Abstract

   Sepsis is a life-threatening condition characterized by a dysregulated
   immune response to infection, leading to systemic inflammation and
   organ dysfunction. Macrophage polarization plays a critical role in
   pathogenesis of sepsis, and the influence of B lymphocyte-induced
   maturation protein-1 (Blimp-1) on this polarization is an underexplored
   yet pivotal aspect. This study aimed to elucidate the role of Blimp-1
   in macrophage polarization and metabolism during sepsis. Using a murine
   cecal ligation and puncture model, we observed elevated Blimp-1
   expression in M2 macrophages. Knockdown of Blimp-1 by
   macrophage-targeted adeno-associated virus in this model resulted in
   decreased survival rates, exacerbated tissue damage, and impaired M2
   polarization, underscoring its protective role in sepsis. In vitro
   studies with bone marrow-derived macrophage (BMDM), RAW264.7, and THP-1
   cells further demonstrated Blimp-1 promotes M2 polarization and
   modulates key metabolic pathways. Metabolomics and dual-luciferase
   assays revealed Blimp-1 significantly influences purine biosynthesis
   and the downstream Ornithine cycle, which are essential for M2
   macrophage polarization. In vitro studies with BMDM further suggested
   that the purine biosynthesis and Ornithine cycle metabolic regulation
   is involved in Blimp-1’s effects on M2 macrophage polarization, and
   mediates Blimp-1’s impact on septic mice. Our findings unveil a novel
   mechanism by which Blimp-1 modulates macrophage polarization through
   metabolic regulation, presenting potential therapeutic targets for
   sepsis. This study highlights the significance of Blimp-1 in
   orchestrating macrophage responses and metabolic adaptations in sepsis,
   offering valuable insights into its role as a critical regulator of
   immune and metabolic homeostasis.

   graphic file with name 41419_2025_7405_Figa_HTML.jpg

   Subject terms: Monocytes and macrophages, Infectious diseases

Introduction

   Sepsis, a life-threatening condition characterized by organ dysfunction
   resulting from a dysregulated host response to infection, affects
   approximately 48.9 million individuals globally each year [[36]1]. In
   2017, it was responsible for 11 million deaths, accounting for 19.7% of
   all global mortality [[37]2]. The pathogenesis of sepsis is
   multifaceted, involving intricate interactions between infectious
   microorganisms and the host, alongside various pathways including
   infection, inflammation, hypoxia, and metabolic reprogramming [[38]3].
   A key mechanism underpinning sepsis is the imbalance of inflammatory
   response. During the early stages of sepsis, the host’s immune response
   is highly inflammatory, with numerous inflammatory mediators present
   that damage both tissues and immune cells. This is followed by an
   advanced stage of sepsis, characterized by host immunosuppression
   [[39]4]. Consequently, the primary therapeutic goal for sepsis patients
   is the regulation of staging immunophenotypes and personalized immune
   therapies, which may effectively reduce sepsis-related mortality.

   In sepsis-induced organ dysfunction, monocytes and macrophages are the
   first line of host defense. Recent study indicates that
   circulation-derived Ly6C^hi inflammatory monocytes are recruited to the
   inflamed livers where they differentiated into macrophages. These
   monocyte-derived macrophages can even replace tissue macrophages,
   directly inducing or exacerbating tissue injury [[40]5]. Macrophages
   play a crucial role in mediating inflammatory responses and primarily
   contain two functional subsets: pro-inflammatory (M1) and
   anti-inflammatory (M2) macrophages. In the early stage of sepsis, IFN-γ
   and LPS induce M1 macrophage polarization [[41]6, [42]7]. The
   increasing number of M1 macrophages continue to release large
   quantities of inflammatory factors, such as IL-1β, TNF-α, and IL-6
   [[43]6, [44]7], leading to severe inflammatory responses and subsequent
   organ damage. Conversely, during late-stage sepsis, M2 macrophages can
   be activated by IL-4 and IL-13 [[45]8]. An excessive increase in M2
   macrophages results in the secretion of large quantities of IL-10,
   TGF-β, and other anti-inflammatory cytokines, inducing an
   immunosuppressive state in the host [[46]6, [47]7, [48]9]. This leads
   to immune paralysis and the recurrence of severe infection. Given the
   critical role of the M1/M2 polarization in sepsis [[49]10], it is
   essential to understand the regulatory mechanisms governing macrophage
   polarization. This is crucial for elucidating sepsis pathology and
   exploring new immunotherapy modalities.

   B lymphocyte-induced maturation protein-1 (Blimp-1) is a transcription
   suppressor of IFN-β encoded by the Prdm1 gene [[50]11, [51]12].
   Overexpression of Blimp-1 in pro-monocytic cells induces partial
   macrophage differentiation, characterized by the expression of surface
   markers such as CD11b and CD11c [[52]13]. Blimp-1 was identified as a
   marker for a specific macrophage population located near the
   microbe-exposed surface in the colon [[53]14]. Recent studies have
   shown CCL8 is a direct target of Blimp-1 in mononuclear macrophages,
   with elevated CCL8 levels observed in Prdm1^fl/^fl mice [[54]15]. Our
   previous study claimed Blimp-1 represses the production of several
   inflammatory cytokines, including IL-1β, IL-6, and IL-18, by directly
   binding to the genomic region and restricting the nuclear translocation
   and transcriptional activity of NF-κB [[55]16]. However, the potential
   regulatory role of Blimp-1 in macrophage polarization and its
   significance in sepsis represent a critical gap in current
   understanding.

   Emerging evidence indicates energy metabolism plays a crucial role in
   regulating macrophage polarization and function [[56]17]. Generally,
   pro-inflammatory M1 macrophages primarily rely on glycolysis to
   generate energy for innate immune responses, while M2 macrophages
   depend more on mitochondrial oxidative phosphorylation and the
   tricarboxylic acid cycle to satisfy their energy requirements [[57]18].
   During M2 polarization, transcription factors such as PPARγ and PGC-1β
   are critical for mitochondrial respiration and fatty acid oxidative
   metabolism [[58]19, [59]20]. Additionally, transketolase (TKT) is
   important for glycometabolism [[60]21], while glutamate-pyruvate
   transaminase (GPT) and glycine amidinotransferase (GATM) are involved
   in amino acid metabolism [[61]22]. Notably, Blimp-1 may participate in
   cellular metabolism, thereby modulating inflammatory marker expression
   in quiescent macrophages [[62]23]. For instance, Blimp-1 regulates
   IL-10 expression in group 2 innate lymphoid cells (ILC2s), which
   exhibit a metabolic dependence on glycolysis for IL-10 production
   [[63]24]. These findings suggest that Blimp-1 may regulate macrophage
   polarization through cellular metabolic pathways, thus influencing
   inflammatory responses in sepsis.

   In this study, we aim to elucidate the effects of Blimp-1 on macrophage
   polarization in sepsis from the perspective of cellular metabolism. We
   inhibited Blimp-1 expression and measured its effects on macrophage
   polarization phenotypes and functions both in vitro and in vivo.
   Subsequently, we utilized the metabolomics technology to explore the
   metabolic mechanisms by which Blimp-1 regulates macrophage
   polarization. We conducted dual-luciferase reporter assay to validate
   the metabolic pathways regulated by Blimp-1 in macrophages. In summary,
   our findings clarify a previously unknown mechanism by which Blimp-1
   plays a key role in orchestrating macrophage polarization during
   sepsis.

Results

Blimp-1 expression is elevated in sepsis-associated macrophages

   We had successfully established a cecal ligation and puncture (CLP)
   sepsis model in C57BL/6 J mice. Compared to sham-operated controls, CLP
   mice exhibited significant histopathological lesions in liver and
   lungs, along with increased serum levels of hepatic injury biomarkers
   and inflammatory cytokines (Fig. [64]1A–[65]C). Importantly, the
   peritoneal lavage fluid (PLF) of CLP mice contained a higher percentage
   of macrophages, with Blimp-1 expression specifically elevated in the M2
   subset (Fig. [66]1D–[67]F). The mRNA level of Blimp-1 was also
   significantly increased in these cells (Fig. [68]1G). This selective
   upregulation of Blimp-1 was observed in NK cells, γδ T cells and M2
   macrophages, but not in dendritic cells, CD4^+ T cells, CD8^+ T cells,
   B cells or neutrophils (Fig. [69]1H, Fig. S[70]1A-[71]1C). Given that
   dendritic cells, NK cells, and γδT cells constitute a minor proportion
   of immune cells in the mouse peritoneal cavity, it suggests that
   Blimp-1 is predominantly expressed in M2 macrophages within PLF and
   plays a vital role in sepsis (Fig. S[72]1D).

Fig. 1. Blimp-1 expression was significantly increased in macrophages in CLP
mice.

   [73]Fig. 1
   [74]Open in a new tab

   A H&E staining of liver and lungs in CLP and sham mice. The liver
   tissue displayed a disordered lobular structure, swollen and deformed
   hepatocytes, and dispersed binuclear hepatocytes (red arrow). The
   alveolar space was widened with inflammatory cells infiltration and
   substantial leakage of red blood cells (red arrow). Scale bar = 50 μm.
   The degree of liver injury was evaluated according to Ishak score. The
   degree of lung injury was evaluated based on pulmonary hemorrhage,
   pulmonary edema and widening of pulmonary septum. B Serum levels of ALT
   and AST. C Serum levels of inflammatory cytokines. D Percentage of
   macrophages in peritoneal lavage fluid (PLF), assessed by flow
   cytometry. E Geometric mean of M1 (CD86^+) and M2 (CD206^+ or CD204^+)
   macrophages in PLF, assessed by flow cytometry. F Geometric mean of
   Blimp-1 in macrophages with different polarization phenotypes in PLF,
   analyzed by flow cytometry. G The mRNA levels of Blimp-1 in PLF cells,
   measured by real-time PCR. H Geometric mean of Blimp-1 in CD4^+ T
   cells, CD8^+ T cells, B cells, neutrophils and macrophages in the PLF
   of mice. Data are presented as the mean ± standard error of independent
   experiments. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Blimp-1 knockdown exacerbates sepsis by impairing M2 polarization

   To elucidate the functional significance of Blimp-1 upregulation, we
   knocked down its expression using shRNA-Blimp-1 and assessed the impact
   on CLP mice (as depicted in Fig. [75]2A). Adeno-associated virus (AAV)
   effectively reduced Blimp-1 in mouse PLF cells (Fig. S[76]2A). We also
   found shRNA released into the circulation (Fig. S[77]2B). Three weeks
   post-injection, the survival rate of CLP mice was significantly lower
   in the Blimp-1 knockdown group (Fig. [78]2B). Histopathological
   analysis revealed exacerbated liver and lung tissue damage (Fig.
   [79]2C). These findings underscore the protective role of Blimp-1 in
   modulating the inflammatory response and tissue repair during sepsis.
   Following shRNA-Blimp-1 intervention, the PLF of CLP mice exhibited an
   elevated total macrophage count, yet a diminished proportion of M2
   macrophages, as evidenced by reduced presence of CD206^+ and
   CD206^+Blimp-1^+ cells (Fig. [80]2D–[81]F). In contrast, Blimp-1
   expression in other immune cells, such as CD4^+ T cells, CD8^+ T cells,
   and B cells, remained unaffected in both PLF (Fig. [82]2G) and spleen
   grinding fluid (Fig. [83]2H). These observations suggest that Blimp-1’s
   protective effect in sepsis may depend on its regulatory role in M2
   macrophage polarization.

Fig. 2. ShRNA-Blimp-1 aggravated multiple organ damages in CLP mice.

   [84]Fig. 2
   [85]Open in a new tab

   A Diagram of CLP mouse model construction and experiment timeline. B
   The survival rates of CLP mice intraperitoneally injected with
   adeno-associated virus (AAV) carrying shRNA-NC or shRNA-Blimp-1,
   respectively (n = 10 per group). C H&E staining of liver and lungs in
   mice. Scale bar = 50 μm. Red arrows indicate hepatic lobular structure
   damage, inflammatory cell infiltration and aggregation, and hepatocyte
   swelling and deformation in the liver. Blue arrows indicate destroyed
   alveolar cavity with inflammatory cell infiltration and congestion in
   the lungs. The degree of liver injury was evaluated according to the
   Ishak score. The degree of lung injury was evaluated based on pulmonary
   hemorrhage, pulmonary edema and widening of pulmonary septum. D
   Percentage of total macrophages and M2 macrophages in PLF. E Geometric
   mean of CD206 in macrophages from PLF. F Percent of CD206^+Blimp-1^+
   macrophages in PLF. G Geometric mean of Blimp-1 in CD4^+ T cells, CD8^+
   T cells, and B cells from PLF. H Geometric mean of Blimp-1 in CD4^+ T
   cells, CD8^+ T cells, and B cells in the spleen. I Geometric mean of
   Blimp-1 in macrophage from PLF. J Percentage of M2 macrophages in PLF.
   K H&E staining of liver and lungs in mice. The alveolar space was
   widened with inflammatory cells infiltration and substantial leakage of
   red blood cells (red arrow). Scale bar = 50 μm. The degree of lung
   injury was evaluated based on pulmonary hemorrhage, pulmonary edema and
   widening of pulmonary septum. Data are presented as the mean ± standard
   error of independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001,
   ****P < 0.0001. Mice are intraperitoneally injected with AAV particles
   carrying shRNA-Blimp-1 or shRNA-NC (B–H) and macrophage-targeted
   shRNA-Blimp-1 or shRNA-NC (I–K).

   To establish that the observed effects of Blimp-1 in septic mice are
   specifically mediated by macrophages, we utilized a macrophage-targeted
   AAV to knock down its expression, effectively silencing Blimp-1 in
   macrophages (Fig. [86]2I). Mice injected with AAV exhibited expression
   of GFP in the liver and lung (Fig. S[87]2C). Subsequent to this
   intervention with shRNA-Blimp-1 targeted at macrophages, the PLF from
   CLP mice exhibited a reduced proportion of M2 macrophages (Fig.
   [88]2J). We opted to conduct our time point analysis at an early stage,
   specifically at 18 hours, prior to any recorded animal fatalities. This
   early assessment might explain why histopathological analyses did not
   reveal liver tissue damage; however, an increase in lung tissue damage
   was distinctly evident (Fig. [89]2K).

Blimp-1 promotes monocyte-macrophage differentiation and M2 polarization

   We next assessed the temporal expression of Blimp-1 during macrophage
   differentiation and polarization from bone marrow cells of naïve mice
   (as depicted in Fig. [90]3A). Blimp-1 expression was significantly
   higher in monocytes compared to B cells and granulocytes (Fig. [91]3B),
   and its mRNA levels progressively increased alongside bone
   marrow-derived macrophage (BMDM) differentiation and maturation (Fig.
   [92]3C). Upon polarization, M1 macrophages exhibited high secretion of
   IL-1β, IL-6, TNF-α, and IL-12 (Fig. S[93]3A); whereas M2 macrophages
   secreted high levels of TGF-β, CCL17, and CCL22 (Fig. S[94]3B).
   Notably, Blimp-1 mRNA levels were specifically elevated in M2
   macrophages compared to M0 and M1 counterparts (Fig. [95]3D),
   indicating a potential role for Blimp-1 in the differentiation and M2
   polarization of macrophages.

Fig. 3. Overexpression of Blimp-1 promoted macrophage differentiation and M2
polarization.

   [96]Fig. 3
   [97]Open in a new tab

   A Diagram of experimental process of bone marrow-derived macrophage
   (BMDM) differentiation and polarization. B Geometric mean of Blimp-1 in
   B cells, granulocytes, and monocytes from bone marrow of naïve mice. C
   The mRNA levels of Blimp-1 during BMDM differentiation. D The mRNA
   levels of Blimp-1 in different polarization types of BMDMs. E
   Morphology of THP-1 cells, differentiated THP-1 cells (dTHP-1), and
   differentiated THP-1 cells with Blimp-1 overexpression (dTHP-1 +
   bLIMP-1). Scale bar = 20 μm. F The mRNA levels of Blimp-1 in dTHP-1
   cells transfected with Blimp-1 plasmid or vehicle. G Geometric mean of
   CD206 in dTHP-1 cells with Blimp-1 overexpression. H The mRNA levels of
   MRC1, MSR1, ARG1 and IL10 in dTHP-1 cells with Blimp-1 overexpression.
   I The mRNA levels of Blimp-1 in RAW264.7 cells transfected with Blimp-1
   plasmid or vehicle. J Morphology of RAW264.7 cells. Scale bar = 10 μm.
   K The mRNA levels of M2 marker genes in RAW264.7 cells. Data are
   presented as the mean ± standard error of independent experiments.
   *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

   To further dissect Blimp-1’s influence on macrophage polarization,
   THP-1 monocytic cells were induced to differentiate into macrophages
   and subsequently transfected with a Blimp-1 expressing plasmid. The
   expression level of Blimp-1 was identified by real-time fluorescent
   quantitative PCR (Fig. [98]3E). Overexpression of Blimp-1 causes
   morphological changes consistent with M2 macrophages, including the
   appearance of extended pseudopods (Fig. [99]3F), and a significant
   increase in CD206 expression, as well as increased mRNA levels of M2
   markers such as MRC1, MSR1, ARG1 and IL10 (Fig. [100]3G, [101]H).
   Similar results were obtained with RAW264.7 macrophages overexpressing
   Blimp-1, where an increase in mRNA levels of Msr1, Mrc1, Arg1,
   Ppargc1b, Il10, Ido1, Pdl1, and Pdl2 was observed (Fig.
   [102]3I–[103]K). These findings collectively highlight the critical
   role of Blimp-1 in driving macrophage differentiation and the
   acquisition of the M2 phenotype.

Blimp-1 is essential for energy metabolism in M2 macrophages

   Further investigation of Blimp-1’s regulatory role in macrophage
   polarization was conducted by knocking down its expression in RAW264.7
   cell (Fig. [104]4A). Compared to scrambled non-specific control
   (shRNA-NC), shRNA-Blimp-1 effectively reduced Blimp-1 expression,
   particularly in M2 macrophages (Fig. [105]4B). This knockdown resulted
   in a significant decrease in the proportion of M2 (CD206^+) macrophages
   and a dramatic reduction in the mRNA levels of M2 markers Mrc1, Msr1,
   and Arg1 (Fig. [106]4C, [107]D). In contrast, the proportion of M1
   (CD86^+) macrophages and the mRNA levels of M1 markers Il12b, Nos2, and
   Tnfa remained unchanged (Fig. [108]4C–[109]E), underscoring the
   specific role of Blimp-1 in M2 macrophage polarization and function.

Fig. 4. ShRNA-Blimp-1 attenuated M2 macrophage polarization.

   [110]Fig. 4
   [111]Open in a new tab

   A Diagram of inducing polarization of RAW264.7 cells. B The mRNA levels
   of Blimp-1 in macrophages infected with lentivirus carrying shRNA-NC or
   shRNA-Blimp-1. C Percentages of M1 and M2 macrophages. D The mRNA
   levels of Mrc1, Msr1 and Arg1 in M0 and M2 macrophages. E The mRNA
   levels of Il12b, Nos2 and Tnfa in M0 and M1 macrophages. Data are
   presented as the mean ± standard error of independent experiments.
   *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

   We also examined the expression of genes pivotal to mitochondrial
   respiration, fatty acid oxidative metabolism, and glycometabolism. The
   robust expression levels of Pparg, Ppargc1b, Tkt, Gatm, and Gpt2 in M2
   macrophages, which were notably diminished following Blimp-1 knockdown
   (Fig. [112]5A). Conversely, overexpression of Blimp-1 in RAW264.7
   macrophages resulted in elevated mRNA levels for these genes (Fig.
   [113]5B). Employing the Seahorse XF analyzer, we conducted metabolic
   assays substantiated the impact of Blimp-1 on mitochondrial function.
   The knockdown of Blimp-1 corresponded with a reduction in mitochondrial
   oxygen consumption rates (OCRs) in M2 macrophages (Fig. [114]5C). This
   reduction is indicative of a decrease in the maximal respiration,
   mitochondrial ATP production, and spare respiratory capacity (Fig.
   [115]5D). Furthermore, the extracellular acidification rate (ECAR), a
   measure of glycolytic activity, was diminished after shRNA-Blimp-1
   intervention (Fig. [116]5E), reflecting lower basal and maximum
   glycolysis levels (Fig. [117]5F). Collectively, these findings
   underscore the capacity of Blimp-1 to facilitate M2 macrophage
   polarization through the modulation of key metabolic enzyme
   expressions.

Fig. 5. Blimp-1 regulated energy metabolism and metabolic marker expressions
in M2 macrophages.

   [118]Fig. 5
   [119]Open in a new tab

   A The mRNA levels of Pparg, Ppargc1b, Tkt, Gatm and Gpt2 in M2
   macrophages infected with lentivirus carrying shRNA-NC or
   shRNA-Blimp-1. B The mRNA levels of Pparg, Ppargc1b, Tkt, Gatm and Gpt2
   in RAW264.7 macrophages transfected with or without Blimp-1 expressing
   plasmid. C Oxygen consumption rate (OCR) curve of the Seahorse Mito
   Stress Test in M2 macrophages infected with lentivirus carrying
   shRNA-NC or shRNA-Blimp-1. D Quantification of OCR parameters in M2
   macrophages. E Extracellular acidification rate (ECAR) curve of the
   Seahorse Glycolytic Test in M2 macrophages infected with lentivirus
   carrying shRNA-NC or shRNA-Blimp-1. F Quantification of ECAR parameters
   in M2 macrophages. FCCP: Trifluoromethoxy carbonyl cyanide
   phenylhydrazone; Rot: rotenone; AA: antimycin A; 2-DG:
   2-Deoxy-d-Glucose. Data are presented as the mean ± standard error of
   independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001,
   ****P < 0.0001.

Blimp-1 modulates the metabolome of M2 macrophages

   We furtherly conducted a non-targeted metabolomics analysis on RAW264.7
   cells with Blimp-1 knockdown and identified a total of 121 metabolites
   (Fig. [120]6A), involving amino acids, carbohydrates, lipids, fatty
   acids, nucleotides, organic acids, and others. An orthogonal partial
   least squares discriminant analysis (OPLS-DA) model clearly
   distinguished the metabolite profiles between the shRNA-Blimp-1 and
   shRNA-NC groups (Fig. [121]6B). The heatmap revealed significant
   differences in amino acids and nucleotides between the two groups (Fig.
   [122]6C). The differential metabolites were displayed in Fig. [123]6D &
   Table [124]S1, with those exceeding a variable importance in projection
   (VIP) scores of 1.0 considered significant (blue and red dots).
   Notably, the levels of Ornithine and multi-amino acids increased
   following Blimp-1 inhibition (Fig. [125]6D, above panel).

Fig. 6. ShRNA-Blimp-1 altered metabolome in M2 macrophages.

   [126]Fig. 6
   [127]Open in a new tab

   RAW264.7 cells were induced by IL-4 and IL-13 to polarize towards M2
   macrophages, followed by infection with lentivirus carrying shRNA-NC or
   shRNA-Blimp-1. A Classification of the 121 metabolites identified in
   non-targeted metabolome analysis. B OPLS-DA Model Discrimination for
   non-targeted metabolome. C Heatmap of all metabolites for quantitative
   identification in the non-targeted metabolome. The color scale
   indicates the level of metabolite accumulation, while the class
   indicates the metabolites. D VIP scores comparing shRNA-Blimp-1 and
   shRNA-NC groups in non-targeted metabolome analysis. The green dots
   represented metabolites with VIP < 1.0 & |Corr.Coeffs. | < 0.9; the
   blue dots represented metabolites with VIP ≥ 1.0, and 0.9 ≤
   |Corr.Coeffs.| < 1.0; the red dots represented metabolites with
   VIP ≥ 1.0, and |Corr.Coeffs.| ≥ 1.0. E OPLS-DA Model Discrimination for
   nucleotide-targeted metabolome. F OPLS-DA Model Discrimination for
   amino acid-targeted metabolome. G SMPDB metabolic pathway enrichment
   analysis of the differential metabolites in the nucleotide-targeted
   metabolome (P < 0.05). H KEGG metabolic pathway enrichment analysis of
   the differential metabolites in the amino acid-targeted metabolome
   (P < 0.05).

   Subsequently, we conducted a quantitative metabolomics analysis
   targeting nucleotide metabolism and amino acid metabolism and
   identified 9 differential nucleotide metabolites and 17 differential
   amino acid metabolites between groups (VIP ≥ 1.0 & false discovery rate
   [FDR] < 0.05; Table [128]S2). OPLS-DA clearly distinguished the
   nucleotide and amino acid metabolite of shRNA-Blimp-1 and shRNA-NC
   groups (Fig. [129]6E, [130]F). Metabolic pathway enrichment analysis
   using the Small Molecule Blimp-1 knockdown significantly impacted GDP
   exchange, UDP exchange and ATP diphosphate hydrolase in M2 macrophages
   (Fig. [131]6G), as well as ammonia recycling, glutamine metabolism,
   urea cycle and glycine and serine metabolism (Fig. [132]6H). These
   findings suggest that Blimp-1 influences the metabolic pathways of
   nucleotides and amino acids in M2 macrophages.

Blimp-1 enhances purine biosynthesis and the downstream Ornithine cycle in M2
macrophages

   Building on the differential metabolite analysis, we concentrated on
   purine biosynthesis, nucleotide metabolism, and the Ornithine cycle.
   Blimp-1 knockdown led to a down-regulation of most metabolites involved
   in purine biosynthesis (Fig. [133]7A). The expression levels of genes
   encoding key enzymes in this pathway, such as Ppat, Pnp, Gda, and Xdh,
   were significantly reduced with Blimp-1 inhibition (Fig. [134]7B).
   Following Blimp-1 knockdown, the content of metabolites in the
   Ornithine cycle and the mRNA levels of key enzymes were significantly
   diminished (Fig. [135]7C, [136]D). Additionally, the concentrations of
   nucleotide metabolites GDP, GTP, ATP, and ADP were markedly reduced
   (Fig. [137]7E). Blimp-1 directly enhances the transcriptional activity
   of promoters for PPAT, GDA, XDH, APRT, ASS1, ASL, and SMA (Fig.
   [138]7F). Collectively, these results demonstrated that Blimp-1
   regulates purine biosynthesis and the downstream Ornithine cycle by
   promoting the transcription of key enzyme genes in M2 macrophages (Fig.
   [139]7G).

Fig. 7. Blimp-1 regulated purine biosynthesis and the Ornithine cycle in M2
macrophages.

   [140]Fig. 7
   [141]Open in a new tab

   RAW264.7 cells were induced by IL-4 and IL-13 to polarize towards M2
   macrophages, followed by infection with lentivirus carrying shRNA-NC or
   shRNA-Blimp-1. A Differential abundance of metabolites in the purine
   biosynthesis pathway. B The mRNA levels of key metabolic enzymes in
   purine biosynthesis pathway. C Differential abundance of metabolites in
   the Ornithine cycle. D The mRNA levels of key metabolic enzymes in the
   Ornithine cycle. E Differential abundance of nucleotides GTP, GDP, ATP,
   and ADP. F Dual-luciferase reporter assay of metabolic enzyme gene
   promoters in 293 T cells. The cells were transfected with Blimp-1
   expressing plasmid or the vehicle, along with reporters of PRPS1, PPAT,
   PNP, GDA, XDH, APRT, ASS1, ASL, and AMS promoters, respectively. G
   Diagram of the differential metabolic pathway in M2 macrophages. Green
   indicates enzymes with upregulated expression by Blimp-1; yellow
   indicates metabolites with increased abundance by Blimp-1. Data are
   presented as the mean ± standard error of independent experiments.
   *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Purine biosynthesis and Ornithine cycle metabolic regulation is involved in
M2 macrophage polarization promoted by Blimp-1

   We further investigated the potential role of purine biosynthesis and
   Ornithine cycle metabolic regulation in Blimp-1’s promotion of M2
   macrophage polarization. As glutamine serves as a hub of purine
   biosynthesis and the Ornithine cycle, thus linking the metabolism of
   two distinct biological macromolecules, we transfected a plasmid
   expressing Blimp-1 in BMDM, and we observed a significant decrease in
   the mRNA levels of the M2 macrophage marker Arg1 in BMDM
   post-transfection with Blimp-1, following administration of BPTES, an
   inhibitor of the glutamine metabolism pathway (Fig. [142]8A).
   Conversely, the mRNA levels of Arg1 in BMDM transfected with
   shRNA-Blimp-1 was found to be elevated upon glutamine supplementation
   (Fig. [143]8B). Additionally, glutamine supplementation was shown to
   rescue the mortality of CLP mice post intervention with
   macrophage-targeted AAV shRNA-Blimp-1 (Fig. [144]8C), along with an
   increased proportion of M2 macrophages (Fig. [145]8D). A schematic
   diagram summarizing our proposed regulatory mechanism of Blimp-1 on the
   metabolism, polarization, and function in macrophages, along with its
   protective effects in sepsis, is presented in Fig. [146]8E. Taken
   together, these results suggest that play key roles in Blimp-1’s
   influence on M2 macrophage polarization and mediate Blimp-1’s impact on
   septic mice.

Fig. 8. Purine biosynthesis and Ornithine cycle metabolic regulation is
involved in Blimp-1-mediated M2 macrophage polarization.

   [147]Fig. 8
   [148]Open in a new tab

   A The mRNA level of Arg1 in BMDM transfected with Blimp-1 plasmid or
   vehicle and subsequently treated with BPTES. B The mRNA level of Arg1
   in BMDM infected with lentivirus carrying shRNA-NC or shRNA-Blimp-1,
   and treated with glutamine (Gln). C The survival rates of CLP mice
   intraperitoneally injected with macrophage-targeted AAV carrying
   shRNA-NC or shRNA-Blimp-1, and treated with Gln, respectively (n = 10
   per group). D Percentage of M2 macrophages in PLF cells. E A proposed
   schematic diagram of the regulatory mechanism of Blimp-1 on the
   metabolism, polarization, and function in macrophages, and its
   protective effect in sepsis. When the body is suffered from
   polymicrobial infection, the Blimp-1 expression in macrophages will be
   enhanced and resulted in promoting M2 polarization, as well as
   invigorating purine biosynthesis and the downstream Ornithine cycle
   through Glutamine pathway, which furtherly suppressed secretion of the
   inflammatory cytokine, and finally protecting the organs (liver and
   lung) from damage in sepsis. Data are presented as the mean ± standard
   error of independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001,
   ****P < 0.0001.

Discussion

   A central component of sepsis pathophysiology is the polarization of
   macrophages. Macrophages undergo distinct polarization states (M1 and
   M2) that significantly influence sepsis outcomes. Nrf2 has been shown
   to play a protective role in sepsis-induced pulmonary injury and
   inflammation by modulating autophagy and NF-κB/PPARγ-mediated
   macrophage polarization [[149]25]. Similarly, Krüppel-like
   transcription factors reduce macrophage glycolysis and inflammatory
   cytokine secretion by suppressing the M1 macrophage polarized phenotype
   [[150]26]. Our recent work revealed that Blimp-1 inhibits the secretion
   of inflammatory cytokines via multiple toll-like receptors [[151]16],
   which elucidate a novel role for Blimp-1 in macrophage polarization
   during sepsis.

   Macrophage metabolic remodeling is essential for adapting to different
   polarized states and their functional changes [[152]21]. The
   polarization of M2 macrophages involves alterations in several
   metabolic pathways, including glucose, amino acid, fatty acid, and
   nucleotide metabolism. Specifically, in fatty acid metabolism, PGC-1β
   collaborates with PPARγ to inhibit the secretion of inflammatory
   factors in M1 macrophages and to promote the M2 polarization [[153]27].
   The mechanism may involve PPARγ and PGC-1β directly regulating ARG1
   expression, with their activation playing a crucial role in
   mitochondrial respiration and fatty acid oxidation in M2 macrophages
   [[154]19, [155]20].

   In glucose metabolism, TKT, a key enzyme in the pentose phosphate
   pathway downstream of D-fructose-1.6 diphosphate, is upregulated in M1
   macrophages along with glycolysis-related enzymes; however, this does
   not affect PPAT-related purine synthesis or cell proliferation
   [[156]28]. In amino acid metabolism, GATM and GPT2, which are involved
   in the glutamate pathway, facilitate glutaminolysis, leading to
   α-ketoglutarate accumulation and M2 polarization [[157]29, [158]30]. In
   our study, Blimp-1 suppression led to a significant reduction in the
   mRNA expression of Pparg, Ppargc1b, Tkt, Gatm, and Gpt2 in M2
   macrophages, while Blimp-1 overexpression resulted in a marked increase
   in these genes. These findings suggest that Blimp-1 regulates M2
   macrophage polarization by modulating key metabolic enzyme expressions.

   Enhanced glycolysis during M2 macrophage polarization provides a
   metabolic basis for improved mitochondrial oxidative phosphorylation.
   M2 macrophages sustain their metabolism primarily through the
   tricarboxylic acid cycle and mitochondrial oxidative phosphorylation,
   resulting a more robust oxidative metabolic profile compared to M1
   macrophages [[159]31]. Our results demonstrated that Blimp-1 knockdown
   led to substantial inhibition of both basal respiration and maximum
   respiratory capacity in mitochondrial oxidative phosphorylation, as
   well as glycolytic function. These results confirm that Blimp-1
   regulates both mitochondrial oxidative phosphorylation and glycolysis
   in macrophages.

   Our results claimed that Blimp-1 enhances the transcription of PPAT,
   ASS1, and ASL, affecting purine biosynthesis and the Ornithine cycle in
   M2 macrophages. Previous studies have linked the Ornithine cycle with
   M2 polarization [[160]18]. Notably, the expression of ARG1, a marker of
   M2 macrophages, is associated with the Ornithine cycle, while
   citrulline consumption is indicative of M1 polarization [[161]32].
   Although research on purine biosynthesis regulation in macrophage
   polarization is limited, both purine metabolism and the Ornithine cycle
   depend on glutamine, a star amino acid that regulates macrophage
   function [[162]30, [163]33]. Glutamine contributes α-ketoglutaric acid
   to the tricarboxylic acid cycle and facilitates acetylation
   modification of histones through acetyl-CoA, activating α-ketoglutaric
   acid-dependent epigenetic regulation and thereby influencing M2
   macrophage polarization [[164]34]. Actually, glutamine is the hub of
   purine biosynthesis and Ornithine cycle, connecting the metabolism of
   two kinds of biological macromolecules. Our study not only confirmed
   the regulatory mechanism of glutamine metabolism by Blimp-1, but also
   more comprehensively mapped the characteristics of glutamine-centered
   purine biosynthesis and Ornithine cycle metabolism in M2 macrophages.
   By elucidating the transcriptional regulation mechanism of Blimp-1 on
   various metabolic enzymes, this study revealed the metabolic regulation
   of M2 polarization and the pathological mechanism of sepsis
   progression.

   While our study provides valuable insights into the role of Blimp-1 in
   sepsis pathogenesis, several limitations warrant consideration. First,
   the study predominantly utilized animal models, necessitating
   validation in human samples. Second, due to the lack of suitable ChIP
   antibodies, the direct binding site and strength of Blimp-1 to target
   gene promoters cannot be evaluated in vivo. Third, longitudinal studies
   are needed to assess the long-term effects of Blimp-1 modulation on
   immune function and host recovery in sepsis.

   In conclusion, our study unveils a novel regulatory role for Blimp-1 in
   macrophage polarization during sepsis, mediated through modulation of
   purine biosynthesis and the Ornithine cycle. These findings expand our
   understanding of the complex interplay between metabolism and immune
   responses in sepsis pathology.

Materials and Methods

Culture of BMDMs

   Bone marrow cells were isolated from C57BL/6 male mice aged 6–8 weeks,
   which were euthanized via cervical dislocation. The femurs and tibia,
   along with lower limb muscles, were removed, sterilized in alcohol, and
   promptly transferred to a biosafety cabinet. After rinsing with PBS to
   eliminate any remaining alcohol, the lower extremity muscles were
   carefully stripped using scissors and a cotton cloth. The femur and
   tibia were fully exposed and removed, and the epiphysis on both sides
   were cut with scissors. The PBS buffer was extracted from the bone
   marrow cavity cells using a 1 mL syringe into a 15 mL centrifuge tube.
   The red bone marrow strips were gently blown several times until a
   uniform cell suspension was achieved. This was then centrifuged at
   800 rpm for 5 min, and the supernatant was discarded. The cells were
   cultured in DMEM complete culture medium containing 10 ng/mL M-CSF
   (315-02, Thermo Fisher Scientific, USA) for 7 days to achieve adherent
   growth of M0 macrophages. Subsequently, 100 ng/mL LPS (297-473-0,
   Sigma-Aldrich, Germany) and 40 ng/mL INF-γ (IF005, Sigma-Aldrich,
   Germany) were used to induce 2-day polarization into M1 macrophages.
   Alternatively, 40 ng/mL IL-4 (SAB1402819, Sigma-Aldrich, Germany) and
   20 ng/mL IL-13 (I1896, Sigma-Aldrich, Germany) were used to induce
   2-day polarization into M2 macrophages. A 10 mM GLS inhibitor BPTES
   (HY-12683, MCE, USA) was selected to inhibit the glutamine metabolic
   pathway, while 20 mM L-Glutamine (HY-N0390, MCE, USA) was used to
   supplement it. The mRNA levels of the M2 cell marker gene were measured
   by real-time PCR after 48 h.

Cell culture and treatment

   RAW264.7 and 293 T cells were cultured with DMEM medium (11965092,
   Gibco, USA) supplemented with 10% FBS (10091148, Gibco, USA) and 1%
   penicillin/streptomycin solution (15240112, Gibco, USA). THP-1 cells
   were cultured and induced to differentiate into M0 macrophages by
   50 ng/mL PMA (P8139, Sigma-Aldrich, Germany) for 48 h. Differentiated
   THP-1 or RAW264.7 cells were transiently transfected with Blimp-1
   expressing plasmid using Lipofectamine 3000 Transfection Reagent
   (L3000150, Thermo Fisher Scientific, USA) for 48 h. The construction of
   lentiviral vector shRNA-Blimp-1 or shRNA-NC was described as our
   previous report [[165]16].

Construction of AAV

   AAV particles carrying shRNA-Blimp-1 or shRNA-NC were customized by
   Genechem Co., Ltd. (Shanghai, China). The Blimp-1 RNAi sequence is
   5’-GCCTCTTCCCTAGGTTGTATC-3’, and the scrambled NC sequence is
   5’-TTCTCCGAACGTGTCACGT-3’. The shRNA was ligated into the
   pAAV-ApoE/hAATp-EGFP GV698 plasmid. The expressing plasmid was
   co-transfected with pHelper (carrying adenoviral origin genes) and
   pAAV-RC (carrying AAV replication and capsid genes) plasmids into
   AAV-293 cells for packaging recombinant AAV9 particles. The virus
   particles were purified through two rounds of CsCl ultracentrifugation
   and ultrafiltration. The shRNA-Blimp-1 AAV particles have a titer of
   6.56 × 10^13 v.g./mL determined by real-time PCR. The concentration of
   AAV gDNA was detected by quantitative PCR, and the virus solution was
   identified by lamp test. The macrophage-targeted recombinant AAV was
   constructed using the pAAV-F4/80p-EGFP-MIR155(RNAi)-SV40 PolyA plasmid
   with the same RNAi sequence and preparation procedure, and the virus
   titer was 9.93 × 10^12 v.g./mL.

Mice, sepsis model and in vivo intervention

   All animal procedures were conducted in accordance with protocols
   approved by the Animal Care Research Ethics Committee of the Capital
   Medical University of China (AEEI-2021-050) and housed under
   pathogen-free conditions in a barrier facility. Male C57BL/6 mice, aged
   6–8 weeks, were sourced from the Institute of Laboratory Animal
   Science, Chinese Academy of Medical Science (Beijing, China). A sepsis
   model was established using the CLP procedure, wherein the cecum was
   exposed, ligated at half the distance between the distal pole and the
   base of the cecum with a 3-0 silk suture, and punctured with an
   18-gauge needle. The abdominal wall and skin were closed with 4-0
   polypropylene sutures. Sterile saline (1 mL) was administered
   subcutaneously for fluid resuscitation during the postoperative period.
   A scoring system, based on three visual parameters—respiratory status,
   activity and response to stimuli, and eye appearance—was employed to
   monitor the mice over a 7-day period. The survival rate of the CLP
   model was about 50%, with the surviving CLP 7 d mice showing signs of
   recovery. Given the abdominal infection in CLP mice, we opted to
   intraperitoneally inject AAV particles carrying shRNA-Blimp-1 or
   shRNA-NC (Macrophage-targeted shRNA-Blimp-1 or Macrophage-targeted
   shRNA-NC) at a titer of 5.00E + 11 v.g/mL. The mice were fed for three
   weeks post-AAV injection before undergoing CLP surgery. Intraperitoneal
   injection of L-Glutamine (HY-N0390, MCE, USA 0.75 g/kg) was chosen to
   be completed within 24 h post-CLP surgery, with mortality observed over
   a 7-day period. CLP mice selected for Blimp-1 knockdown in macrophages
   using macrophage-targeted AAV were sampled at 18 h postoperatively.

Splenic cell isolation

   Spleen was excised and minced in PBS. Minced spleen suspension was
   filtered through a 70-μm filter, spun at 1200 rpm for 5 min. The cell
   precipitation was collected and centrifuged at 1200 rpm for 5 min. The
   cell precipitates were collected and split in RBC lysate (420302,
   BioLegend, USA) for 3 min and then rotated at 1200 rpm with PBS for
   5 min. The supernatant was discarded and washed again with PBS, and an
   appropriate number of cells were resuspended in the FACS buffer for
   flow cytometry staining.

Peritoneal lavage fluid (PLF) extraction

   The abdominal skin of mice was cut open, the peritoneum was exposed,
   the abdominal cavity was irrigated with 5 mL 0.9% NaCl. The normal
   saline was sucked out after full shaking for 90 s, the supernatant was
   rotated at 1200 rpm for 5 min and the supernatant was discarded. The
   cells were resuspended with FACS buffer and stained for flow cytometry.

Flow cytometry

   Cells were washed and collected by centrifuging at 1200 rpm for 5 min,
   and resuspended with FACS buffer. Single cell suspensions were stained
   with LIVE/DEAD Zombie Green (423111, BioLegend, USA) and ɑ-CD16/CD32
   (BE0307-1, BioXcell, China) before tetramer staining. The following
   mAbs were used: Blimp-1-PE (12-9850-82, Biosciences, USA), CD86-PE-Cy7
   (60-0862, Tonbo Bioscience San Diego, USA), CD45-APC-Cy7 (109824,
   BioLegend, USA), CD45-eFluor™ 450 (60-0862, Tonbo Bioscience San Diego,
   USA), F4/80-FITC (11-4801-82, eBioscience, USA), F4/80-APC (17-4801-82,
   eBioscience, USA), F4/80-Alexa Fluor® 700 (123129, BioLegend, USA),
   CD11b-BV510 (101263, BioLegend, USA),CD11b-PerCP-Cy5.5 (65-0112, Tonbo
   Bioscience San Diego, USA), CD3- eFluor450 (48-0032-82, eBioscience,
   USA),CD3-BV421 (100228, BioLegend, USA), CD19-PE-Cy7 (115520,
   BioLegend, USA), CD8-FITC (MA1-10303, eBioscience, USA), CD4-Percp
   (100432, BioLegend, USA), CD206-APC (141708, BioLegend, USA),
   CD206-BV421 (141717, BioLegend, USA), Gr-1-FITC (551460, BD Bioscience,
   USA), CD204-PE (154709, BioLegend, USA), CD48-Percp-Cy5.5 (103422,
   BioLegend, USA), CD48-PE-Cy7 (103424, BioLegend, USA), Gr-1-APC
   (108412, BioLegend, USA), CD11c-BV605 (117333, BioLegend, USA),
   NK1.1-BV785 (108749, BioLegend, USA), γδT-FITC (118105, BioLegend,
   USA). For transcription factor Blimp-1 staining, cells were fixed and
   permeabilized with Fix/perm buffer (00-5523-00, Thermo Fisher
   Scientific, USA) prior to intracellular staining. Data acquisition and
   analysis were performed using a CantoII flow cytometer (BD Bioscience,
   USA) or SA3800 Spectral Analyzer (SONY, JPN) and FlowJo Software
   (version 10.1; Tree Star, USA). The flow cytometry gating strategy is
   as follows: M2 macrophages (CD45^+F4/80^+CD11b^+CD206^+ or
   F4/80^+CD11b^+CD204^+), M1 macrophages (CD45^+F4/80^+CD11b^+CD80^+),
   neutrophils(CD45^+Gr-1^+CD11b^+CD48^-), CD4 T
   cells(CD45^+CD3^+CD4^+CD8^-), CD8 T cells(CD45^+CD3^+CD4^-CD8^+), B
   cells(CD45^+CD3^-CD19^+), NK cells(CD45^+NK1.1^+), dendritic
   cells(CD45^+CD11c^+), and γδT cells(CD45^+γδT^+).

Multiple cytokine assay

   The LEGENDplexTM MU Macrophage/Microglia Panel (13-plex) and VbP
   Reagent (740846, BioLegend, USA) were used to detect cytokines. Add
   25 μL standard product or plasma sample was added into the 96-well
   plate, add 25 μL/well of detection fluid and 25 μL/well detection
   microsphere and paste the sealing plate film. Shake for 2 h in the dark
   on a horizontal shaker. Centrifuge at 1,300 rpm for 5 min and discard
   the supernatant. Wash with 200 μL/well wash buffer, centrifuge at
   1,300 rpm for 5 min, and remove the supernatant. Add 25 μL/well
   detection antibody reagent and shake vigorously for 1 h at room
   temperature, avoiding light. Add 25 μL SA-PE reagent to each well and
   shake vigorously for 30 min at room temperature, avoiding light. Wash
   twice, add 150 μL/well was buffer, fully suspended the microspheres,
   transfer to the flow tube for machine detection. Duplicate the flow
   data and enter the BioLegend website, generate the standard curve and
   perform data analysis and processing.

Serum alanine aminotransferase and aspartate aminotransferase measurement

   Mouse serum samples were collected and the activities of serum
   aspartate transaminase (AST) and alanine transaminase (ALT) were
   determined using an automatic analyzer (Model 7600 Series, Hitachi,
   Japan).

Hematoxylin-eosin (H&E) staining

   After the mice were sacrificed, liver and lung samples were separated
   and fixed with 4% paraformaldehyde (BL539A, Biosharp, China) for 24 h,
   subsequently embedded in paraffin, sliced (4–5 μm). Histological
   sections of tissues were subjected to H&E staining. For each tissue
   section, 5 to 7 visual fields were randomly chosen for scoring. The
   Ishak score was employed to assess the degree of inflammation and
   necrosis in liver H&E staining. Lung tissues were graded on a scale
   from 0 to 4 based on inflammation, edema, hemorrhage, and alveolar
   septum thickening. The final score represented a comprehensive
   evaluation of all four injury parameters.

RNA isolation, reverse transcription, and real-time PCR

   Total RNA was extracted with RNA extraction Kit (ET101-01-V2,
   Transgenbiotech, China) and reverse transcribed using the PrimeScript™
   RT reagent Kit (RR047A, TaKaRa, Japan). Amplification was performed
   using the Power SYBR^® Green PCR Master Mix (4367659, Thermo Fisher
   Scientific, USA). The relative expression level of each transcript was
   normalized to murine GAPDH. The primers are listed in Table [166]S3.
   The relative expression levels of target genes were calculated by
   2-ΔΔCT (Livak method). Calculation formula: ΔCt = Ct target gene -Ct
   reference gene; ΔΔCt = ΔCt experimental group - ΔCt control group.

Plasmid construction, transfection, and dual-luciferase assay

   The promoter cDNA sequences of PRPS1, PPAT, PNP, GDA, XDH, APRT, ASS1,
   ASL, and SMS were constructed using the pGL3-basic-Luc-wt plasmid.
   Blimp-1 cDNA was cloned into pcDNA3.1+ plasmid as previous report
   [[167]35]. Appropriate concentration of 293 T cells was inoculated into
   a 24-well plate, with each well containing 1 × 10^5 cells. The total
   DNA transfected into each 24-well plate was 500 ng. Solution A
   consisted of 50 μL Opti-MEM medium, including 245 ng of pGL3-basic
   luciferase reporter carrying indicated gene promoter, 245 ng of
   pcDNA-BLIMP1, and 10 ng pRL-TK plasmid expressing Renilla luciferase
   (internal reference). Solution B was prepared by mixing 1 μL PEI-40K
   transfection reagent (G1802, Servicebio, China) with 49 μL Opti-MEM
   medium (31985062, Gibco, USA), and allowed to stand for 5 min. Solution
   B was then added to Solution A, mixed thoroughly, and incubated at room
   temperature for 15 min. The cell medium in the 24-well plate was
   replaced with Opti-MEM medium, and the AB mixture was then added to the
   respective well. This was incubated in a cell incubator at 37 °C for
   4–6 h, after which the medium was replaced with DMEM complete medium.
   After 24 h, cell lysate was collected and detected using the
   Dual-Luciferase® Reporter Assay System (E1910, Promega, USA). The
   microplate luminescence detector was used for detection (Turner
   Biosystems, USA) was used for detection. The calculation formula is as
   follow: Normalized value = Firefly luciferase reading value (reporter
   gene)/Renilla luciferase reading value (internal reference gene).
   Relative expression ratio=experimental group normalized value/control
   group normalized value.

Seahorse XF glycolytic stress and mitochondrial stress test

   Mitochondrial respiratory capacity and glycolytic function of M2
   macrophages were measures using Seahorse XF Glycolysis Stress Test Kit
   (103020-100, Agilent, USA) and Seahorse XF Cell Mito Stress Test Kit
   (103015-100, Agilent, USA) on a Seahorse XFe24 Analyzer (Agilent, USA).
   Inoculate cells at 1 × 10^4/well, culture overnight in a CO[2]-free
   cell incubator at 37 °C, and the coverage of cells before loading
   should reach about 80–90%. Add 200 μL/well ddH[2]O to the hydration
   plate, and incubate the whole probe plate overnight in a CO[2]-free
   cell incubator at 37 °C. Discard the sterile distilled water in the
   hydration plate, and add 200 μL XF calibration solution to each
   hydration plate well. Hydrate the whole probe plate for 1 h in a
   CO[2]-free cell incubator at 37 °C. Prepare Seahorse test solution:
   97 mL pH 7.4 Seahorse XF DMEM medium, 1 mL pyruvate (1 mM), 1 mL
   glutamine (2 mM) and 1 mL glucose (10 mM). Incubate the prepared test
   solution in a CO[2]-free cell incubator at 37 °C. Add 160 μL detection
   solution to cell culture wells, and place it in a CO[2]-free cell
   incubator for 1 h. Preheated drugs are sequentially added to the probe
   plate dosing holes, preparing for machine testing. Run the XF
   experiment on the computer, open the suite template corresponding to
   the wave software, set up the experimental grouping, and collect data
   for analysis after the experiment is completed.

Non-targeted and targeted metabolomics analysis

   The preparation of cell sample was consistent for non-targeted and
   targeted metabolomics detection. Briefly, 1 × 10^7 RAW264.7 cells were
   centrifuged at 4 °C at 200–1000 rcf for 10 min. The cell pellets were
   lysed ultrasonically and centrifuged at 18,000 g for 20 min. Ten
   microliter supernatants were mixed with 70 μl borate buffer and 20 μl
   6-minoquinolyl-N-hydroxysuccinimidyl carbamate derivatization reagent.
   The mixture was shaken at 1200 rpm for 10 min, then added 900 μl
   ultrapure water and mixed well. Using liquid chromatography mass
   spectrometry (LC-MS) and Metabo-Profile’s self-built database,
   non-targeted metabolomics analysis of metabolites in samples was
   performed qualitatively and quantitatively. The target metabolomics of
   amino acids and nucleotides in the samples were subjected to
   ultra-performance liquid chromatography/tandem mass spectrometry
   (UPLC-MS/MS) assay for quantification.

   XploreMET software (V3.0, Metabo-Profile, Shanghai, China) was used for
   quality control of samples and annotation of metabolites in
   non-targeted metabolomics profiling. In both non-targeted and targeted
   metabolomics profiling, we firstly conducted the principal component
   analysis to assess the overall distribution of samples and to verify
   the consistency of the analytical process. Subsequently, OPLS-DA and
   univariate analysis (Student t-test or Mann-Whitney test) were utilized
   to discern the variations in metabolic profiles and to identify
   metabolites with differential expression among groups. We identified
   differentially expressed metabolites according to VIP score ≥ 1.0 in
   the non-targeted metabolomics, while VIP ≥ 1.0 and FDR < 0.05 in
   targeted metabolomics. The P values were adjusted with FDR method in
   targeted metabolomics analysis. The VIP score of each metabolite
   comparing shRNA-Blimp-1 and shRNA-NC groups was displayed with
   “ggplot2” package. The heatmap of all metabolites was drawn by
   “pheatmap” package. The SMPDB and KEGG metabolic pathway enrichment
   analysis of the differential nucleotide- and amino acid-metabolites was
   performed with MetaboAnalyst tool.

Statistics analysis

   A two-sided P value < 0.05 was considered significant. All statistical
   analyses were performed using GraphPad Prism 9.0 software. The
   Kolmogorov-Smirnov test was used to inspect the normality and
   homogeneity of variance of all data. For two-group comparison, P values
   were derived from the Student t-test to determine differences between
   groups with normally distributed data and Mann-Whitney nonparametric
   test with other data.

Supplementary information

   [168]Supplemental Tables and Figures legends^ (23.4KB, docx)
   [169]Supplemental Figures^ (1.8MB, pdf)

Acknowledgements