosuppressive functions.

Glucose metabolism

MDSC maturation relies primarily on glycolysis and glucose-derived carbon for anabolic metabolism and ATP production, with oxidative phosphorylation maintaining low activity mainly to regenerate NADPH ( 37 ). In vitro experiments demonstrated that MDSCs were able to differentiate into mature neutrophils in glucose-containing medium, whereas their differentiation was hindered in the absence of glucose or in the presence of the glycolysis inhibitor 2-deoxy-D-glucose ( 38 ).

Lipid and FAO

Although MDSCs primarily rely on glycolysis for energy production and biosynthetic precursors, they also possess the ability to utilize other metabolic pathways, such as fatty acid β-oxidation. This pathway has been demonstrated in research models like solid tumors, where fatty acids are broken down to generate energy. This metabolic plasticity suggests that under pathological conditions such as sepsis, MDSCs may adapt to different microenvironmental demands by activating FAO.

Amino acid metabolism

During sepsis, signaling axes such as PERK–CHOP and IRE1α–XBP1 play a pivotal role in upregulating canonical MDSC effector molecules—particularly ARG1 and iNOS—which underpin their immunosuppressive functions ( 39 ). MDSCs suppress T-cell activity by depleting L-arginine, an amino acid essential for T-cell proliferation and function ( 36 ). They markedly upregulate the expression of ARG1 and iNOS, which catalyze the breakdown of arginine into ornithine/urea and nitric oxide (NO), respectively. This localized nutrient deprivation directly inhibits T-cell proliferation and downregulates expression of the TCR ζ chain ( 36 , 40 ).

Redox balance and ROS

ROS serve as central regulators of the function and fate of MDSCs. They are involved not only in metabolic reprogramming, proliferation, and differentiation of MDSCs, but also act as key effector molecules mediating immune suppression. Through the regulation of crucial transcription factors such as Nrf2 and HIF-1α, ROS drives transcriptional and metabolic remodeling of MDSCs. Furthermore, MDSCs employ ROS to suppress immune responses via multiple pathways, including direct inhibition of T cells, NK cells, and B cells, as well as indirect modulation of antigen presentation and chemokine activity ( 41 ).

Inhibition of NK cells

The suppressive reach of MDSCs extends to the innate immune system, particularly NK cells. MDSCs inhibit the cytotoxic function of NK cells via both membrane-bound TGF-β and soluble factors like IL-10. In sepsis models, MDSCs have been shown to downregulate activating receptors (such as NKG2D) on NK cells, reducing their release of perforin and granzymes and thus weakening their ability to kill infected cells ( 42 ). Furthermore, NO released by MDSCs can directly inhibit intracellular signaling in NK cells and blunt their IFN-γ secretion ( 43 ).

The MDSC–Treg axis

Beyond direct suppression, MDSCs actively sculpt the immune microenvironment by promoting the differentiation and expansion of Foxp3 + Tregs. This is achieved through two routes:

Cell-to-cell contact: MDSCs highly express molecules such as glucocorticoid-induced leucine zipper protein, which, upon direct contact with T cells, induces the generation of Foxp3 + Tregs ( 44 , 45 ).

Cytokine secretion: MDSC-derived IL-10 and TGF-β activate the STAT3 and SMAD signaling pathways in naive T cells, skewing their differentiation away from effector fates and towards the Treg lineage ( 46 , 47 ).

This interplay establishes a potent “MDSC-Treg” immunosuppressive network, creating a self-amplifying loop of immune dysregulation that perpetuates the immunosuppressive state of late sepsis.

THERAPEUTIC STRATEGIES: TARGETING MDSCS IN SEPSIS

Given the central role of MDSCs in postsepsis immunosuppression, they represent a high-value therapeutic target. Current strategies can be broadly categorized by their mechanism of action, ranging from upstream blockade of generation to downstream inhibition of function. The therapeutic pipeline for targeting MDSCs in sepsis is summarized in Table 2 .

Modulating pathological myelopoiesis

This strategy aims to “nip it in the bud” by preventing the emergency granulopoiesis that generates MDSCs.

A primary strategy involves intercepting the pathological myelopoiesis that fuels MDSC accumulation. Given that granulocyte colony-stimulating factor (G-CSF) is a primary driver of pathological myelopoiesis, blocking the G-CSF receptor (G-CSFR) theoretically offers a way to limit MDSC expansion. However, true pharmacologic G-CSFR antagonism has not been successfully demonstrated as a therapeutic strategy in sepsis models. The application of such blockade is severely limited by the risk of inducing profound neutropenia and impairing bacterial clearance ( 48 , 49 ). Consequently, rather than blunt upstream blockade of G-CSFR, current research focuses on modulating downstream signaling effectors (e.g., STAT3 or specific metabolic enzymes) to selectively inhibit the suppressive phenotype while preserving essential granulopoiesis ( 50 ). The JAK/STAT3 pathway is a master regulator of MDSC expansion and cytokine production. Consequently, small-molecule JAK inhibitors (e.g., ruxolitinib) have been proposed as potential therapeutics. It is crucial, however, to distinguish the rationale in sepsis from that in cancer immunotherapy. In oncology, immune-activating therapies (e.g., CAR-T) carry the risk of triggering a “cytokine storm” (Cytokine Release Syndrome). In stark contrast, the objective of JAK inhibition in sepsis is to dampen the endogenous cytokine storm and limit hyperinflammation. However, the clinical translation of this strategy faces a major dilemma regarding the net effect on immune homeostasis. While early inhibition might mitigate tissue damage and MDSC generation, broad suppression of the JAK/STAT pathway carries the significant risk of deepening sepsis-associated immunosuppression by impairing T-cell and NK-cell signaling. Due to this uncertainty—whether the benefit of controlling inflammation outweighs the risk of impairing host defense—the role of JAK inhibitors in sepsis remains strictly investigational. An alternative approach is to force their terminal differentiation. Retinoids, particularly all-trans retinoic acid, can induce MDSCs to differentiate into mature myeloid cells (e.g., macrophages and DCs) that lack immunosuppressive activity ( 51 ). In preclinical sepsis models, all-trans retinoic acid administration successfully decreased circulating MDSC numbers, restored T-cell functionality, and attenuated organ damage ( 52 , 53 ).

Inhibiting suppressive effector functions

Rather than depleting the cells, this approach seeks to neutralize their toxic weaponry while preserving their antimicrobial potential.

A second major approach focuses on disarming MDSCs by inhibiting their canonical immunosuppressive arsenal. MDSCs exert potent T-cell suppression through the enzymatic activity of ARG1 and iNOS, and the production of ROS. Pharmacological inhibition of these pathways is therefore a logical intervention. ARG1 inhibitors, such as Nω-hydroxy-L-arginine, can abrogate MDSC-mediated T-cell suppression by restoring local L-arginine levels, which has been correlated with improved outcomes in septic mice ( 46 , 54 ). Similarly, selective iNOS inhibitors like 1400W can curtail the production of suppressive NO, thereby rescuing T-cell proliferation and cytokine secretion ( 55 , 56 ). PDE-5 inhibitors downregulate MDSC activity by suppressing the expression of ARG1 and NOS2, yet their role in tuberculosis or other infectious diseases warrants further investigation ( 57 ). Furthermore, antioxidants, including N-acetylcysteine, can neutralize ROS-mediated suppression, alleviating the inhibitory pressure MDSCs exert on surrounding immune cells ( 58 ). In sepsis-induced immunosuppression, MDSCs likely exert a critical inhibitory effect on T cells via the PD-L1/PD-1 axis, thus making it a potential therapeutic target ( 59 ).

Metabolic reprogramming

MDSCs rely on specific metabolic pathways for their survival and function.

Metabolic reprogramming is a hallmark of MDSC activation. However, it is crucial to recognize that the majority of our mechanistic understanding stems from the chronic, sterile inflammation of the tumor microenvironment. In contrast, sepsis induces an acute, pathogen-driven metabolic storm. While similarities likely exist, the following mechanisms—largely defined in oncology—serve as a conceptual framework that requires rigorous validation in the context of sepsis. MDSCs exhibit a high reliance on FAO for energy production to sustain their suppressive activities ( 60 ). Inhibiting this pathway with agents like etomoxir, a carnitine palmitoyltransferase 1 inhibitor, can effectively block FAO, impair MDSC function, and increase survival in preclinical sepsis models ( 61 ). In addition to FAO, aberrant glucose metabolism is crucial for MDSC regulation. Drugs that target the glycolytic pathway, such as 2-deoxy-D-glucose, have been shown to inhibit MDSC proliferation and function, offering another strategy to ameliorate immune dysregulation in sepsis ( 62 , 63 ). Nrf2 serves as a key regulator of metabolic reprogramming in MDSCs. It activates antioxidant pathways, enhances glucose metabolism and mitochondrial function, thereby driving MDSCs into a state of high metabolism and proliferation. The Nrf2-mediated expansion of MDSCs can be reversed by mTOR inhibitors such as rapamycin ( 41 ). Additionally, glucocorticoid receptors have been shown to regulate MDSCs’ function via mitochondrial metabolism in immune thrombocytopenia. Their role in sepsis warrants further investigation ( 61 ).

Subset-specific depletion

Precision medicine requires targeting pathogenic subsets without wiping out protective neutrophils.

While the heterogeneity of MDSCs complicates broad targeting, the identification of subset-specific surface markers offers a path toward precision medicine. A prime example is the LOX-1, which is highly expressed on PMN-MDSCs from human sepsis patients but not on normal neutrophils ( 64 ). This differential expression makes LOX-1 an ideal target for antibody-based therapies designed to specifically deplete the pathogenic PMN-MDSC population without inducing neutropenia ( 65 ). Similarly, CD84 has emerged as a potential target of interest. Although primarily studied in the context of chronic malignancy (e.g., multiple myeloma), where it mediates MDSC-macrophage crosstalk, CD84 is known to be upregulated on myeloid cells during inflammation. While direct evidence in sepsis models is currently lacking, it is hypothesized that blocking CD84 could disrupt the suppressive microenvironment. Thus, anti-CD84 therapies represent an investigational avenue that warrants testing to see if the preclinical success observed in oncology can be translated to the acute septic setting ( 66 ).

Indirect modulation

Targeting the environment that fosters MDSC development.

Growing evidence implicates the gut microbiota as a critical regulator of systemic immunity. While direct evidence in sepsis models is emerging, proof-of-concept studies in other inflammatory conditions suggest a strong link. For instance, in models of cardiac allograft rejection and renal fibrosis, probiotic administration or fecal microbiota transplantation successfully corrected dysbiosis and reduced MDSC frequency ( 67 , 68 ). Extrapolating to sepsis, where gut barrier failure and dysbiosis are ubiquitous, it is plausible that similar interventions could limit the gut-derived signals (e.g., LPS translocation) that drive pathological myelopoiesis. However, specific studies confirming that microbiome restoration improves sepsis survival via an MDSC-dependent mechanism are still needed.

INTEGRATING MDSCS INTO THE SEPSIS IMMUNOSUPPRESSIVE NETWORK

Multifaceted roles of MDSCs in mediating sepsis-associated immunosuppression

The continuum of HLA-DR low /S100A high monocytes

Clinically, the downregulation of HLA-DR on monocytes is the gold standard biomarker for sepsis-associated immunoparalysis. It is crucial to recognize that M-MDSCs (CD14 + HLA-DR −/lo ) and these “immunoparalyzed monocytes” likely represent the same pathophysiological entity or a highly overlapping continuum. Recent high-dimensional analyses have further refined this signature, identifying a population of HLA-DR low /S100A high monocytes that drive immune suppression. These cells share the fundamental MDSC characteristic: the inability to present antigen coupled with the active secretion of suppressive mediators. Thus, the expansion of M-MDSCs is not an isolated event but the cellular basis of the broadly observed “monocyte deactivation” in septic patients.

Driving T-cell exhaustion (PD-1, TIM-3, and LAG-3)

MDSCs are potent drivers of the T-cell exhaustion phenotype. Beyond the well-established PD-L1/PD-1 axis, MDSCs create a microenvironment that upregulates alternative checkpoint receptors. Through the secretion of Galectin-9 (which binds to TIM-3) and the distinct metabolic pressure of amino acid deprivation, MDSCs promote the expression of LAG-3 and TIM-3 on CD4 + and CD8 + T cells. This results in a “deeply exhausted” T-cell phenotype characterized by proliferative arrest and impaired cytokine production, rendering the host vulnerable to secondary infections ( 69 ).

Linking metabolism to lymphopenia and apoptosis

MDSCs are mechanistically linked to sepsis-associated lymphopenia, a key predictor of mortality. The immense consumption of L-arginine and L-cysteine by MDSCs starves bystander lymphocytes, blocking their cell cycle progression (arrest in G0-G1). Furthermore, the release of ROS and peroxynitrite by PMN-MDSCs induces oxidative stress that triggers caspase-dependent apoptosis in T cells and NK cells. Therefore, MDSCs act as a “double hit” to the adaptive immune system: they functionally silence surviving T cells while actively driving their apoptotic depletion.

MDSCs and the legacy of sepsis: long-term outcomes

For the growing population of sepsis survivors, hospital discharge often marks the beginning of a chronic decline rather than a full recovery. This state, clinically defined as persistent inflammation, immunosuppression, and catabolism syndrome, is characterized by recurrent infections, physical weakness, and high long-term mortality. Emerging evidence implicates MDSCs as key drivers of this postsepsis legacy.

Recurrent infections and late mortality

While acute MDSC expansion may be protective against early cytokine storms, the failure of MDSC levels to normalize is strongly associated with poor long-term outcomes. Studies indicate that patients with persistently elevated MDSCs at discharge are significantly more prone to recurrent nosocomial infections (e.g., pneumonia, C. difficile) and have higher 6-month and 1-year mortality rates. This suggests that MDSCs actively maintain a state of “immunological scarring,” preventing the restoration of immunocompetence.

Muscle weakness and catabolism

Furthermore, the systemic effects of MDSCs may extend to metabolic and physical frailty. The chronic inflammatory milieu sustained by MDSCs—specifically through the secretion of cytokines like IL-6 and TNF-α—drives muscle catabolism and cachexia. This aligns with the clinical observation of profound intensive care unit-acquired weakness that persists for months. Thus, targeting the resolution of MDSC expansion may be critical not only for immune restoration but also for improving the functional quality of life in sepsis survivors.

Lessons from clinical trials targeting MDSC-related pathways

Although no therapies specifically designed to deplete MDSCs have been approved for sepsis, several clinical interventions have targeted pathways integral to MDSC biology—namely, myelopoiesis regulation, immune checkpoint inhibition, and arginine metabolism. The mixed results of these trials offer critical insights into the complexity of MDSC modulation.

Growth Factors (G-CSF/GM-CSF)

G-CSF and GM-CSF drive the generation of myeloid cells. Historically, these agents were tested in sepsis to boost host defense against pathogens. However, preclinical evidence suggests that chronic G-CSF stimulation is a potent driver of MDSC expansion and pathological myelopoiesis. In clinical settings, while some meta-analyses suggest GM-CSF may improve infection clearance in specific immunosuppressed phenotypes (e.g., sepsis-associated immunosuppression), large multicenter trials of G-CSF have generally failed to reduce mortality. This failure may stem from the “double-edged” outcome of stimulating myelopoiesis: while increasing neutrophil numbers, G-CSF may concurrently expand immunosuppressive PMN-MDSCs, thereby neutralizing the potential benefit.

Immune checkpoint blockade (PD-1/PD-L1)

MDSCs potently suppress T cells via the PD-L1/PD-1 axis. Consequently, blocking this pathway is a promising strategy to reverse sepsis-induced immunosuppression. Clinical trials (e.g., Nivolumab in sepsis) have shown that PD-1 blockade is well-tolerated and can restore HLA-DR expression and T-cell function. Although primarily aimed at exhausted T cells, these agents indirectly antagonize the suppressive machinery of MDSCs. However, patient selection remains a hurdle; applying checkpoint inhibitors in the hyperinflammatory phase could be detrimental, necessitating precise timing guided by biomarkers of immunosuppression.

Arginine and NO pathway modulation

MDSCs deplete L-arginine via ARG1 and iNOS, leading to T-cell cycle arrest. Theoretically, L-arginine supplementation should restore immune function. However, clinical results have been contradictory. While some enteral nutrition studies enriched with arginine showed benefits, others raised concerns about increased mortality, possibly due to excessive NO production leading to refractory hypotension. This paradox highlights the metabolic heterogeneity of MDSCs: simply supplying the substrate (arginine) without controlling the enzymatic machinery (ARG1/iNOS) expressed by MDSCs may fuel pathological vasodilation rather than immune restoration.

KEY CHALLENGES AND FUTURE PERSPECTIVES

Despite progressing from a niche concept to a central player in sepsis pathophysiology, the study of MDSCs remains beset by fundamental challenges. Translating preclinical successes into clinical realities is contingent upon resolving critical gaps in our understanding of their basic biology, metabolic regulation, and clinical identification.

The conundrum of phenotypic definition and functional heterogeneity

Technical variability and standardization barriers

The potential of MDSCs as prognostic biomarkers is currently compromised by significant technical heterogeneity across studies. Unlike stable soluble biomarkers (e.g., CRP or Procalcitonin), MDSCs are fragile, live cellular products highly sensitive to ex vivo manipulation.

Preanalytical variables

The most critical source of variability lies in sample handling. PMN-MDSCs, which constitute the majority of the MDSC population in sepsis, have a short half-life and are sensitive to apoptosis. Prolonged sample processing times or storage at room temperature can lead to significant cell loss or phenotypic alteration (e.g., shedding of CD62L or CD16). Furthermore, the method of isolation is a major confounder. Many early studies utilized density gradient centrifugation (e.g., Ficoll) to isolate peripheral blood mononuclear cells. However, as PMN-MDSCs differ in density from lymphocytes/monocytes, they often co-sediment with erythrocytes and granulocytes, leading to their inadvertent exclusion from analysis. This explains why some studies report low PMN-MDSC counts. Current consensus recommends using fresh whole blood with minimal manipulation to accurately capture the total MDSC pool. Additionally, the choice of anticoagulant (EDTA vs. heparin) can influence surface marker stability and must be standardized.

Analytical variability and gating strategies

There is no universal “MDSC marker.” Identification relies on multicolor flow cytometry panels that vary between laboratories. Distinguishing PMN-MDSCs (CD11b + CD14 − CD15 + /CD66b + ) from mature, activated neutrophils remains technically challenging, often requiring the specific marker LOX-1, which is not yet routinely included in clinical panels. Similarly, defining M-MDSCs (CD14 + HLA-DR − /lo) requires a standardized cutoff for “HLA-DR low,” which varies based on the fluorochrome intensity and cytometer settings.

Lack of standardized controls

Finally, interpretation is hindered by the choice of control groups. While MDSC expansion is robust compared with healthy donors, fewer studies compare septic patients to those with sterile inflammation (e.g., trauma or major surgery). Without such controls, it is difficult to distinguish whether MDSC expansion is a specific response to infection or a generic reaction to critical illness. Future prognostic studies must adhere to the minimal phenotypic standards proposed by the myeloid biological community to reduce this noise ( 27 ).

Beyond the blood: tissue compartmentalization and sampling limitations

Most clinical studies rely exclusively on peripheral blood analysis due to accessibility. However, experimental models indicate that circulating MDSCs may be merely a proxy for more profound events occurring in tissue compartments.

The spleen and CCI

Murine models of CCI—characterized by persistent inflammation and immunosuppression—reveal that the spleen acts as a critical reservoir. In late-stage sepsis, the spleen often shifts from an immune-activating organ to a site of extramedullary hematopoiesis, generating and harboring massive numbers of MDSCs. This splenic accumulation is thought to be the primary engine driving long-term lymphopenia and susceptibility to secondary infections, a phenomenon that blood analysis may underestimate.

Bone marrow and lung dynamics

Similarly, the bone marrow is the “factory” where the initial skewing of emergency myelopoiesis occurs, often preceding changes in the circulation. In the lungs, MDSC recruitment plays a dichotomous role. While early recruitment correlates with tissue damage, during the recovery phase, specific MDSC subsets are essential for the resolution of inflammation and tissue repair. Disruption of this homing can lead to persistent lung injury or fibrosis.

The sampling barrier

A major translational hurdle is the lack of “tissue visibility” in humans. Unlike animal models, where organs can be harvested, obtaining splenic or bone marrow biopsies from hemodynamically unstable septic patients is ethically and practically challenging. Consequently, it remains unclear whether the peripheral blood MDSC count accurately reflects the tissue burden or if “trapped” tissue-resident MDSCs continue to exert suppression even after blood levels normalize. Future postmortem studies or novel imaging techniques are needed to bridge this knowledge gap.

Impact of developmental age: neonates as a high-risk group

The host’s developmental stage critically influences MDSC biology. Notably, healthy neonates—particularly preterm infants—exhibit physiologically elevated levels of circulating MDSCs compared with adults ( 70 , 71 ). This preexisting high baseline may profoundly shape the immune response in neonatal sepsis. While potentially offering initial anti-inflammatory protection, it likely predisposes these immunologically immature hosts to a rapid and severe shift towards immunosuppression and immunoparalysis, possibly contributing to their high sepsis mortality. The specific role and dynamics of MDSCs in neonatal sepsis, especially in preterm infants, remain poorly understood and warrant dedicated investigation to guide age-specific therapeutic strategies.

Deconvoluting metabolic causality and plasticity

The causal link between MDSC metabolic reprogramming and immunosuppressive function in sepsis is debated. While cancer MDSCs often exhibit a stable phenotype reliant on enhanced glycolysis and FAO ( 60 , 72 ), septic MDSCs appear far more metabolically plastic. Preclinical models suggest a spatiotemporal shift: an early glycolytic upregulation (likely supporting rapid migration) transitions to a reliance on mitochondrial OXPHOS for long-term survival in later phases ( 62 ). The molecular mechanisms governing this metabolic switch, and whether master regulators like HIF-1α are central as they are in cancer, remain to be validated.

Furthermore, the immunoregulatory role of specific metabolites is conflicting. Methylglyoxal, a key T-cell inhibitor in tumor MDSCs, shows no correlation with MDSC numbers in sepsis patients ( 73 ). Conversely, while pathways like glutamine metabolism may generate anti-inflammatory mediators (e.g., succinate), how this function is balanced against systemic immunosuppression is unknown ( 74 ). This ambiguity extends to interventions; while FAO inhibitors like etomoxir improve survival in septic mice, it remains unclear if this is a direct effect on MDSC function or an indirect, off-target effect on other immune cells ( 75 ).

Future perspective: A critical need exists to “parse” these metabolic trajectories. This will require applying multiomics (metabolomics, lipidomics) and advanced technologies like in vivo metabolic flux imaging to track MDSC metabolic reprogramming across different sepsis stages. Elucidating these pathways and their key regulators (e.g., HIF-1α and NRF2) will be paramount for developing metabolic therapies that precisely target MDSC function without disrupting systemic metabolic homeostasis.

The dual challenge of clinical translation: diagnostics and therapeutics

The potential role of MDSCs as prognostic biomarkers remains a subject of debate. Early single-center studies (e.g., Janols et al., n = 41; Mathias et al., n = 66) reported strong associations between high circulating MDSC levels and increased nosocomial infections or 28-day mortality. However, these findings must be interpreted with caution due to small sample sizes and limited adjustment for confounders such as age, comorbidities, and sepsis severity scores. Notably, some subsequent larger analyses have failed to consistently replicate the association between MDSC frequency and organ dysfunction (sequential organ failure scores), suggesting that MDSC levels alone may not be an independent predictor of outcome across all septic populations ( 76 , 77 ).

Therapeutically, targeting MDSCs presents a significant therapeutic dilemma. The “double-edged sword” effect is a real and present danger. For example, the TRAIL-R2 agonist DS-8273a, effective in cancer, reduced MDSC numbers in septic mice but its function in sepsis is not yet understood ( 78 ). Similarly, the JAK inhibitor restored T-cell function but at the cost of an increased risk of cytokine storm ( 79 ). This highlights a critical problem: broad depletion strategies may eliminate the protective, anti-inflammatory functions of MDSC subsets while failing to resolve immunosuppression.

Future perspective: Overcoming this translational gap requires a precision medicine approach. 1) For diagnostics, protocols must be optimized and standardized, moving toward high-parameter, single-cell methods (e.g., mass cytometry) and validated in large-scale, multicenter cohorts. 2) For therapeutics, the goal must shift from “depletion” to “modulation.” Based on subset-specific functional differences, the development of sophisticated, reversible inhibitors (e.g., selectively targeting ARG1 but not iNOS) is needed to balance immune regulation and inflammation control.

Emerging Frontiers: microenvironmental crosstalk and epigenetic regulation

Finally, our understanding of how septic MDSCs are regulated is still in its infancy. The mechanisms of their interaction with other immune cells are not well-defined. While the MDSC–Treg immunosuppressive axis is well-established in cancer, its spatiotemporal synergy in sepsis remains unverified. Furthermore, nascent explorations into platelet-MDSC aggregates suggest a role in sepsis-associated coagulopathy, but whether platelets actively promote MDSC activation via S100A8/A9 release is purely speculative ( 80 ).

Perhaps the most significant “black box” is the role of epigenetic regulation. While extensive mapping of the epigenetic landscape (e.g., DNA methylation, histone H3K27 acetylation) has defined MDSC ontogeny in cancer, this area remains almost entirely unexplored in sepsis ( 81 ). It is hypothesized that the “emergency myelopoiesis” of sepsis may induce transient epigenetic marks distinct from the stable reprogramming seen in chronic malignancy. Thus, cancer-derived epigenetic mechanisms serve as informative but not yet validated models for understanding the transient expansion of MDSCs in septic patients.

Future perspective: These areas represent the next critical Frontier. Applying single-cell epigenomic techniques (e.g., scATAC-seq) to septic MDSCs is a necessary next step to identify sepsis-specific epigenetic marks. Understanding this regulatory layer, combined with the microenvironmental crosstalk, will be essential for identifying novel, upstream therapeutic targets to prevent MDSC generation rather than merely targeting their downstream effects.

CONCLUSIONS AND OUTLOOK

Sepsis-induced immunosuppression, a dominant global health challenge, is pivotally mediated by MDSCs. This review has articulated their profound duality: MDSCs act as “brakes” on early, excessive inflammation but later drive persistent immunosuppression and mortality, creating a central therapeutic dilemma. Clinical translation of MDSC-targeted therapies has been stalled by fundamental bottlenecks, including ambiguous phenotypic definitions, a poor understanding of their metabolic plasticity, and the “double-edged sword” risk of broad-acting interventions (Fig. 1 B).

The future of this field thus hinges on a paradigm shift from broad depletion toward precision immune modulation. This requires leveraging high-dimensional, single-cell platforms (e.g., scRNA-seq and mass cytometry) to build a standardized, sepsis-specific MDSC atlas and resolve their heterogeneity. Concurrently, multiomic approaches must deconstruct their spatiotemporal metabolic reprogramming and uncover the upstream epigenetic regulation dictating their fate.

The ultimate goal is no longer to simply eliminate MDSCs, but to modulate them—precisely dialing down their immunosuppressive activities while preserving their protective functions. Achieving this delicate balance is the critical next hurdle and will provide the rational basis for developing the first successful, targeted immunotherapies for this devastating syndrome.

Abbreviations: ARDS Acute Respiratory Distress Syndrome ARG1 Arginase 1 ATRA All-Trans Retinoic Acid CCI Chronic Critical Illness CLP Cecal Ligation and Puncture DAMPs Damage-Associated Molecular Patterns DCs Dendritic Cells e-MDSCs Early-stage Myeloid-Derived Suppressor Cells FAO Fatty Acid Oxidation G-CSF Granulocyte Colony-Stimulating Factor GM-CSF Granulocyte-Macrophage Colony-Stimulating Factor HIF-1α Hypoxia-Inducible Factor 1-Alpha HLA-DR Human Leukocyte Antigen-DR IL Interleukin IMCs Immature Myeloid Cells iNOS Inducible Nitric Oxide Synthase JAK Janus Kinase LOX-1 Lectin-type Oxidized LDL Receptor-1 MDSCs Myeloid-Derived Suppressor Cells M-MDSCs Monocytic Myeloid-Derived Suppressor Cells mTOR Mammalian Target of Rapamycin NK cells Natural Killer cells NO Nitric Oxide Nrf2 Nuclear Factor Erythroid 2-Related Factor 2 PAMPs Pathogen-Associated Molecular Patterns PD-1 Programmed Cell Death Protein 1 PD-L1 Programmed Death-Ligand 1 PMN-MDSCs Polymorphonuclear Myeloid-Derived Suppressor Cells ROS Reactive Oxygen Species STAT3 Signal Transducer and Activator of Transcription 3 TCR T-Cell Receptor TGF-β Transforming Growth Factor-Beta Tregs Regulatory T cells

This work was supported by the National Natural Science Foundation of China (No. 82202387).

The authors report no conflicts of interest.

S.Q., Z.K., and Q.W. are cofirst authors of this article.

REFERENCES

1. Markwart R Saito H Harder T . Epidemiology and burden of sepsis acquired in hospitals and intensive care units: a systematic review and meta-analysis. Intensive Care Med . 2020 ; 46 ( 8 ): 1536 – 1551 . 32591853 10.1007/s00134-020-06106-2 PMC7381455

2. Luhr R Cao Y Söderquist B . Trends in sepsis mortality over time in randomised sepsis trials: a systematic literature review and meta-analysis of mortality in the control arm, 2002-2016. Crit Care . 2019 ; 23 ( 1 ): 241 . 31269976 10.1186/s13054-019-2528-0 PMC6610784

3. Chousterman BG Swirski FK Weber GF . Cytokine storm and sepsis disease pathogenesis. Semin Immunopathol . 2017 ; 39 ( 5 ): 517 – 528 . 28555385 10.1007/s00281-017-0639-8

4. D’Elia RV Harrison K Oyston PC . Targeting the “cytokine storm” for therapeutic benefit. Clin Vaccine Immunol . 2013 ; 20 ( 3 ): 319 – 327 . 23283640 10.1128/CVI.00636-12 PMC3592351

5. Reizine F Lesouhaitier M Gregoire M . SARS-CoV-2-Induced ARDS associates with MDSC expansion, lymphocyte dysfunction, and arginine shortage. J Clin Immunol . 2021 ; 41 ( 3 ): 515 – 525 . 33387156 10.1007/s10875-020-00920-5 PMC7775842

6. Shi W Lin Q Zhang M . Herpes simplex virus-1 susceptibility as a risk factor for sepsis, with cytomegalovirus susceptibility elevating severity: insights from a bidirectional mendelian randomization study. Shock . 2024 ; 61 ( 6 ): 894 – 904 . 38662585 10.1097/SHK.0000000000002351

7. Janols H Bergenfelz C Allaoui R . A high frequency of MDSCs in sepsis patients, with the granulocytic subtype dominating in gram-positive cases. J Leukoc Biol . 2014 ; 96 ( 5 ): 685 – 693 . 24929004 10.1189/jlb.5HI0214-074R

8. Zhang W Fang X Gao C . MDSCs in sepsis-induced immunosuppression and its potential therapeutic targets. Cytokine Growth Factor Rev . 2023 ; 69 : 90 – 103 . 35927154 10.1016/j.cytogfr.2022.07.007

9. Gabrilovich DI . Myeloid-derived suppressor cells. Cancer Immunol Res . 2017 ; 5 ( 1 ): 3 – 8 . 28052991 10.1158/2326-6066.CIR-16-0297 PMC5426480

10. Mathias B Delmas AL Ozrazgat-Baslanti T ; the Sepsis, Critical Illness Research Center Investigators . Human myeloid-derived suppressor cells are associated with chronic immune suppression after severe sepsis/septic shock. Ann Surg . 2017 ; 265 ( 4 ): 827 – 834 . 27163951 10.1097/SLA.0000000000001783 PMC5102824

11. Si Y Merz SF Jansen P . Multidimensional imaging provides evidence for down-regulation of T cell effector function by MDSC in human cancer tissue. Sci Immunol . 2019 ; 4 ( 40 ): eaaw9159 . 31628161 10.1126/sciimmunol.aaw9159

12. Ferrer G Jung B Chiu PY . Myeloid-derived suppressor cell subtypes differentially influence T-cell function, T-helper subset differentiation, and clinical course in CLL. Leukemia . 2021 ; 35 ( 11 ): 3163 – 3175 . 33935280 10.1038/s41375-021-01249-7 PMC8550941

13. Cui J Cai W Lin J . Monocytic myeloid-derived suppressor cells as an immune indicator of early diagnosis and prognosis in patients with sepsis. BMC Infect Dis . 2024 ; 24 ( 1 ): 399 . 38609858 10.1186/s12879-024-09290-4 PMC11015644

14. Uhel F Azzaoui I Grégoire M . Early expansion of circulating granulocytic myeloid-derived suppressor cells predicts development of nosocomial infections in patients with sepsis. Am J Respir Crit Care Med . 2017 ; 196 ( 3 ): 315 – 327 . 28146645 10.1164/rccm.201606-1143OC

15. Loftus TJ Ungaro R Dirain M . Overlapping but disparate inflammatory and immunosuppressive responses to SARS-CoV-2 and bacterial sepsis: an immunological time course analysis. Front Immunol . 2021 ; 12 : 792448 . 34956225 10.3389/fimmu.2021.792448 PMC8696010

16. Bizymi N Matthaiou AM Matheakakis A . New perspectives on myeloid-derived suppressor cells and their emerging role in haematology. J Clin Med . 2022 ; 11 ( 18 ): 5326 . 36142973 10.3390/jcm11185326 PMC9504532

17. Kapor S Santibanez JF . Myeloid-derived suppressor cells and mesenchymal stem/stromal cells in myeloid malignancies. J Clin Med . 2021 ; 10 ( 13 ): 2788 . 34202907 10.3390/jcm10132788 PMC8268878

18. Hegde S Leader AM Merad M . MDSC: markers, development, states, and unaddressed complexity. Immunity . 2021 ; 54 ( 5 ): 875 – 884 . 33979585 10.1016/j.immuni.2021.04.004 PMC8709560

19. Cheng P Corzo CA Luetteke N . Inhibition of dendritic cell differentiation and accumulation of myeloid-derived suppressor cells in cancer is regulated by S100A9 protein. J Exp Med . 2008 ; 205 ( 10 ): 2235 – 2249 . 18809714 10.1084/jem.20080132 PMC2556797

20. Waight JD Netherby C Hensen ML . Myeloid-derived suppressor cell development is regulated by a STAT/IRF-8 axis. J Clin Invest . 2013 ; 123 ( 10 ): 4464 – 4478 . 24091328 10.1172/JCI68189 PMC3784535

21. Peranzoni E Zilio S Marigo I . Myeloid-derived suppressor cell heterogeneity and subset definition. Curr Opin Immunol . 2010 ; 22 ( 2 ): 238 – 244 . 20171075 10.1016/j.coi.2010.01.021

22. Crook KR Jin M Weeks MF . Myeloid-derived suppressor cells regulate T cell and B cell responses during autoimmune disease. J Leukoc Biol . 2015 ; 97 ( 3 ): 573 – 582 . 25583578 10.1189/jlb.4A0314-139R PMC4338846

23. Alattar H Xu H Zenke M . Fully functional monocytic MDSC generation from the murine HoxB8 cell line. Eur J Immunol . 2023 ; 53 ( 9 ): e2350466 . 37367431 10.1002/eji.202350466

24. Yang F Li Y Zou W . Adoptive transfer of IFN-γ-induced M-MDSCs promotes immune tolerance to allografts through iNOS pathway. Inflamm Res . 2019 ; 68 ( 7 ): 545 – 555 . 31055608 10.1007/s00011-019-01237-9

25. Yang F Li Y Wu T . TNFα-induced M-MDSCs promote transplant immune tolerance via nitric oxide. J Mol Med (Berl) . 2016 ; 94 ( 8 ): 911 – 920 . 26936474 10.1007/s00109-016-1398-z

26. Zhang Y Wang X Zhang R . MDSCs interactions with other immune cells and their role in maternal-fetal tolerance. Int Rev Immunol . 2022 ; 41 ( 5 ): 534 – 551 . 34128752 10.1080/08830185.2021.1938566

27. Bronte V Brandau S Chen SH . Recommendations for myeloid-derived suppressor cell nomenclature and characterization standards. Nat Commun . 2016 ; 7 : 12150 . 27381735 10.1038/ncomms12150 PMC4935811

28. Zheng R Chen S Chen S . Correlation between myeloid-derived suppressor cells and S100A8/A9 in tumor and autoimmune diseases. Int Immunopharmacol . 2015 ; 29 ( 2 ): 919 – 925 . 26508452 10.1016/j.intimp.2015.10.014

29. Yan D Adeshakin AO Xu M . Lipid metabolic pathways confer the immunosuppressive function of myeloid-derived suppressor cells in tumor. Front Immunol . 2019 ; 10 : 1399 . 31275326 10.3389/fimmu.2019.01399 PMC6593140

30. Nan J Xing YF Hu B . Endoplasmic reticulum stress induced LOX-1(+) CD15(+) polymorphonuclear myeloid-derived suppressor cells in hepatocellular carcinoma. Immunology . 2018 ; 154 ( 1 ): 144 – 155 . 29211299 10.1111/imm.12876 PMC5904716

31. Brudecki L Ferguson DA McCall CE . Myeloid-derived suppressor cells evolve during sepsis and can enhance or attenuate the systemic inflammatory response. Infect Immun . 2012 ; 80 ( 6 ): 2026 – 2034 . 22451518 10.1128/IAI.00239-12 PMC3370575

32. Hollen MK Stortz JA Darden D . Myeloid-derived suppressor cell function and epigenetic expression evolves over time after surgical sepsis. Crit Care . 2019 ; 23 ( 1 ): 355 . 31722736 10.1186/s13054-019-2628-x PMC6854728

33. Schrijver IT Herderschee J Théroude C . Myeloid-derived suppressor-like cells as a prognostic marker in critically ill patients: insights from experimental endotoxemia and intensive care patients. Cells . 2024 ; 13 ( 4 ): 314 . 38391927 10.3390/cells13040314 PMC10887109

34. O’Keefe GE Wu Y Mirabadi N . A pilot study assessing the utility of quantitative myeloid-derived suppressor cell measurements in detecting posttraumatic infection. Crit Care Explor . 2025 ; 7 ( 3 ): e1228 . 40100961 10.1097/CCE.0000000000001228 PMC11918587

35. Liu D Fu W Zhang T . Eliminating myeloid-derived suppressor cells alleviates immunosuppression and reduces susceptibility to secondary infections in a two-hit sepsis model. Cytokine . 2025 ; 191 : 156955 . 40339353 10.1016/j.cyto.2025.156955

36. Darcy CJ Minigo G Piera KA . Neutrophils with myeloid derived suppressor function deplete arginine and constrain T cell function in septic shock patients. Crit Care . 2014 ; 18 ( 4 ): R163 . 25084831 10.1186/cc14003 PMC4261583

37. Koutroulis I Batabyal R McNamara B . Sepsis immunometabolism: from defining sepsis to understanding how energy production affects immune response. Crit Care Explor . 2019 ; 1 ( 11 ): e0061 . 32166242 10.1097/CCE.0000000000000061 PMC7063962

38. Dietrich O Heinz A Goldmann O . Dysregulated immunometabolism is associated with the generation of myeloid-derived suppressor cells in staphylococcus aureus chronic infection. J Innate Immun . 2022 ; 14 ( 3 ): 257 – 274 . 34763332 10.1159/000519306 PMC9149459

39. Lou X Gao D Yang L . Endoplasmic reticulum stress mediates the myeloid-derived immune suppression associated with cancer and infectious disease. J Transl Med . 2023 ; 21 ( 1 ): 1 . 36593497 10.1186/s12967-022-03835-4 PMC9809056

40. Li H Dai F Peng Q . Myeloid-derived suppressor cells suppress CD4 + and CD8 + T cell responses in autoimmune hepatitis. Mol Med Rep . 2015 ; 12 ( 3 ): 3667 – 3673 . 25998831 10.3892/mmr.2015.3791

41. Ohl K Fragoulis A Klemm P . Nrf2 Is a central regulator of metabolic reprogramming of myeloid-derived suppressor cells in steady state and sepsis. Front Immunol . 2018 ; 9 : 1552 . 30034396 10.3389/fimmu.2018.01552 PMC6043652

42. Müller I Janson L Sauter M . Myeloid-derived suppressor cells restrain natural killer cell activity in acute coxsackievirus B3-induced myocarditis. Viruses . 2021 ; 13 ( 5 ): 889 . 34065891 10.3390/v13050889 PMC8151145

43. Stiff A Trikha P Mundy-Bosse B . nitric oxide production by myeloid-derived suppressor cells plays a role in impairing fc receptor-mediated natural killer cell function. Clin Cancer Res . 2018 ; 24 ( 8 ): 1891 – 1904 . 29363526 10.1158/1078-0432.CCR-17-0691 PMC7184799

44. Wang Y Tian J Wang S . The potential therapeutic role of myeloid-derived suppressor cells in autoimmune arthritis. Semin Arthritis Rheum . 2016 ; 45 ( 4 ): 490 – 495 . 26272193 10.1016/j.semarthrit.2015.07.003

45. Cari L De Rosa F Nocentini G . Context-dependent effect of glucocorticoids on the proliferation, differentiation, and apoptosis of regulatory T cells: a review of the empirical evidence and clinical applications. Int J Mol Sci . 2019 ; 20 ( 5 ): 1142 . 30845709 10.3390/ijms20051142 PMC6429178

46. Sehgal R Maiwall R Rajan V . Granulocyte-macrophage colony-stimulating factor modulates myeloid-derived suppressor cells and treg activity in decompensated cirrhotic patients with sepsis. Front Immunol . 2022 ; 13 : 828949 . 35720398 10.3389/fimmu.2022.828949 PMC9205181

47. Wu D Zhang H Miao C . The homeostasis and heterogeneity of regulatory T cells in sepsis. Burns Trauma . 2025 ; 13 : tkaf047 . 40873496 10.1093/burnst/tkaf047 PMC12378600

48. Paschall AV Zhang R Qi CF . IFN regulatory factor 8 represses GM-CSF expression in T cells to affect myeloid cell lineage differentiation. J Immunol . 2015 ; 194 ( 5 ): 2369 – 2379 . 25646302 10.4049/jimmunol.1402412 PMC4340766

49. Wang R Chen M Fu M . A research on the influence of G-CSF mobilization on donor’s peripheral blood MDSCs and its relationship with patient prognosis. Int Immunopharmacol . 2023 ; 124 ( Pt B ): 110998 . 37832238 10.1016/j.intimp.2023.110998

50. Dai J Kumbhare A Williams DA . Nfia deletion in myeloid cells blocks expansion of myeloid-derived suppressor cells during sepsis. Innate Immun . 2018 ; 24 ( 1 ): 54 – 65 . 29172874 10.1177/1753425917742956 PMC6240914

51. Nefedova Y Fishman M Sherman S . Mechanism of all-trans retinoic acid effect on tumor-associated myeloid-derived suppressor cells. Cancer Res . 2007 ; 67 ( 22 ): 11021 – 11028 . 18006848 10.1158/0008-5472.CAN-07-2593

52. Martire-Greco D Rodriguez-Rodrigues N Castillo LA . Novel use of all-trans-retinoic acid in a model of lipopolysaccharide-immunosuppression to decrease the generation of myeloid-derived suppressor cells by reducing the proliferation of CD34+ precursor cells. Shock . 2017 ; 48 ( 1 ): 94 – 103 . 27922552 10.1097/SHK.0000000000000812

53. Martire-Greco D Landoni VI Chiarella P . All-trans-retinoic acid improves immunocompetence in a murine model of lipopolysaccharide-induced immunosuppression. Clin Sci (Lond) . 2014 ; 126 ( 5 ): 355 – 365 . 24053409 10.1042/CS20130236

54. Köstlin N Vogelmann M Spring B . Granulocytic myeloid-derived suppressor cells from human cord blood modulate T-helper cell response towards an anti-inflammatory phenotype. Immunology . 2017 ; 152 ( 1 ): 89 – 101 . 28464218 10.1111/imm.12751 PMC5543463

55. Mazzoni A Bronte V Visintin A . Myeloid suppressor lines inhibit T cell responses by an NO-dependent mechanism. J Immunol . 2002 ; 168 ( 2 ): 689 – 695 . 11777962 10.4049/jimmunol.168.2.689

56. Su L Xu Q Zhang P . Phenotype and function of myeloid-derived suppressor cells induced by porphyromonas gingivalis infection. Infect Immun . 2017 ; 85 ( 8 ): e00213 – e00217 . 28533469 10.1128/IAI.00213-17 PMC5520439

57. Leukes V Walzl G du Plessis N . Myeloid-derived suppressor cells as target of phosphodiesterase-5 inhibitors in host-directed therapeutics for tuberculosis. Front Immunol . 2020 ; 11 : 451 . 32269568 10.3389/fimmu.2020.00451 PMC7109258

58. Fraternale A Green KA Schiavano GF . Inhibition of myeloid-derived suppressor cell (MDSC) activity by redox-modulating agents restores T and B cell proliferative responses in murine AIDS. Int Immunopharmacol . 2023 ; 124 ( Pt A ): 110882 . 37659111 10.1016/j.intimp.2023.110882

59. Ruan WS Feng MX Xu J . Early activation of myeloid-derived suppressor cells participate in sepsis-induced immune suppression via PD-L1/PD-1 axis. Front Immunol . 2020 ; 11 : 1299 . 32719675 10.3389/fimmu.2020.01299 PMC7347749

60. Hossain F Al-Khami AA Wyczechowska D . Inhibition of fatty acid oxidation modulates immunosuppressive functions of myeloid-derived suppressor cells and enhances cancer therapies. Cancer Immunol Res . 2015 ; 3 ( 11 ): 1236 – 1247 . 26025381 10.1158/2326-6066.CIR-15-0036 PMC4636942

61. Hou Y Xie J Wang S . Glucocorticoid receptor modulates myeloid-derived suppressor cell function via mitochondrial metabolism in immune thrombocytopenia. Cell Mol Immunol . 2022 ; 19 ( 7 ): 764 – 776 . 35414712 10.1038/s41423-022-00859-0 PMC9243139

62. Ohl K Tenbrock K . Reactive oxygen species as regulators of MDSC-mediated immune suppression. Front Immunol . 2018 ; 9 : 2499 . 30425715 10.3389/fimmu.2018.02499 PMC6218613

63. Zhu X Ji W Guo S . Glycolytic and lipid oxidative metabolic programs are essential for freshly-isolated regulatory T cells in mice with sepsis. RSC Adv . 2020 ; 10 ( 35 ): 21000 – 21008 . 35517767 10.1039/d0ra01947j PMC9054272

64. Condamine T Dominguez GA Youn JI . Lectin-type oxidized LDL receptor-1 distinguishes population of human polymorphonuclear myeloid-derived suppressor cells in cancer patients. Sci Immunol . 2016 ; 1 ( 2 ): aaf8943 . 28417112 10.1126/sciimmunol.aaf8943 PMC5391495

65. Li X Liu J Xing Z . Polymorphonuclear myeloid-derived suppressor cells link inflammation and damage response after trauma. J Leukoc Biol . 2021 ; 110 ( 6 ): 1143 – 1161 . 34636072 10.1002/JLB.3MA0821-029R

66. Zhou LX Jiang YZ Li XQ . Myeloid-derived suppressor cells-induced exhaustion of CD8 + T-cell participates in rejection after liver transplantation. Cell Death Dis . 2024 ; 15 ( 7 ): 507 . 39013845 10.1038/s41419-024-06834-z PMC11252260

67. Bromberg JS Hittle L Xiong Y . Gut microbiota-dependent modulation of innate immunity and lymph node remodeling affects cardiac allograft outcomes. JCI Insight . 2020 ; 5 ( 15 ): e142528 . 32759493 10.1172/jci.insight.142528 PMC7455058

68. Wang Y Chen Y Xiao Z . Fecal microbiota transplantation modulates myeloid-derived suppressor cells and attenuates renal fibrosis in a murine model. Ren Fail . 2025 ; 47 ( 1 ): 2480749 . 40141007 10.1080/0886022X.2025.2480749 PMC11951334

69. Zhang W Yu Q Gao X . Myeloid-derived suppressor cells induce exhaustion-like CD8(+) T cells during JEV infection. Int J Biol Sci . 2024 ; 20 ( 15 ): 5959 – 5978 . 39664572 10.7150/ijbs.102372 PMC11628328

70. Olin A Henckel E Chen Y . Stereotypic immune system development in newborn children. Cell . 2018 ; 174 ( 5 ): 1277 – 1292.e14 . 30142345 10.1016/j.cell.2018.06.045 PMC6108833

71. Olaloye O Gu W Gehlhaar A . A single-cell atlas of circulating immune cells over the first 2 months of age in extremely premature infants. Sci Transl Med . 2025 ; 17 ( 788 ): eadr0942 . 40043141 10.1126/scitranslmed.adr0942

72. Al-Khami AA Zheng L Del Valle L . Exogenous lipid uptake induces metabolic and functional reprogramming of tumor-associated myeloid-derived suppressor cells. Oncoimmunology . 2017 ; 6 ( 10 ): e1344804 . 29123954 10.1080/2162402X.2017.1344804 PMC5665069

73. Baumann T Dunkel A Schmid C . Regulatory myeloid cells paralyze T cells through cell-cell transfer of the metabolite methylglyoxal. Nat Immunol . 2020 ; 21 ( 5 ): 555 – 566 . 32327756 10.1038/s41590-020-0666-9

74. Kaneki M . Metabolic inflammatory complex in sepsis: septic cachexia as a novel potential therapeutic target. Shock . 2017 ; 48 ( 6 ): 600 – 609 . 28520694 10.1097/SHK.0000000000000906 PMC8675641

75. Ettel P Weichhart T . Not just sugar: metabolic control of neutrophil development and effector functions. J Leukoc Biol . 2024 ; 116 ( 3 ): 487 – 510 . 38450755 10.1093/jleuko/qiae057 PMC7617515

76. Waeckel L Venet F Gossez M . Delayed persistence of elevated monocytic MDSC associates with deleterious outcomes in septic shock: a retrospective cohort study. Crit Care . 2020 ; 24 ( 1 ): 132 . 32264937 10.1186/s13054-020-02857-y PMC7137224

77. Schrijver IT Karakike E Théroude C . High levels of monocytic myeloid-derived suppressor cells are associated with favorable outcome in patients with pneumonia and sepsis with multi-organ failure. Intensive Care Med Exp . 2022 ; 10 ( 1 ): 5 . 35146585 10.1186/s40635-022-00431-0 PMC8831012

78. Dominguez GA Condamine T Mony S . Selective targeting of myeloid-derived suppressor cells in cancer patients using DS-8273a, an agonistic TRAIL-R2 antibody. Clin Cancer Res . 2017 ; 23 ( 12 ): 2942 – 2950 . 27965309 10.1158/1078-0432.CCR-16-1784 PMC5468499

79. Lu C Talukder A Savage NM . JAK-STAT-mediated chronic inflammation impairs cytotoxic T lymphocyte activation to decrease anti-PD-1 immunotherapy efficacy in pancreatic cancer. Oncoimmunology . 2017 ; 6 ( 3 ): e1291106 . 28405527 10.1080/2162402X.2017.1291106 PMC5384417

80. Moghaddasnejad MR Sherafat NS Saki N . The role of platelets in the regulation of myeloid-derived suppressor cells in immune thrombocytopenia. Thromb Res . 2025 ; 256 : 109506 . 41138350 10.1016/j.thromres.2025.109506

81. Dai H Xu H Wang S . Connections between metabolism and epigenetic modification in MDSCs. Int J Mol Sci . 2020 ; 21 ( 19 ): 7356 . 33027968 10.3390/ijms21197356 PMC7582655