MALAT1 regulates human macrophage metabolism by interacting with HADHB

TL;DR

This study shows how MALAT1, a noncoding RNA, interacts with HADHB to regulate metabolism in human macrophages, influencing inflammation.

Contribution

The novel interaction between MALAT1 and HADHB in regulating macrophage metabolism during inflammation is uncovered.

Findings

MALAT1 enhances HADHB thiolase activity via mitochondrial targeting by HuR-MTCH2.

Knockdown of MALAT1 increases glycolysis and fatty acid synthesis while reducing fatty acid oxidation.

MALAT1-HADHB interaction supports inflammation resolution by dampening pro-inflammatory activation.

Abstract

Long noncoding RNAs (lncRNAs) are critical regulators of immune responses and cellular metabolism. Here, we report a previously unrecognized interaction between MALAT1 and HADHB, which reveals additional regulatory roles for MALAT1 in human macrophages. Our findings demonstrate that MALAT1-HADHB interaction significantly enhances HADHB thiolase activity during the late phase of inflammation via HuR-MTCH2-mediated mitochondrial targeting of MALAT1. MALAT1 also negatively regulates the pro-inflammatory macrophage activation via HADHB. Knockdown of MALAT1 induces metabolic reprogramming, characterized by enhanced glycolysis, increased fatty acid synthesis, and reduced fatty acid oxidation, suggesting that MALAT1 suppresses inflammatory metabolic pathways. This study uncovers the MALAT1-HADHB interaction and demonstrates that MALAT1 regulates macrophage metabolic reprogramming, offering new insights into the metabolic control of inflammation and highlighting MALAT1 as a potential therapeutic target for inflammatory diseases.

•MALAT1 binds HADHB to boost its thiolase activity in human macrophages•HuR-MTCH2 targets MALAT1 to mitochondria•MALAT1 dampens pro-inflammatory macrophage activation via HADHB•MALAT1-HADHB-driven metabolic switching supports inflammation resolution

MALAT1 binds HADHB to boost its thiolase activity in human macrophages

HuR-MTCH2 targets MALAT1 to mitochondria

MALAT1 dampens pro-inflammatory macrophage activation via HADHB

MALAT1-HADHB-driven metabolic switching supports inflammation resolution

Molecular mechanism of gene regulation; Cell biology; Metabolic flux analysis

Figures

Mechanistic illustration of *MALAT1*-mediated regulation in THP-1-derived macrophages during inflammationUnder inflammatory conditions, *MALAT1* expression in the nucleus increases and drives the metabolic reprogramming of macrophages through several mechanisms. First, *MALAT1* enhances HADHB expression and *trans*-localizes from the nucleus to mitochondria via the HuR-MTCH2 axis, where it binds HADHB and elevates its thiolase activity to promote FAO. *MALAT1* also upregulates LDHB, facilitating the lactate-to-pyruvate conversion, while downregulating LDHA to reduce lactate production and dimi

Full text

Introduction

Long noncoding RNAs (lncRNAs) are RNA molecules typically more than 200 nucleotides in length that regulate gene expression and cellular processes through diverse mechanisms and interactions.1^,^2^,^3 One of the first lncRNAs linked to cancer is metastasis-associated lung adenocarcinoma transcript 1 (MALAT1), also known as NEAT2.4^,^5 MALAT1 is highly conserved and approximately 8,700 nucleotides in length.6 Extensive research indicates that that MALAT1 not only promotes tumor progression and metastasis,4^,^7^,^8^,^9^,^10 it also regulates lipid metabolism in hepatocellular carcinoma cells. MALAT1 promotes the expression of multiple genes in the adenosine 5′-monophosphate-activated protein kinase signaling pathway and the unsaturated fatty acid metabolism pathway, thereby regulating pre-mRNA splicing and transcription and, ultimately, enhancing glucose uptake and lipogenesis.11

In the context of inflammation, MALAT1 functions as an endogenous regulator in mesenchymal stem cells (MSCs). Studies have shown that MALAT1 influences MSC proliferation and angiogenesis and exerts immunosuppressive effects.12 MALAT1’s involvement in inflammation extends to the modulation of molecular pathways such as the mitogen-activated protein kinase/nuclear factor κB (NF-κB) cascade, thereby facilitating pro-inflammatory cytokine production.13 This regulation is particularly notable in chronic inflammatory conditions such as chronic obstructive pulmonary disease and certain viral infections, where MALAT1 modulates inflammatory gene expression and is associated with increased tissue damage and disease severity. During respiratory infections, including those caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2),14 MALAT1 is highly expressed and has been linked to elevated cytokine levels and greater lung tissue damage. For instance, in coronavirus disease 2019 (COVID-19) pneumonia, MALAT1 expression is associated with a compromised airway epithelial barrier function and an intensified inflammatory response, contributing to the severity of lung damage.15 Moreover, MALAT1 interacts with specific microRNAs, such as miR-125b and miR-203, thereby enhancing inflammatory signaling pathways and contributing to tissue injury in affected organs.13

Hydroxy acyl-CoA dehydrogenase trifunctional multienzyme complex subunit β (HADHB) plays a vital role in the final step of fatty acid oxidation (FAO), producing acetyl-CoA and a shortened acyl-CoA molecule. Recent studies suggest that HADHB is pivotal in regulating metabolic shifts in macrophages. Some studies indicate that exposure to multi-walled carbon nanotubes reduces HADHB mRNA levels, thereby inducing chronic pulmonary granulomatosis.16 Beyond its essential role in FAO, HADHB also interacts with lncRNAs, modulating inflammation and the metabolic processes that are vital for energy production and inflammatory responses.

In our prior studies, we identified a novel lncRNA, termed long noncoding FAO (Lncfao), which functions as an anti-inflammatory regulator in mouse bone marrow-derived macrophages (BMDMs). Our findings indicated that Lncfao upregulation in response to inflammation promoted resolution by suppressing pro-inflammatory cytokines such as Il1b and Il6 and reducing pro-inflammatory macrophages’ activation. Additionally, Lncfao bound to mouse HADHB, enhancing its thiolase activity, which then promoted FAO in macrophages and facilitated their differentiation into anti-inflammatory alternatively activated macrophages (M2). This effect was prominent during the resolution and tissue repair stages of inflammation, where Lncfao expression was observed in Ly6C^hi^ macrophages. Notably, Lncfao knockout in mice led to impaired inflammation resolution and wound healing.17

LncRNAs exhibit a high degree of species specificity, shaped by their unique sequence characteristics, evolutionary adaptations, and functional roles in different organisms. While Lncfao exerts these functions in mice, it is not present in human cells, and little is known about human lncRNAs that interact with HADHB to regulate inflammation during the resolution stage. Therefore, this study aimed to identify whether there are other lncRNAs that bind to human HADHB and exert a similar anti-inflammatory function during the resolution phase of inflammation. We hypothesized that human macrophages express an lncRNA that functionally parallels Lncfao’s anti-inflammatory role in mice by interacting with HADHB. Our findings highlight MALAT1’s capacity to regulate HADHB-mediated FAO and suppress pro-inflammatory activation, providing insights into its function in immunometabolic regulation.

Results

MALAT1 is identified as a transcript that binds to HADHB

To identify human lncRNAs that bind HADHB, we performed RNA immunoprecipitation (RIP) in phorbol 12-myristate 13-acetate (PMA)-differentiated THP-1 macrophages stimulated with lipopolysaccharide (LPS) (Figure 1A). Western blot analyses confirmed HADHB recovery in the RIP fraction, but not in the IgG control (Figures 1B and S1). Based on these findings, we performed RNA immunoprecipitation sequencing (RIP-seq) to identify lncRNAs that bind HADHB. According to the sequencing results, we obtained multiple HADHB-bound lncRNAs (Table S1). Among these, MALAT1 emerged as the top candidate, exhibiting the highest peak score among the HADHB-binding lncRNAs (Figure 1C).Figure 1MALAT1 binds to HADHB(A) Schematic representation of the RNA immunoprecipitation (RIP) of LPS-stimulated THP-1-derived macrophages differentiated with PMA for 24 h.(B) Western blot of RIP samples using an anti-HADHB antibody, with IgG as a negative control. The experiment was repeated; n = 3 independent experiments (independent cell cultures on different days).(C) Top ten candidate non-coding RNAs binding to HADHB, ranked by peak score from RIP-seq analysis. The experiment was repeated; n = 3 independent experiments (independent cell cultures on different days).(D) Fold enrichment of MALAT1 in RIP samples. The experiment was repeated; n = 3 independent experiments (independent cell cultures on different days); data are presented as mean ± SD and analyzed with the two-tailed unpaired Student’s t test.

Consistent with the RIP-seq data, we validated MALAT1 enrichment in HADHB RIP versus IgG control. Based on these results, we observed a markedly higher MALAT1 fold enrichment in HADHB RIP than in IgG (Figure 1D), indicating that MALAT1 specifically interacts with HADHB.

Taken together, we determine that the lncRNA that binds to HADHB is MALAT1.

MALAT1 translocates to mitochondria via the HuR-MTCH2 axis and enhances the thiolase activity of HADHB

Since HADHB is primarily localized in the mitochondria, we next examined whether MALAT1 was also present in the mitochondria of THP-1-derived macrophages. Subcellular fractionation showed that MALAT1 was predominantly localized in the nucleus, though it was also present in the mitochondria, albeit at much lower levels (Figure 2A). Based on the above experiments, we confirmed that MALAT1 was localized in both the nucleus and mitochondria.Figure 2MALAT1 translocates to mitochondria via the HuR-MTCH2 axis and enhances HADHB thiolase activity(A) MALAT1 transcript levels in the nucleus, cytoplasm, and mitochondria, normalized to U6 (nuclear RNA), GAPDH (cytoplasmic RNA), and COX2 (mitochondrial RNA). The experiment was repeated; n = 9 independent experiments (independent cell cultures on different days); data are shown as mean ± SD and analyzed with the two-way ANOVA followed by Tukey’s post hoc test.(B–D) MALAT1 transcript levels in the nucleus, cytoplasm, and mitochondria of PMA-differentiated THP-1 macrophages stimulated with LPS for various durations. For each LPS stimulation time point, n = 9 independent experiments (independent cell cultures and stimulated on different days); data are shown as mean ± SD and analyzed with the one-way ANOVA followed by Tukey’s post hoc test.(E) Western blot of RIP samples from THP-1-derived macrophages stimulated with LPS for 24 h or 48 h, probed with anti-HuR and anti-MTCH2, input and IgG controls shown. For each LPS stimulation time point, n = 3 independent experiments (independent cell cultures and stimulated on different days).(F) Fold-change enrichment of MALAT1 in HuR and MTCH2 RIP samples from THP-1-derived macrophages after LPS stimulation (24 h and 48 h). The experiment was repeated; n = 3 independent experiments (independent cell cultures on different days); data are presented as mean ± SD and analyzed with the two-tailed unpaired Student’s t test.(G) Reciprocal RIP-western confirms HuR-MTCH2 association in THP-1-derived macrophages after 24- and 48-h LPS stimulation. For each LPS stimulation time point, n = 3 independent experiments (independent cell cultures and stimulated on different days).(H) MALAT1 transcript levels in siRNA-transfected THP-1-derived macrophages (the experiment was repeated; n = 9 independent experiments) and MALAT1 overexpression THP-1 cells (the experiment was repeated n = 3 independent experiments), normalized to GAPDH and compared with siNC. Data are shown as mean ± SD and analyzed with one-way ANOVA followed by Dunnett’s post hoc test.(I) Schematic representation of FAO. HADHB, the β-subunit of the mitochondrial trifunctional protein (MTP), possesses thiolase activity and catalyzes the final step of FAO. Diagram created with BioRender.(J) Flow cytometric analysis of HADHB thiolase activity measured by mean fluorescence intensity (MFI) in THP-1-derived macrophages transfected with siRNA (siMALAT1) or stably overexpressing MALAT1 (Oe-MALAT1). The experiment was repeated; n = 3 independent experiments (independent cell cultures on different days); data are shown as mean ± SD and analyzed with the two-tailed unpaired Student’s t test.

Furthermore, a time-course analysis of MALAT1 localization following LPS stimulation revealed dynamic changes in its distribution. At 1 h post-LPS stimulation, MALAT1 transcription increased in the nucleus, while its cytoplasmic and mitochondrial levels decreased transiently. Between 1 and 6 h, nuclear MALAT1 decreased, while the mitochondrial level increased slightly; however, after 6 h of LPS stimulation, nuclear MALAT1 transcription increased rapidly. In parallel, due to its subcellular distribution, MALAT1 also increased in the cytosol and mitochondria during this interval (Figures 2B–2D). These findings suggest that MALAT1 initially accumulated in the nucleus during the early phase of inflammation but subsequently redistributed to the cytoplasm and mitochondria as inflammation progressed.

Emerging evidence indicates that the RNA-binding protein human antigen R (HuR; ELAVL1) participates in the trafficking of lncRNAs from the nucleus to mitochondria,18 while the outer mitochondrial membrane protein mitochondrial carrier homolog 2 (MTCH2) has been reported to interact with noncoding RNAs and modulate cellular functions.19^,^20 Motivated by these observations, we hypothesized that MALAT1 translocates from the nucleus to the mitochondria with the assistance of HuR and MTCH2. Using PMA-differentiated THP-1-derived macrophages stimulated with LPS for 24 or 48 h, we performed RIP followed by immunoblotting (RIP-WB) and RT-qPCR. In both HuR and MTCH2 RIP samples, the respective target proteins were readily detected by western blotting, whereas no signal was observed in IgG controls (Figure 2E). Consistently, MALAT1 was robustly enriched in both HuR and MTCH2 RIP samples, relative to IgG controls (Figures 2F, S2A, and S2B). To determine whether HuR and MTCH2 act in concert, we conducted reciprocal RIP-WB: after 24 and 48 h of LPS stimulation, HuR was detected in MTCH2 immunoprecipitates and, reciprocally, MTCH2 was detected in HuR immunoprecipitates (Figures 2G, S2C, and S2D). Together, these data support a model in which MALAT1 engages a HuR-MTCH2 axis to mediate its translocation between the nucleus and mitochondria.

To explore the function of MALAT1 in differentiated macrophages derived from THP-1 cells, we first knocked down MALAT1 in THP-1-derived macrophages using MALAT1 siRNA (siMALAT1) (Figure S3A), leading to a significant reduction in its transcription level. Concurrently, we generated stable MALAT1-overexpressing (Oe-MALAT1) cell lines via lentiviral transduction. Gene expression analyses confirmed that MALAT1 transcript levels were significantly increased in the overexpression group compared with the negative control (NC) group (Figure 2H).

We next explored whether MALAT1 regulates the thiolase activity of HADHB, which is responsible for the final step of FAO to get acetyl-CoA (Figure 2I). After adding FAO detection reagent FAOBlue to THP-1-derived macrophages, we evaluated the thiolase activity of HADHB using fluorescence-activated cell sorting to measure the mean fluorescence intensity (MFI) (Figure S3B). The experimental results were clear: after knocking down MALAT1, the thiolase activity of HADHB decreased significantly, compared with the negative control group. Conversely, in MALAT1-overexpressing cells, the thiolase activity of HADHB increased significantly, compared with the negative control (Figure 2J).

To determine whether this phenomenon extends to mouse macrophages, we screened two siRNAs targeting mouse Malat1 (siMalat1-1 and siMalat1-2) in RAW 264.7 cells. RT-qPCR confirmed that siMalat1-2 achieved superior knockdown and was, therefore, used in subsequent experiments (Figure S3C). Knockdown of Malat1 in RAW 264.7 cells, BMDMs, and liver-resident Kupffer cells significantly reduced the thiolase activity of Hadhb, relative to the negative control, in all three macrophage populations (Figure S3D).

Accordingly, MALAT1 translocates from the nucleus to the mitochondria via the HuR-MTCH2 axis, thereby promoting the thiolase activity of HADHB.

MALAT1 restrains pro-inflammatory activation of macrophages via HADHB

We next investigated the role of MALAT1 in regulating LPS-induced inflammatory gene expression. We first confirmed the transcript levels of MALAT1, different inflammatory cytokines, and macrophage polarization markers at various time points after LPS stimulation. The results indicated that MALAT1 was upregulated after 24 h, whereas the pro-inflammatory cytokines, the gene expressions of TNFA and IL1B, were downregulated. The gene expression of TGFB, an anti-inflammatory cytokine, increased after 24 h of LPS stimulation. Interestingly, HADHB transcript levels also increased following 24 h of LPS stimulation (Figures 3A, 3B, and S4A–S4C).Figure 3MALAT1 suppresses and restrains pro-inflammatory phenotype in macrophages(A and B) Time-course gene expression of MALAT1 (A) and HADHB (B) in PMA-differentiated THP-1-derived macrophages following LPS stimulation at the indicated time points. For each LPS stimulation time point, n = 9 independent experiments (independent cell cultures and stimulated on different days); data are shown as mean ± SD and analyzed with one-way ANOVA followed by Tukey’s post hoc test.(C–H) Time-course gene expression of inflammatory cytokines, macrophage-polarization markers, and HADHB in THP-1-derived macrophages transfected with siMALAT1 or siNC and stimulated with LPS at the indicated time points. For each LPS stimulation time point, n = 9 independent experiments (independent cell cultures and stimulated on different days); data are shown as mean ± SD and analyzed with two-way ANOVA followed by Tukey’s post hoc test.(I) Western blot showing HADHB protein levels in THP-1-derived macrophages after siMALAT1 transfection and LPS stimulation at the indicated time points; adjacent quantification shows HADHB band intensity normalized to GAPDH (loading control). The experiment was repeated n = 3 independent experiments (independent cell cultures on different days); data are shown as mean ± SD and analyzed with one-way ANOVA followed by Tukey’s post hoc test.

When we measured changes in the gene expression levels of macrophage polarization markers during inflammation, we found that MRC1 transcript levels increased significantly 24 h after LPS stimulation. By contrast, NOS2 declined after stimulating by LPS, but then increased thereafter (Figures S4D and S4E).

In BMDMs, we profiled the same markers at multiple time points after LPS stimulation and observed temporal patterns that mirrored those observed in PMA-differentiated THP-1-derived macrophages (Figures S4F–S4H).

The inverse relationship between MALAT1 and pro-inflammatory cytokines such as TNFA, coupled with its temporal concordance with MRC1 from 12 to 24 h, suggests that MALAT1 functions as a protective regulator during the late phase of LPS stimulation by suppressing pro-inflammatory macrophages’ activation.

We next examined the effects of MALAT1 knockdown on cytokines and polarization markers’ expression. Compared with the negative control group, we found that MALAT1 knockdown significantly increased transcript levels of pro-inflammatory cytokines such as TNFA and IL1B, while it reduced those of the anti-inflammatory cytokine TGFB. Consistent with a shift in polarization, the transcript level of MRC1 was downregulated and NOS2 was upregulated in MALAT1 knockdown cells. Additionally, the gene expression of HADHB decreased after MALAT1 knockdown (Figures 3C–3H).

We performed western blotting to assess HADHB protein levels in MALAT1 knockdown cells following LPS stimulation at multiple time points. We observed reduced HADHB protein in the 24 h after LPS stimulation (Figures 3I and S5A). These findings are consistent with our previous results, which showed reduced HADHB transcripts under the same conditions (Figure 3H).

In RAW 264.7 macrophages and BMDMs, the effects of mouse Malat1 on inflammatory responses and macrophage polarization mirrored those observed in THP-1-derived macrophages (Figures S6A–S6D and S7A–S7C).

Conversely, in the stable MALAT1-overexpressing THP-1 cells generated by lentiviral transduction, we observed the opposite pattern: transcript levels of pro-inflammatory cytokines, such as IL1B and TNFA, reduced relative to the negative control, whereas the anti-inflammatory cytokine TGFB increased. Consistent with a shift toward an anti-inflammatory polarization state, the transcript level of MRC1 increased and NOS2 decreased. As expected, the transcript level of HADHB was elevated in the overexpression group (Figures S7D–S7G).

To optimize HADHB silencing in PMA-differentiated THP-1-derived macrophages, we evaluated two independent siRNAs (siHADHB-1 and siHADHB-2). The gene expression results showed that siHADHB-2 achieved superior knockdown (Figure S7H), so it was used in subsequent experiments. A time-course western blot revealed that HADHB protein levels began to decline on day 4 post-transfection (Figure S7I); accordingly, LPS stimulation was initiated on day 4 for downstream assays. Following siRNA-mediated HADHB knockdown and LPS stimulation at the indicated time points, transcript levels of pro-inflammatory cytokines IL1B and TNFA were elevated relative to the negative control, whereas TGFB was reduced. Consistent with a shift toward pro-inflammatory polarization, the transcript level of MRC1 decreased while NOS2 increased (Figures 4A and 4B). Notably, this phenotype recapitulated that which we observed upon MALAT1 knockdown.Figure 4*MALAT1-*HADHB pathway constrains inflammation and attenuates macrophage pro-inflammatory responses(A and B) Temporal expression profiles of IL1B, TNF, TGFB, MRC1, NOS2, and HADHB in THP-1-derived macrophages transfected with siHADHB or siNC and stimulated with LPS at the indicated time points. For each LPS stimulation time point, n = 3 independent experiments (independent cell cultures and stimulated on different days); data are shown as mean ± SD and analyzed with one-way ANOVA followed by Tukey’s post hoc test.(C–E) Rescue experiment, Oe-HADHB THP-1-derived macrophages were transfected with siMALAT1; figures show LPS-evoked expression changes in inflammatory, polarization, and HADHB markers relative to negative control (siNC). For each LPS stimulation time point, n = 3 independent experiments (independent cell cultures and stimulated on different days); data are shown as mean ± SD and analyzed with one-way ANOVA followed by Tukey’s post hoc test.

We performed a rescue experiment to test whether MALAT1 regulates inflammatory activation and macrophage polarization via HADHB. First, we generated stable HADHB-overexpressing (Oe-HADHB) THP-1 cells using lentiviral transduction; immunoblotting confirmed markedly elevated HADHB proteins relative to the negative control (Figure S7J). We then differentiated THP-1 cells into macrophages, transfected them with siMALAT1, and stimulated them with LPS at the indicated time points. In HADHB-overexpressing cells, concurrent MALAT1 knockdown reversed the siMALAT1-induced pro-inflammatory phenotype: 24 h after LPS stimulation, transcript levels of IL1B and TNF decreased, which TGFB expression increased. Concurrently, with respect to macrophage polarization, MRC1 transcript levels increased, whereas NOS2 decreased. These data indicated that HADHB overexpression rescued the effects of MALAT1 depletion, supporting a model in which HADHB mediates the anti-inflammatory action of MALAT1 (Figures 4C–4E).

Together, these results demonstrate that MALAT1 restrains pro-inflammatory macrophages’ activation via its interaction with HADHB.

MALAT1-mediated global transcriptional reprogramming in macrophage activation

To gain a comprehensive view of MALAT1’s impact on the metabolic reprogramming of macrophage polarization, we performed mRNA sequencing on MALAT1-knockdown-differentiated macrophages derived from THP-1 cells (Figure S8A). After 24 h of LPS stimulation, we identified 35 upregulated and 43 downregulated genes (Figure 5A). Genes differentially expressed after 12 and 48 h of LPS stimulation were listed (Figures S8B and S8C). Furthermore, we performed gene set enrichment analyses (GSEA) for genes with differential expression. GSEA revealed that, after 24 h of LPS stimulation, MALAT1 knockdown significantly upregulated the NF-κB signaling pathway, while the oxidative phosphorylation (OXPHOS) and FAO pathways were significantly downregulated (Figures 5B, 5C, S9, and S10).Figure 5MALAT1 orchestrates global transcriptional and metabolic reprogramming in macrophages(A) Differentially expressed genes (DEGs) identified by RNA-seq in LPS-stimulated macrophages after 24 h. The experiment was repeated; n = 3 independent experiments (independent cell cultures on different days).(B and C) Hallmark gene set scores from gene set enrichment analysis (GSEA). The experiment was repeated; n = 3 independent experiments (independent cell cultures on different days).(D and E) Intracellular and extracellular L-lactate levels in PMA-differentiated macrophages transfected with siMALAT1 or siNC and subsequently stimulated with LPS; measurements were taken at the indicated post-LPS time points. For each LPS stimulation time point, n = 3 independent experiments (independent cell cultures and stimulated on different days); data are shown as mean ± SD and analyzed with two-way ANOVA followed by Tukey’s post hoc test.(F, G, J, and L) The gene expression levels of LDHA, LDHB, ACLY, and CD36 in LPS-stimulated siRNA-transfected THP-1-derived macrophages. For each LPS stimulation time point, n = 9 independent experiments (independent cell cultures and stimulated on different days); data are shown as mean ± SD and analyzed with two-way ANOVA followed by Tukey’s post hoc test.(H, I, and K) The protein expression of LDHA, LDHB, ACLY, and p-ACLY in LPS-stimulated THP-1-derived macrophages transfected with siMALAT1 or siNC at the indicated time points. The experiment was repeated; n = 3 independent experiments (independent cell cultures on different days).

We also performed the Kyoto Encyclopedia of Genes and Genomes enrichment analysis. The results showed that the differentially expressed genes of differentiated macrophages after stimulation for 24 h were involved in the TNF signaling pathway and cytokine-cytokine receptor interaction. In the differential gene enrichment results for 12 h and 48 h after LPS stimulation, we found cells that were related to these pathways (Figure S11).

In addition, we performed a gene ontology enrichment analysis of differential genes 24 h after LPS stimulation and found that differential genes were involved in inflammatory responses. In the enrichment results of differential genes 12 and 48 h after LPS stimulation, we observed that differential genes were related to the inflammatory, immune, and innate immune responses (Figure S12).

Based on these findings, we confirm that MALAT1 plays an anti-inflammatory role by promoting OXPHOS and FAO during the resolution phase of inflammation, processes that are essential for M2 macrophages’ polarization.

Given the distinct metabolic profiles of the classical activated macrophages (M1) and M2 macrophages, L-lactate is higher in M1 macrophages and lower in M2 macrophages. We first measured intracellular and extracellular L-lactate levels at various time points after LPS stimulation of MALAT1-knockdown macrophages. After 24 h of LPS stimulation, the concentration of intracellular L-lactate in macrophages increased significantly, while the extracellular concentration of L-lactate decreased slightly (Figures 5D and 5E). Therefore, we detected transcription changes in lactate dehydrogenase A (LDHA) and lactate dehydrogenase B (LDHB), key enzymes in lactate-pyruvate conversion. Through these results, we confirmed that MALAT1 knockdown resulted in a decreased expression level of LDHB, which promotes the conversion of lactate to pyruvate and an increased expression level of LDHA, which promotes the production of lactate (Figures 5F and 5G). Additionally, we assessed LDHA and LDHB protein levels in MALAT1 knockdown cells following LPS stimulation. The western blot results were consistent with the gene expression changes of LDHA and LDHB, confirming that MALAT1 knockdown influenced the lactate-pyruvate metabolic pathway in this context (Figures 5H, 5I, S5B, and S13).

Based on the above, we suggest that MALAT1 inhibits glycolysis by promoting the conversion of lactate into pyruvate and suppressing the lactate production.

Despite the above findings, according to the mRNA sequencing results, we also found that the expression of ATP-citrate lyase (ACLY), a key enzyme involved in fatty acid synthesis21 in M1 macrophages, was upregulated following MALAT1 knockdown (Figures 5A and S8C). Therefore, we wanted to clarify whether MALAT1 would also inhibit the fatty acid synthesis processes by regulating the expression and activity of ACLY. We found that, after MALAT1 knockdown, ACLY transcription increased (Figure 5J). At the same time, since ACLY activity is mainly positively regulated via phosphorylation,22 we also detected the protein expression of phosphorylated ACLY (p-ACLY) via western blot. This shows that the protein expression of p-ACLY increased significantly after MALAT1 knockdown (Figures 5K and S14).

We also detected the expression of CD36, the fatty acid transporter, which facilitates the uptake of fatty acids into cells,23 and found that the transcription of CD36 remained unchanged (Figure 5L).

Using stable MALAT1-overexpressing THP-1 cells, we profiled the gene expressions described above. Conversely, MALAT1 overexpression decreased the transcript levels of LDHA and ACLY while increasing the transcript levels of LDHB, relative to the negative control (Figures S15A–S15C). In RAW 264.7 cells, the Acly transcript level increased following siMalat1-mediated knockdown, relative to siNC (Figure S6E). CD36 transcript levels were higher than the negative control at 24 h in MALAT1-overexpressing THP-1-derived macrophages after LPS stimulation, whereas no significant difference was observed between groups at 48 h (Figure S15D).

These results suggest that MALAT1 supports FAO and OXPHOS while restraining glycolysis and fatty acid synthesis pathways that drive pro-inflammatory responses.

Discussion

This study was prompted by the question of whether molecular mechanisms analogous to the interaction between the long non-coding RNA (lncRNA) Lncfao and HADHB—previously characterized in mouse models—also exist in human macrophages. In mice, Lncfao has been implicated in promoting FAO and facilitating inflammation resolution.17 However, since no direct homolog of Lncfao had yet been identified in humans, it remained unclear whether another lncRNA might fulfill a similar role. Here, we provide the evidence that, in human macrophages, the HuR-MTCH2 axis promotes MALAT1 trafficking to mitochondria, enabling MALAT1-HADHB binding that negatively regulates the inflammatory response. Specifically, MALAT1 knockdown leads to reduced thiolase activity of HADHB, accompanied by increasing glycolysis and fatty acid synthesis pathways, thereby amplifying inflammatory cytokine production. These findings suggest that the MALAT1-HADHB axis in human macrophages provides a molecular foundation for inflammation resolution, unveiling a previously unrecognized layer of macrophage functional regulation (Figure 6).Figure 6. Mechanistic illustration of MALAT1-mediated regulation in THP-1-derived macrophages during inflammationUnder inflammatory conditions, MALAT1 expression in the nucleus increases and drives the metabolic reprogramming of macrophages through several mechanisms. First, MALAT1 enhances HADHB expression and trans-localizes from the nucleus to mitochondria via the HuR-MTCH2 axis, where it binds HADHB and elevates its thiolase activity to promote FAO. MALAT1 also upregulates LDHB, facilitating the lactate-to-pyruvate conversion, while downregulating LDHA to reduce lactate production and diminish glycolysis. Furthermore, MALAT1 downregulates ACLY expression and lowers phosphorylated ACLY, thereby inhibiting the fatty acid synthesis pathway typically associated with the pro-inflammatory phenotype.

Although Lncfao appears to be mouse-specific and lacks an obvious human ortholog, it is plausible that functional modules—rather than a primary sequence, per se—are conserved across species. Despite limited sequence homology, human MALAT1 harbors conserved structural elements, such as the 3′ triple-helix stability module, which shows strong evidence of structural conservation across vertebrates,24 and RNA G-quadruplexes (rG4s), which have been reported as thermodynamically stable and conserved.25 These features could provide motif-level convergence with murine Lncfao. With this in mind, we hypothesize that MALAT1 may encode structural or short motif features analogous to those in Lncfao that enable coupling to mitochondrial FAO pathways, even if the underlying primary sequences have diverged. This view aligns with the broader principle that lncRNA function can be preserved at the level of secondary structure, short patches, synteny, or RBP-binding modules, despite rapid sequence evolution. This hypothesis will require in silico motif scans and in-cell structure probing in the future.

The activation state of macrophages is closely linked to shifts between glycolysis and FAO. In M1 macrophages, the rate of the tricarboxylic acid cycle is limited, leading to the accumulation of metabolites such as succinic acid and citric acid.26 Mitochondrial citrate carriers (CICs) transport citric acid from mitochondria to the cytoplasm, where ACLY converts it to acetyl-CoA and oxaloacetic acid. In response to inflammatory stimuli such as LPS, the expression of CIC and ACLY are upregulated, maintaining the proinflammatory function of M1 macrophages.27 In general, M1 macrophages stimulated by LPS rely predominantly on aerobic glycolysis, rather than mitochondrial respiration. In contrast, macrophages transitioning toward inflammation resolution and tissue repair—such as M2 macrophages—primarily generate ATP through OXPHOS and FAO, metabolic programs that support their anti-inflammatory role.28^,^29^,^30 This metabolic reprogramming is governed by a complex network of transcription factors and signaling molecules. Recently, lncRNAs have emerged as key regulators of metabolic enzymes, acting not only at the transcriptional level but also through direct RNA-RNA and RNA-protein interactions. MALAT1 is a well-characterized lncRNA, originally studied for its role in cancer cells and endothelial cells, where it has been shown to promote proliferation and metastasis via its high expression.4^,^31 In immunology, recent studies have highlighted the involvement of MALAT1 in macrophages’ inflammatory responses and immune regulation. For instance, MALAT1 has been reported to fine-tune NF-κB signaling and inflammasome activity, and to influence lipid metabolism, including foam cell formation and cholesterol efflux.32 However, its role in metabolic processes such as FAO and OXPHOS during macrophage polarization toward an inflammation-resolving phenotype remained largely unexplored, particularly with respect to its potential physical interaction with HADHB.

Foam cell formation in macrophages and chronic inflammation are essential for the progression of atherosclerotic lesions, and MALAT1 serves as a multifaceted regulator in this process. In Apoe^−/−^ mice with the MALAT1 deficiency in bone marrow-derived cells, atherosclerotic lesions were enlarged and inflammatory cytokine production was enhanced, compared to controls,33 suggesting that MALAT1 deficiency exacerbates atherosclerosis through immune-mediated mechanisms. This finding implies that MALAT1 plays a protective role by suppressing chronic inflammation and slowing plaque progression; however, under short-term oxidized low-density lipoprotein (oxLDL) stimulation, MALAT1 has been shown to promote inflammatory responses and foam cell formation, indicating that its role may vary depending on the stage of atherosclerosis. For instance, in early-stage lesions, MALAT1 induces CD36 and inflammatory genes, facilitating lipid accumulation and inflammation.34^,^35 As inflammation progresses, MALAT1 is upregulated and functions to suppress excessive inflammation—potentially by regulating NF-κB activity and promoting autophagy-mediated cellular protection.36 This biphasic model suggests that MALAT1 exerts stage-dependent effects in atherosclerosis. Regarding its involvement in human cardiovascular diseases, some studies suggest that MALAT1 exacerbates pathological conditions. However, given its apparent biphasic functionality, further clinical research is needed before drawing definitive conclusions.32 In any case, MALAT1 is considered a key factor that modulates the inflammatory and metabolic states of macrophages within atherosclerotic lesions, influencing plaque formation and stability—for example, by regulating cell survival via autophagy and apoptosis, as well as efferocytosis efficiency.35^,^36^,^37^,^38 The MALAT1-HADHB axis elucidated in this study may contribute to future investigations into the dual role of MALAT1 in macrophages.

In this context, our study has suggested that MALAT1 interacts with HADHB and facilitates FAO activation in macrophages. While previous studies have demonstrated MALAT1’s impact on lipid metabolism and inflammation, most of these mechanisms have been attributed to its function as an miRNA sponge, or as a scaffold for transcription factor recruitment.31 In contrast, our findings reveal that the MALAT1-HADHB interaction modulates mitochondrial function directly, thereby enabling macrophages to shift their metabolic state and transition toward inflammation resolution and tissue repair. MALAT1 was primarily detected in the nucleus, but also in mitochondria. The functional significance of its mitochondrial localization remains unclear; however, given its interaction with HADHB, it may play a role in the direct regulation of mitochondrial metabolism.

In summary, this study advances our understanding of the molecular mechanisms underlying inflammation resolution in macrophages by elucidating the role of MALAT1 in HADHB-mediated FAO activation. Our findings suggest that MALAT1 serves as a key regulator of inflammation resolution in human macrophages. These insights provide a foundation for considering lncRNAs as potential therapeutic targets in immune regulation and inflammatory diseases. Further elucidation of macrophage metabolic reprogramming, including the MALAT1-HADHB axis, may pave the way for novel treatment strategies for chronic inflammatory diseases, atherosclerosis, and cancer.

Limitations of the study

Although our data support a functional association between MALAT1 and HADHB in which the modulation of the thiolase activity of HADHB restrains inflammatory responses and pro-inflammatory macrophage activation, the current evidence does not establish a direct, specific RNA-protein interaction. Future studies employing in vitro reconstitution with purified components crosslinking-guided mutational mapping will be required to determine whether MALAT1 binds HADHB directly, to define the interaction sites and structural motifs, and to assess whether bridging factors mediate complex assembly.

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Yuxiang Liu (lyxhmu@gmail.com).

Materials availability

This study did not generate new unique reagents.

Data and code availability

• •The RIP-seq data and RNA-seq data have been made available at the Gene Expression Omnibus (GEO) and are publicly available as of the date of publication. Accession numbers are listed in the key resources table.
• •This article does not report original code.
• •Any additional information required to reanalyze the data reported in this article is available from the lead contact, upon request.

Acknowledgments

We sincerely appreciate Ms. M. Hayashi for their excellent technical assistance. We also express our deep gratitude to Dr. Ryoko Uchida and Dr. Jun-ichi Okata for their scientific oversight during the process of updating the figures, refining the methods section, and preparing the supplementary information. At the same time, we sincerely appreciate Professor Issei Komuro for contributing to the conceptual refinement of the revised discussion section. This study was supported by grants from the Grant-in-Aid for Scientific Research from the 10.13039/501100001691Japan Society for the Promotion of Science, the Japan Agency for Medical Research and Development, and the Japan Science and Technology Agency (24H00631, 22gm6510010h0001, 23ek0210193h0001, 25gm4010028h0001, and JPMJMS2023) to K.F. and the Grant-in-Aid for Early-Career Scientists from the 10.13039/501100001691Japan Society for the Promotion of Science (25K19413) to Y.L. This study was supported by Nanken-Kyoten, Science Tokyo; the Multilayered Stress Diseases program (JPMXP1323015483), Science Tokyo; and the Medical Research Center Initiative for High Depth Omics, Science Tokyo.

Author contributions

Conceptualization, Y.L., Y.N., and K.F.; data curation, Y.L., Y.N., and J.S.; methodology, Y.L., Y.N., J.S., I.M., and K.F.; writing – original draft, Y.L.; writing – review and editing, Y.N., Y.I., and K.F.; visualization, J.S., T.O., K.K., A.K., and N.S.; funding acquisition, K.F. and Y.L.

Declaration of interests

All authors declare that they have no competing interests.

STAR★Methods

Key resources table

REAGENT or RESOURCESOURCEIDENTIFIERAntibodiesPolyclonal Rabbit anti-Human HADHB AntibodyLSBio (LifeSpan)Cat# LS-B4668-50; RRID: AB_10801829Rabbit Anti-GAPDH Monoclonal Antibody, Unconjugated (Clone 14C10)Cell Signaling TechnologyCat# 2118; RRID:AB_561053Polyclonal Rabbit anti-Human HADHB AntibodyLSBio (LifeSpan)Cat# LS-B4668-50; RRID: AB_10801829ATP-Citrate Lyase AntibodyCell Signaling TechnologyCat# 4332; RRID:AB_2223744Rabbit Anti-GAPDH Monoclonal Antibody, Unconjugated (Clone 14C10)Cell Signaling TechnologyCat# 2118; RRID:AB_561053Phospho-ATP-Citrate Lyase (Ser455) AntibodyCell Signaling TechnologyCat# 4331; RRID:AB_2257987ATP-Citrate Lyase AntibodyCell Signaling TechnologyCat# 4332; RRID:AB_2223744Rabbit Anti-LDHA Monoclonal Antibody, Unconjugated (Clone C4B5)Cell Signaling TechnologyCat# 3582; RRID:AB_2066887Phospho-ATP-Citrate Lyase (Ser455) AntibodyCell Signaling TechnologyCat# 4331; RRID:AB_2257987Lactate Dehydrogenase antibody [EP1565Y]AbcamCat# ab53292; RRID:AB_2234531Rabbit Anti-LDHA Monoclonal Antibody, Unconjugated (Clone C4B5)Cell Signaling TechnologyCat# 3582; RRID:AB_2066887Anti-rabbit IgG, HRP-linked AntibodyCell Signaling TechnologyCat# 7074; RRID:AB_2099233Lactate Dehydrogenase antibody [EP1565Y]AbcamCat# ab53292; RRID:AB_2234531Mouse Anti-rabbit IgG (Conformation Specific) (L27A9) mAb (HRP Conjugate)Cell Signaling TechnologyCat# 5127; RRID:AB_10892860Anti-rabbit IgG, HRP-linked AntibodyCell Signaling TechnologyCat# 7074; RRID:AB_2099233Mouse Anti-rabbit IgG (Conformation Specific) (L27A9) mAb (HRP Conjugate)Cell Signaling TechnologyCat# 5127; RRID:AB_10892860MTCH2 antibodyProteintechCat# 16888-1-AP; RRID: AB_2266733Anti-HuR/ELAVL1 antibody [EPR17397]AbcamCat# ab200342; RRID: AB_2784506Chemicals, peptides, and recombinant proteinsFAOBlue (Fatty Acid Oxidation Detection Reagent)FunakoshiCat# FDV-0033Lipopolysaccharides (LPS)SigmaCat# L5293Phorbol-12-myristate-13-acetate (PMA)SigmaCat# P1585Restore Plus Western Blot Stripping BufferThermoscientificCat# 46430RPMI-1640GibcoCat# 11875093PBS 1xSigmaCat# D8537-500MLFBSCytivaCat# SH30396Penicillin-Streptomycin (10,000 U/mL)GibcoCat# 15140122RIPA Lysis and EXtraction BufferThermoscientificCat# 89900Halt Protease and Phosphatase Inhibitor CocktailThermoscientificCat# 78440Novex Sharp Pre-Stained Protein StandardNovexCat# LC5800MagicMark XP Western Protein StandardNovexCat# LC5602Lane Marker Sample BufferThermoscientificCat# 39000Clarity Max Western ECL SubstrateBioRadCat# 1705061Propidium Iodide (1.0 mg/mL Solution in Water)InvitrogenCat# P1304MPTransfer BufferBioRadCat# 161-0734MethanolWakoCat# 137-01823Nitrocellulose Blotting MembraneCytivaCat# 10600124EveryBlot Blocking BufferBioradCat# 12010020TBS-T Buffer 20xThermoscientificCat# 28360HiPerFect Transfection ReagentQiagenCat# 301705DMEM/F-12GibcoCat# 11320033DMEM, high glucoseGibcoCat# 11965092Recombinant Mouse M-CSF (carrier-free)BiolegendCat# 576406Lenti-X ConcentratorTakaraCat# 631232Lipofectamin 2000ThermofisherCat# 11668019Critical commercial assaysRNeasy Mini KitsQiagenCat# 74106SuperScript III First-Strand Synthesis SystemInvitrogenCat# 18080051Magna RIP RNA-Binding Protein Immunoprecipitation KitMerckCat# 17-700Takara BCA Protein Assay KitTakaraCat# T9300APARIS KitInovitrogenCat# AM1921Qproteome Mitochondria Isolation KitQiagenCat# 37612NEBNext Ultra II RNA Library Prep KitNEBCat# E7770LL-lactate Assay Kit (Colorimetric)AbcamCat# ab65331Deproteinizing Sample Preparation Kit-TCAAbcamCat# ab204708MycoFluor Mycoplasma Detection KitThermo FisherCat# M7006Deposited dataRIP-seq dataGene Expression Omnibus[Database]: [GSE281530](GSE281530)RNA-seq dataGene Expression Omnibus[Database]: [GSE281531](GSE281531)Experimental models: Cell linesTHP-1ATCCTIB-202; RRID: CVCL_0006RAW 264.7ATCCTIB-71; RRID: CVCL_0493BMDMFreshly IsolatedIn MouseKupffer CellsFreshly IsolatedIn Mouse293TATCCCRL-3216; RRID:CVCL_0063Experimental models: Organisms/strainsMouse: WT C57BL/6NJOriental Yeast Company (Tokyo, Japan)In Bred; RRID: IMSR_JAX:005304OligonucleotidesSilencer Select Pre-desigened siRNA of MALAT1:siMALAT1#1, 5′- CCGCUGCUAUUAGAAUGCATT-3′ (sense)AmbionsiRNA ID: n511399Silencer Select Pre-desigened siRNA of MALAT1:siMALAT1#1, 5′- UGCAUUCUAAUAGCAGCGGGA-3′ (antisense)AmbionsiRNA ID: n511399Silencer Select Pre-desigened siRNA of MALAT1:siMALAT1#2, 5′- GGCUUAUACUCAUGAAUCUTT-3′ (sense)AmbionsiRNA ID: n272231Silencer Select Pre-desigened siRNA of MALAT1:siMALAT1#2, 5′- AGAUUCAUGAGUAUAAGCCTG-3′ (antisense)AmbionsiRNA ID: n272231Silencer Select Negative Control#1 siRNAAmbionCat# 4390843Silencer Select Pre-desigened siRNA of Malat1: siMalat1#1, 5′- CAGCUAGCAUGUGAUGUAATT-3′ (sense)AmbionsiRNA ID: n520782Silencer Select Pre-desigened siRNA of Malat1: siMalat1#1, 5′- UUACAUCACAUGCUAGCUGCT-3′ (antisense)AmbionsiRNA ID: n520782Silencer Select Pre-desigened siRNA of Malat1: siMalat1#2, 5′- GUAUUGUAUCGAGACCAAATT-3′ (sense)AmbionsiRNA ID: n4392420Silencer Select Pre-desigened siRNA of Malat1: siMalat1#2, 5′- UUUGGUCUCGAUACAAUACTG-3′ (antisense)AmbionsiRNA ID: n4392420Silencer Select Pre-desigened siRNA of HADHB: siHADHB#1, 5′- CUAAGGAAGUAGUUGAUUATT-3′ (sense)AmbionsiRNA ID: s6444Silencer Select Pre-desigened siRNA of HADHB: siHADHB#1, 5′- UAAUCAACUACUUCCUUAGGG-3′ (antisense)AmbionsiRNA ID: s6444Silencer Select Pre-desigened siRNA of HADHB: siHADHB#2, 5′- GGAACAGGAUGAAUAUGCATT-3′ (sense)AmbionsiRNA ID: s6443Silencer Select Pre-desigened siRNA of HADHB: siHADHB#2, 5′- UGCAUAUUCAUCCUGUUCCAG-3′ (antisense)AmbionsiRNA ID: s6443Recombinant DNApsPAX2Addgene#12260; RRID: Addgene_12260pMD2.GAddgene#12259; RRID: Addgene_12259pCDH-EF1-copGFP-T2A-PuroAddgene#72263; RRID: Addgene_72263pCDH-hMALAT1Addgene#118580; RRID: Addgene_118580pLV[Exp]-EGFP:T2A:Puro-SFFV>hHADHB[NM_000183.3]VectorBuilder#VB250724-1701qfqpLV[Exp]-EGFP/Puro-EF1A>ORF_StufferVectorBuilder#VB010000-9493jsySoftware and algorithmsGraphPad Prismhttps://www.graphpad.com/featuresversion 10.0BiorenderBiorenderwww.biorender.comRR Core Teamhttps://www.r-project.org/GSEAhttp://www.gsea-msigdb.orgversion 4.3.3KEGGhttp://www.genome.jp/kegg/version 4.12.6GOhttps://geneontology.org/docs/go-enrichment-analysis/version 2.56.0FlowJo softwareFlowJohttps://www.flowjo.com/

Experimental model and study participant details

Animals

C57BL/6NJ female mice were purchased from Oriental Yeast company (Tokyo, Japan) and raised at the Laboratory Animal Resource Center of Nippon Medical School (Tokyo, Japan). They were bred and maintained under pathogen-free conditions with free access to food and water. Mice were ready at the age of 6∼8 weeks. All experiments were meticulously conducted and approved by the Nippon Medical School Laboratory Animal Care and Use Committee (the ethical approval reference number: 2022-12).

Cell culture

Human THP-1 cells and RAW 264.7 cells

Human THP-1 cell lines and Raw 264.7 cell lines were purchased from the American Type Culture Collection (ATCC, TIB-202, TIB-71). THP-1 cells were cultured in RPMI 1640 and differentiated into macrophages by treatment with 100 ng/mL phorbol-12-myristate-13-acetate (PMA) for 48 hours, and they were then cultured in fresh medium to rest for 24 hours prior to further treatment.

The RAW 264.7 cells were cultured in Dulbecco’s Modified Eagle medium (DMEM). To induce the inflammation response, the cells were treated with LPS (100 ng/mL) for different hours. All the cell media was added by 10% (vol/vol) of fetal bovine serum (FBS) (Gibco) and 1% penicillin-streptomycin (Thermo Fisher).

All cell lines were rigorously screened for mycoplasma contamination using the MycoFluor Mycoplasma Detection Kit (Thermo Fisher) and were cultured at 37 °C, 5% CO_2_ in the fully humidified air. Cells were sub-cultured every 3-4 days to maintain optimal growth conditions. THP-1 cell line was authenticated by STR profiling, and both cell lines were used at low passage after thawing.

Bone marrow derived macrophages (BMDMs)

Bone marrow cells were flushed from femurs and tibias of male mice between 6 and 8 weeks and were grown in DMEM/F12 (Thermo Fisher) supplemented with 10% (vol/vol) of FBS (Gibco), 1% penicillin–streptomycin (Thermo Fisher), 1% L-glutamine (Thermo Fisher), and 40 ng/mL of recombinant mouse M-CSF (BioLegend) for 7 days. The medium was changed every 3 d. Differentiated BMDMs were detached from plates using the cell scraper and replated into 24-well culture plate at a density of 1 × 10^5^ cells per well prior to cell transfection. For LPS treatment, 100 ng/mL of LPS was added to the medium after transfection.

Liver Kupffer cells (KCs)

KCs were generated by isolating liver non-parenchymal cells (LNPCs) from the mouse liver as described.39 Briefly, the mouse liver was digested by digestion medium and centrifuged by at 50 rcf for 3 min. The supernatant was then centrifuged by 400 rcf for 5 min to deposit the LNPCs. After removing the red blood cells by ACK buffer and counting the cell number, resuspend the LNPCs to a density of 1-3 × 10^7^/mL in RPMI medium contained 10% FBS (Gibco), 1% penicillin–streptomycin (Thermo Fisher) and 1% L-glutamine (Thermo Fisher). Incubate the 6-well culture plate contained 1 × 10^7^ LNPCs per well for 2 hours at 37°C with 5% CO_2_. After incubation for 2 hours, remove the cell debris and non-adherent cells by cold PBS, continue to add fresh complete medium and 50 ng/mL M-CSF (BioLegend) for further culture.

Method details

Purification of RNA and real–time PCR

Total RNA was purified from cells using RNeasy Mini kits (Qiagen) according to the manufacturer’s instructions. RNA purity and concentration were evaluated using a Nanodrop (Thermo Fisher Scientific) before proceeding with subsequent experiments. Total RNA was reverse transcribed into cDNA using SuperScript III (Thermo Fisher). Quantitative real-time PCR analyses were conducted using a Lightcycler 480 system (Roche), with GAPDH serving as an internal control of human genes and Gapdh or Actb serving as an internal control of mouse genes.

For qPCR, the following primer sequences were used for human genes: COX2 was amplified using the forward primer TACCTGTCCTGCGTGTTGAA and the reverse primer TCTTTGGGTAATTTTTGGGATCT. HADHB was amplified using the forward primer CAGCCTCTTCTCCTTCCTGAT and the reverse primer GCCAGAGGGCTGATTAGAGA. MRC1 was amplified using the forward primer ACTACTACGCCAAGGAGGTCAC and the reverse primer TGCTTGAACTTGTCATAGATTTCG. NOS2 was detected using the forward primer TTGGACCTGAACCTTGCTCC and the reverse primer ACAGCTCAGAGGTCTTATGGAAAA. LDHA was amplified using the forward primer CAGCGCTTGTGATCTTCATT and the reverse primer TACCCCTGCTCCTGGTTTTT. LDHB was detected using the forward primer CATCCTCTTTGCGACAGAGAC and the reverse primer GCAGCTCAGCCTGTACTTATC. ACLY expression was analyzed with the forward primer TTGACCTACGTGGCTTGGAAG and the reverse primer GGTAACGGAATCGGGCTGAAT. CD36 was detected using the forward primer CCTCAGATCGTCAAGTACAGTCC and the reverse primer ATCACGCGGTGTTTGGGTAAT. IL1B expression was determined using the forward primer ACTTCGGCAGAGACAGGTAG and the reverse primer CCCGAGCATACTTGAACCGA. TNF expression was analyzed using the forward primer CGGCTGCAGGTCAACCTATT and the reverse primer CACCAATGGTCCCAGTCTCA. TGFB was amplified using the forward primer TGCGTAATGGAAAGTAAAGCCC and the reverse primer CAAACACCTCACAAAACCCCC. MALAT1 was detected using the forward primer AGCCACATCGCTCAGACAC and the reverse primer GCCCAATACGACCAAATCC. GAPDH was amplified using the forward primer CTCGCTTCGGCAGCACAT and the reverse primer TTTGCGTGTCATCCTTGCG. Finally, U6 expression was assessed using the forward primer GCTGTCCCCACATTAGGCTT and the reverse primer ACCGTAGTATACCCCCGGTC.

The following primer sequences were used for mouse genes: Malat1 was amplified using the forward primer ATAGCCCAGGAAAGAGTGCG and the reverse primer GCTTCACCACCACATCCGTA. Il1b was amplified using the forward primer TGCCACCTTTTGACAGTGATG and the reverse primer ATGTGCTGCTGCGAGATTTG. Tnf was amplified using the forward primer TAGCCCACGTCGTAGCAAAC and the reverse primer ACAAGGTACAACCCATCGGC. Tgfb was amplified using the forward primer ACGTCACTGGAGTTGTACGG and the reverse primer TTTGGGGCTGATCCCGTTG. Mrc1 was amplified using the forward primer GGATGGCTCTGGTGTGGAAC and the reverse primer TCTCGCTTCCCTCAAAGTGC. Arg1 was amplified using the forward primer CGTAGACCCTGGGGAACACTAT and the reverse primer TCCATCACCTTGCCAATCCC. Nos2 was amplified using the forward primer GGTGAAGGGACTGAGCTGTT and the reverse primer ACGTTCTCCGTTCTCTTGCAG. Hadhb was amplified using the forward primer AGGCAGATTTCAGAATGACTACCA and the reverse primer ACTTGGTCTGGACAGCTGGG. Acly was amplified using the forward primer ATCGACTCCAGCACCCAGTA and the reverse primer ACTTGGGACTGAATCTTGGGG. Gapdh was amplified using the forward primer ACCACAGTCCATGCCATCAC and the reverse primer TCCACCACCCTGTTGCTGTA. Actb was amplified using the forward primer CACTGTCGAGTCGCGTCC and the reverse primer TCATCCATGGCGAACTGGTG.

Human MALAT1, mouse Malat1 and human HADHB knockdown

Macrophages differentiated from THP-1 cells or BMDMs or RAW 264.7 cells were transfected with siRNAs by means of HiPerFect Transfection Reagent (Qiagen) according to the manufacturer’s instructions.

Briefly, 24 hours before transfection, 1 × 10^5^ THP-1 cells or BMDMs or RAW264.7 cells per well were seeded in the 24-well plates. The next day, 100 ng/mL PMA was added to the THP-1 cells to differentiate into macrophages for 48 hours. For silencing the gene expression by means of transfecting siRNAs, the siRNAs used for knocking down the expression of human MALAT1, mouse Malat1 and human HADHB, which were purchased from Thermo Fisher Scientific. After 6 hours for transfection, the cells were added 400 μL culture medium containing serum, and the cells were continued incubating for 24 hours for analyzing gene silencing. The gene expression of successfully transfected cells was analyzed by real-time quantitative PCR (RT–qPCR), and the successful transfected cells were split and plated for subsequent experiments. The sequences of siRNAs were as follows:

Human: siMALAT1#1, 5′- CCGCUGCUAUUAGAAUGCAtt-3′ (sense) and 5′- UGCAUUCUAAUAGCAGCGGga-3′ (antisense). siMALAT1#2, 5′- GGCUUAUACUCAUGAAUCUtt-3′ (sense) and 5′- AGAUUCAUGAGUAUAAGCCtg-3′ (antisense). siHADHB#1, 5′- CUAAGGAAGUAGUUGAUUAtt-3′ (sense) and 5′- UAAUCAACUACUUCCUUAGgg-3′ (antisense). siHADHB#2, 5′- GGAACAGGAUGAAUAUGCAtt-3′ (sense) and 5′- UGCAUAUUCAUCCUGUUCCag-3′ (antisense).

Mouse: siMalat1#1, 5′- CAGCUAGCAUGUGAUGUAAtt-3′ (sense) and 5′- UUACAUCACAUGCUAGCUGct-3′ (antisense). siMalat1#2, 5′- GUAUUGUAUCGAGACCAAAtt-3′ (sense) and 5′- UUUGGUCUCGAUACAAUACtg-3′ (antisense).

Construction of Lentivirus and stable cell line

Human MALAT1 (hMALAT1) -overexpressing plasmid (Addgene plasmid, pCH-hMALAT1, #118580) and negative control plasmid (Addgene plasmid, pCDH-EF1-copGFP-T2A-Puro, #72263) were purchased from Addgene Company (USA). Human HADHB (hHADHB) -overexpressing plasmid (pLV[Exp]-EGFP:T2A:Puro-SFFV>hHADHB[NM_000183.3]) and negative control plasmid (pLV[Exp]-EGFP/Puro-EF1A>ORF_Stuffer) were purchased from Vector Builder Inc. 293T cells (ATCC) in a good growth state were taken and cultured in 10 cm dishes.

293T cells were transfected at 30%∼40% confluence using Lipofectamine 2000 (Thermo Fisher) following the manufacturer’s instructions. For each transfection, plasmids encoding the hMALAT1 or hHADHB overexpression transfer vector (X), the packaging plasmid (psPAX2; Y), and the envelope plasmid (pMD2.G; Z) were mixed at a mass ratio of 4:3:1 in Opti-MEM (1 mL) (Gibco) to prepare solution B. Lipofectamine 2000 (40 μL) was diluted in Opti-MEM (1 mL) to prepare solution A. After gentle mixing, solution B was added to solution A, incubated for 20 min at room temperature, and the complexes were applied dropwise to 293T cells. After transfection for 12 hours, 293T cells were changed to the fresh medium for further culture.

At 36 hours and 60 hours post-transfection, viral supernatants were separately collected and concentrated by Lenti-X Concentrator (Takara) following the manufacturer’s instructions. The concentrated virus was used to transduce human THP-1 cells. Transduction efficiency was assessed by fluorescence microscopy. The transduced cells were selected by puromycin, and overexpression efficiency was quantified by RT-qPCR or western blotting.

RNA immunoprecipitation (RIP)

RIP was performed with the Magna RIP RNA-Binding Protein Immunoprecipitation Kit (Millipore) according to the manufacturer’s instructions.

Briefly, differentiated macrophages induced by THP-1 cells via PMA on the plates were washed twice with ice-cold PBS. Pre-mixed complete RIP lysis buffer was added to the cells. The lysate was dispensed into nuclease-free microcentrifuge tubes and 5 μg of HADHB or 5 μg of HuR or 5 μg of MTCH2 and rabbit IgG (Millipore) antibodies were used for the following experiment. To prepare the immunoprecipitation of RNA-binding protein-RNA complexes, pre-mixed RIP Immunoprecipitation Buffer was added to each tube. Next, 100 μL of the supernatant of RIP lysates was removed and added to each beads-antibody complex in RIP Immunoprecipitation Buffer with rotating at 4°C overnight. At the same time, 10 μL of the supernatant of RIP lysates was removed to a new tube labelled Input and stored at -80°C. In addition, an additional 10 μL of the supernatant of RIP lysates was transferred to another tube for the next Western Blotting experiment. Then, 50 μL each out of 500 μL of the beads’ suspension during the last wash was removed to new tubes for the next Western Blotting experiment.

To purify the RNAs, each immunoprecipitation was re-suspended in 150 μL of proteinase K buffer and incubated at 55°C for 30 minutes with shaking to digest the protein. Next, RNeasy Mini Kit (Qiagen) was applied to purify the RNAs according to the manufacturer’s instructions. The purity and concentration of extract RNAs were assessed by means of Nanodrop in prior to the subsequent experiments.

Isolation of nuclear, cytoplasmic and mitochondrial RNA

PARIS Kit (Invitrogen) was used for isolation of nuclear and cytoplasmic RNAs and Qproteome Mitochondria Isolation Kit (Qiagen) was used for isolation of mitochondrial RNAs according to the manufacturer’s instructions.

First, human THP-1 cells were differentiated into macrophages, and the differentiated macrophages were stimulated by LPS. Then the macrophages were washed once in PBS. After the centrifuge, the cells were gently resuspended in 500 μL ice-cold Cell Fractionation Buffer (Invitrogen) by pipetting. After the centrifuge, the supernatant was transferred to a new RNase-free tube and mixed well with pre-warmed 2X Lysis/Binding Solution (Invitrogen). The remained cell pellets were washed with ice-cold Cell Fraction Buffer (Invitrogen), and the tubes were centrifuged for 1 minute at 500 g and 4°C. To obtain the nuclear lysates, 500 μL ice-cold Cell Disruption Buffer (Invitrogen) was added to the lysates, and the lysates were pipetted vigorously to lyse the nuclei, and pre-warmed 2X Lysis/Binding Solution (Invitrogen) was added to the homogeneous lysates. Subsequently, 500 μL 100% ethanol (Fujifilm) was added to the nuclear and cytoplasmic lysates and the sample mixtures were transferred to the Filter Cartridges (Invitrogen). After the centrifuge, 700 μL pre-warmed Wash Solution I (Invitrogen) was added, and the cartridges were washed twice with 500 μL Wash Solution 2/3 (Invitrogen) and centrifuged. To elute the RNAs, two sequential aliquots of 40 μL and 10 μL pre-heated Elution Solution (Invitrogen) were added to the cartridges. The purity and concentration of extract nuclear and cytoplasmic RNAs were assessed by means of Nanodrop in prior to the subsequent experiments.

For isolation mitochondria RNAs, human THP-1 cells were differentiated into macrophages, and the differentiated macrophages were stimulated by LPS. The cells were washed once in 1 mL of NaCl solution (Otsuka Pharmaceutical). After centrifuging the tubes, the cell pellets were resuspended in 2 mL ice-cold Lysis Buffer (Qiagen). After that, the lysates were centrifuged, and the pellets were resuspended in 1.5 mL ice-cold Disruption Buffer (Qiagen). To complete cell disruption, samples were aspirated with a blunt-ended syringe for 10 times. After centrifuging the new tubes at 6000 g for 10 minutes at 4°C, the mitochondrial pellets were washed with 1 mL Mitochondria Storage Buffer (Qiagen) and the tubes were centrifuged at 6000 g for 20 minutes at 4°C. Subsequently, the mitochondrial pellets were resuspended in 350 μL Buffer RLT (Qiagen) and mitochondrial RNAs were extracted by RNeasy Mini Kit (Qiagen). The purity and concentration of extract mitochondrial RNAs were also assessed by means of Nanodrop in prior to the subsequent experiments.

Western blotting

The sample of protein were extract from THP-1-derived macrophages after stimulating by LPS, which supplemented with RIPA Lysis Buffer (Thermoscientific), proteinase and phosphatase inhibitor mixture (Thermoscientific) and sample buffer (Thermoscientific). Then, the protein samples were boiled. After which the proteins were separated by SDS/PAGE, transferred to a nitrocellulose blotting membrane (GE Healthcare), blocked with Blocking Buffer (BioRad), and incubated with primary antibodies at 4 °C overnight. Proteins were visualized using HRP-linked secondary antibody and Clarity Max Western ECL Substrates (BIORAD) and then captured using C-Digit (LI-COR). The same membrane was then stripped using Stripping Buffer (Thermoscientific) and sequentially incubated with other primary antibodies, as multiple proteins were targeted.

Fluorescence activated cell sorting (FACS)

The macrophages incubated FAOBlue were treated with trypsin and transferred to the centrifuge tubes, and the tubes were centrifuged. After discarding the supernatant, the macrophages were diluted to 10^6^ single cells with FACS Buffer, and 2 μg/mL Propidium Iodide (Life Technologies) was added to each tube. After adjusting the parameters of the machine, the blue fluorescence was measured and cells with blue fluorescence were collected by FACS machine (BD FACSAria II). The blue fluorescence signal was quantified by mean fluorescence intensity (MFI) signal.

RNA sequencing

RNA samples obtained through RIP were amplified using the NEBNext Ultra II RNA Library Prep Kit (New England Biolabs), and cDNA libraries were prepared following the manufacturer's protocols. The libraries were sequenced using paired end reads on the Illumina HiSeq 1500 (Illumina). Sequencing reads were aligned to the human genome (hg38) using STAR (version 2.7.11).40 MACS2 (version 2.2.9.1) was used to call peaks in RIP-seq data. Gene set enrichment analysis (GSEA, version 4.3.3), KEGG enrichment analysis (version 4.12.6) and Gene Ontology Enrichment Analysis (version 2.56.0). was conducted41 using rank files generated from expression data processed with DESeq2 (version 1.44.0).42 Genes with a P-value less than 0.1 and an absolute log2 fold change greater than 0.5 were considered significantly differentially expressed.

FAO activity detection

The FAO activity was measured by Fatty Acid Oxidation Detection Reagent (FAOBlue, Funakoshi). First, human THP-1 cells were differentiated into macrophages by PMA. Then the differentiated THP-1 cells or BMDMs or RAW 264.7 cells or KCs were washed with PBS twice and added pre-mixed 5 μM FAOBlue-containing RPMI medium (no phenol red). After incubating the cells at 37 °C for 2 hours, the cells were washed with PBS, and the blue fluorescence was detected by FACS.

Measurement of L-lactate concentration

The intracellular and extracellular concentration of L-lactate were detected by L-lactate Assay kit (Colorimetric, ab65331) according to the manufacturers’ protocols. Briefly, the differentiated THP-1 cells were transfected by siRNAs and then stimulated by LPS. The cultured medium was used to detect the extracellular concentration of L-lactate, and the cell lysates were used to detect the intracellular concentration of L-lactate. Then, all the samples were removed the enzyme by using Deproteinizing Sample Preparation kit-TCA (ab204708). After preparing the standard samples, all the samples were mixed with Reaction Mix and measure output on a microplate reader (BIO RAD) at OD 450 nm.

Quantification and statistical analysis

The significance of differences in the experimental data was determined using GraphPad Prism 10.0 software (GraphPad Software Inc.). All data involving statistics were presented as mean ± SD. Experiments were repeated at least three times to ensure reproducibility and used two-tailed unpaired Student’s t-tests, one-way ANOVA or two-way ANOVA, which was used the Tukey's post hoc test, the Sidak's post hoc test or the Dunnett's post hoc test. The number of replicates and the statistical test used were described in the figure legends.