Macrophage-Specific Lactate Dehydrogenase Expression Modulates Inflammatory Function In Vitro

Abstract

Key Points

• Lactate dehydrogenase A deletion alters macrophage function.

• Lactate dehydrogenase A could serve as a potential therapeutic target in AKI.

Background

In AKI, macrophages play a major role in regulating inflammation. Classically activated macrophages (M1) undergo drastic metabolic reprogramming during their differentiation and upregulate the aerobic glycolysis pathway to fulfill their proinflammatory functions. NAD⁺ regeneration is crucial for the maintenance of glycolysis, and the most direct pathway by which this occurs is through the fermentation of pyruvate to lactate, catalyzed by lactate dehydrogenase A (LDHA). Our previous study determined that LDHA is predominantly expressed in the proximal segments of the nephron in the mouse kidney and increases with hypoxia. This study investigates the potential of LDHA as a therapeutic target for inflammation by exploring its role in macrophage function in vitro.

Methods

Bone marrow–derived macrophages (BMDMs) were isolated from myeloid-specific LDHA knockout mice derived from crossbreeding LysM-Cre transgenic mice and LDHA floxed mice. RNA sequencing and LC-MS/MS metabolomics analyses were used in this study to determine the effect of LDHA deletion on BMDMs after stimulation with IFN-γ.

Results

LDHA deletion in IFN-γ BMDMs resulted in a significant alteration of the macrophage activation and functional pathways and change in glycolytic, cytokine, and chemokine gene expression. Metabolite concentrations associated with proinflammatory macrophage profiles were diminished, whereas anti-inflammatory–associated ones were increased in LDHA knockout BMDMs. Glutamate and amino sugar metabolic pathways were significantly affected by the LDHA deletion. A combined multiomics analysis highlighted changes in Rap1 signaling, cytokine–cytokine receptor interaction, focal adhesion, and mitogen-activated protein kinase signaling metabolism pathways.

Conclusions

Deletion of LDHA in macrophages results in a notable reduction in the proinflammatory profile and concurrent upregulation of anti-inflammatory pathways. These findings suggest that LDHA could serve as a promising therapeutic target for inflammation, a key contributor to the pathogenesis of AKI.

Visual Abstract

Introduction

Macrophages are vital for both innate and adaptive immunity, supporting organ function throughout the body. Their metabolic plasticity enables them to play key roles in acute and chronic inflammation, as well as tissue repair and regeneration.¹ Therefore, macrophage flexibility in a wide variety of microenvironments is essential for both the maintenance of homeostasis and optimal responses during pathophysiological conditions.

Macrophages are broadly classified into monocyte-derived macrophages from bone marrow and tissue-resident macrophages from yolk sac progenitor cells.² They exhibit various intermediate phenotypes, but are generally categorized into M1, which drives inflammation, and M2, which aids in tissue remodeling and immune regulation.^3,4 The M1 phenotype, induced by proinflammatory cytokines such as IFN-γ and toll-like receptor agonists, activates macrophages for antigen presentation, cytokine production (IL-1ß, IL-6, IL-12, IL-18), and phagocytosis.^5,6 M1 macrophages facilitate phagocytosis of foreign material and pathogens and eliminate tumor cells, although these may impair tissue regeneration and wound healing because of proinflammatory action.⁷ M2 macrophages, induced by cytokines (e.g., IL-4), glucocorticoids, and immune complexes, support angiogenesis and tissue remodeling.⁸ Notably, the transition from the M1 to M2 phenotype is essential for the termination of inflammatory responses and subsequent regeneration.^9,10 Macrophage polarization is associated with significant metabolic alterations, in particular, glucose metabolism and mitochondrial bioenergetics.^11,12 ATP is produced by glycolysis and oxidative phosphorylation (OXPHOS) as the primary source of energy. In the cytosol, glycolysis breaks down glucose rapidly into pyruvate while generating small amounts of ATP. By contrast, OXPHOS produces large quantities of ATP in the mitochondria by oxidizing pyruvate, as well as other sugars, fats, and amino acids, at a slower rate.¹³ The OXPHOS pathway depends on oxygen and is inactive under anaerobic conditions, shifting cells to rely on glycolysis and rapid pyruvate-to-lactate conversion by lactate dehydrogenase (LDH). Glycolysis can also be upregulated in rapidly proliferating cells, such as cancer cells, even in the presence of oxygen, known as the Warburg effect.¹⁴ During polarization, M1 macrophages shift to glycolytic metabolism with TCA cycle disruptions, accumulating itaconate and succinate, while M2 macrophages maintain reliance on OXPHOS for energy.^15,16

Maintaining NAD⁺ levels is crucial for glycolysis, especially in the absence of OXPHOS, achieved primarily through pyruvate-to-lactate conversion by LDH. LDH is a tetramer composed of subunits encoded by the lactate dehydrogenase A (LDHA) and lactate dehydrogenase B (LDHB) genes.^17,18 The LDHA subunit drives the conversion of pyruvate to lactate and nicotinamide adenine dinucleotide to NAD⁺, whereas the LDHB subunit catalyzes the opposite reaction. The final composition of LDH will produce the most favorable reaction.^19–21 LDHA upregulation has been described in the aerobic glycolytic switch (Warburg effect), in which cells become dependent on glycolysis for their energy needs, even when oxygen is present, such as in the context of cancers or proinflammatory macrophages.¹⁴ In the emerging field of immunometabolism, the administration of 2-deoxyglucose blocks phagocytosis, secretion of proinflammatory cytokines, and formation of reactive oxygen species, which indicates that it is possible to change immune cell function by altering its metabolic state.²² Our previous study demonstrated that LDHA is predominantly expressed in the proximal tubules of healthy mouse kidneys. In addition, exposure of mouse and human renal proximal tubule cells to hypoxia in vitro increased LDHA expression with no change in LDHB levels.²³ In this study, we used metabolomic and transcriptomic approaches to investigate LDHA as a potential target to suppress the M1 macrophage proinflammatory profile in vitro.

Methods

Animals

Female LysM-Cre(−) LDHA fl/fl (wild-type [WT]) and LysM-Cre(+) LDHA fl/fl (knockout [KO]) mice (10 weeks old) were used. Previous studies from the same colony, sex matched for mycobacterium tuberculosis research, found no phenotypic differences in uninfected mice.²⁴ The mice were generated by cross-breeding LysM-Cre transgenic mice with LDHA fl/fl mice, provided by Dr. P. Seth, Beth Israel Deaconess Hospital, Boston, MA.²⁵ All mice were on a C57BL/6J background and lacked functional protein expression in macrophages, dendritic cells, and neutrophils. National Institutes of Health guidelines for animal care were followed, with approval from University of Alabama at Birmingham's Institutional Animal Care and Use Committee.

Isolation and Polarization of Bone Marrow–Derived Macrophages

Bone marrow–derived macrophages (BMDMs) were isolated and cultured according to previously described methods.^26,27 Mouse femurs were flushed and passed through a 40-µm strainer, and red blood cells were lyzed with ammonium-chloride-potassium lysing buffer. Cells were centrifuged, plated, and incubated for 7 days in DMEM with 15% FBS, antibiotics, nonessential amino acids, and 30 ng/ml macrophage colony-stimulating factor. Polarization was induced with IFN-γ for M1 and IL-4 for M2 states over 24 hours.

RNA Sequencing

RNA was extracted with Trizol and quality checked using the Eukaryote Total RNA Nano assay. Sequencing was performed on Illumina NextSeq500, and .fastq files were trimmed with FastQC. Alignment used the STAR pipeline with the GRCm38 mouse genome. Gene counts were obtained with HTseq, generating FPKM values. Differential expression was analyzed with DESeq2/edgeR. Protein ANalysis THrough Evolutionary Relationships pathway and gene ontology enrichment analyses were performed using KEGG and Gene Ontology. Each condition included cells from three animals.

LC-MS/MS Metabolomics

Extraction

BMDMs were cultivated identically to those used for RNA sequencing experiments as they were done in parallel and were isolated according to the Targeted Metabolomics and Proteomics Laboratory protocols.²⁸ Cells were washed with ice-cold PBS, covered with methanol, scraped, and placed on an orbital rotor for 30 minutes. The solution was diluted and centrifuged, and the supernatant was evaporated under N2. LC-MS/MS was performed by the Targeted Metabolomics and Proteomics Laboratory core at University of Alabama at Birmingham following standard protocols. Samples were redissolved in water, combined into a pooled sample, and injected onto a LunaOmega column. Metabolites were eluted with a gradient of acetonitrile and analyzed using a SCIEX TripleTOF 5600plus mass spectrometer. Full-scan spectra were collected every 250 ms, and MS/MS spectra were obtained in a 0.5-second duty cycle. Data were recorded as .wiff and .wiffscan files.

Metabolomic Data Analysis

LC-MS/MS spectral data (.wiff and .wiff.scan files) were analyzed using MS-DIAL version 4.9 with version 17 of the public MSMS database. Mass accuracies were set at 10 ppm for MS and 15 ppm for MSMS, with a retention time variance of 0.10 minutes. Data were normalized using the total ion current of annotated metabolites and then subjected to Pareto scaling, where each data point was mean centered and divided by the SEM to reduce data magnitude.

Statistically significant MS2-matched compounds (ANOVA <0.05) from MS-DIAL were uploaded to MetaboAnalyst 6.0 for impacted pathway analysis.²⁹ Gene with significantly altered expressions (P < 0.05) were analyzed using the g:GOSt Functional profiling software.³⁰ The pathway analysis module integrated results from the KEGG.

mRNA Quantification

Gene mRNA expression was confirmed using the same samples from the RNAseq experiments using RT-PCR with SYBR Green Master Mix in triplicate. Melting curves ensured a single PCR product, and the ΔΔCt method was used for analysis. The primers used are listed in Table 1.

Table 1.

Primers sequences used for RT-PCR

Primer	Direction	Sequence
Il1b	F	TGC​CAC​CTT​TTG​ACA​GTG​ATG
	R	TGA​TGT​GCT​GCT​GCG​AGA​TT
Il18	F	CCT​CTT​GGC​CCA​GGA​ACA​AT
	R	ACA​GTG​AAG​TCG​GCC​AAA​GT
Nlrp3	F	AGA​AGA​GAC​CAC​GGC​AGA​AG
	R	CCT​TGG​ACC​AGG​TTC​AGT​GT
Pkm	F	GCC​GCC​TGG​ACA​TTG​ACT​C
	R	CCA​TGA​GAG​AAA​TTC​AGC​CGA​G

Sequences of primers used in the RT-PCR validation experiments with both F and R directions. F, Forward; R, Reverse.

Western Blot

Cells were collected in radioimmunoprecipitation assay buffer with protease inhibitor, and protein concentration was determined using a the bicinchoninic acid assay. 15 µg of protein was resolved on a 12% SDS-PAGE gel, transferred to a polyvinylidene fluoride membrane, and blocked with 5% nonfat milk. Membranes were incubated overnight with specific antibodies, followed by horseradish peroxidase-linked secondary antibody for 1 hour. Horseradish peroxidase activity was detected by chemiluminescence using the KwikQuant system. Densitometry was performed with Image Studio Lite.

Statistical Analyses

Statistical analyses used ANOVA with a post hoc t test with P values <0.05 reported as statistically significant. n values are the number of animals per group.

Owing to word limitations in this article, full details and information are provided in the Supplemental Materials and Methods document.

Results

LDHA Deletion Induces Transcriptional Changes in Macrophages

To investigate the role of LDHA in macrophage function and polarization, bone marrow was isolated from 10-week-old mice that either expressed (WT) or lacked (KO) LDHA in the myeloid compartment using the LysM-Cre system (Figure 1A). BMDMs were grown in the presence of macrophage colony-stimulating factor for 7 days to ensure macrophage maturity (Figure 1B). Upon maturation, BMDMs were polarized with IFN-γ (100 µg/ml) for 24 hours and subsequently probed by immunoblotting for LDHA and LDHB protein expression. BMDMs that were stimulated with either IFN-γ or IL-4 to polarize them toward M1 or M2, respectively, were probed for the expression of canonical M1 marker inducible nitric oxide synthase and M2 Arginase 1 showing the desired upregulation profiles (Figure 1, C and D, and Supplemental Figure 3A). RNA sequencing of BMDMs displayed a clear transcriptional separation between untreated (Mo), IFN-γ–treated, and IL-4–treated groups (Figure 1E). LDHA deletion remained unchanged after IFN-γ stimulation, while no significant change in LDHB expression was observed across all conditions (Supplemental Figure 1, A and B).

Figure 1.

LDHA deletion produces a distinct transcriptomic profile in BMDMs. (A) Schematic of the Cre-lox system used to produce myeloid-specific LDHA deletion. Cre recombinase expression relies on myeloid-specific Lyz2 (LysM) promoter, which then drives the deletion of exon 2 of LDHA, resulting in a nonfunctioning protein. (B) Schematic of the in vitro experimental design, in which we isolated bone marrow from LDHA KO and WT mice, matured cells in the presence of MCSF for 7 days, stimulated mature BMDMs with IFN-γ for 24 hours, and collected them for RNA-sequencing and LC-MS/MS metabolomics. (C) Representative western blots and (D) quantification data of BMDMs stimulated with either IL-4 (20 μg/ml) or IFN-γ (100 μg/ml) for 24 hours. iNOS was upregulated by IFN-γ stimulation, whereas Arg1 expression was upregulated by the IL-4. (E) PCA2 plot of transcriptomic profiles from the RNA sequencing, highlighting significant separation between the groups. Arg1, Arginase 1; BMDM, bone marrow–derived macrophage; iNOS, inducible nitric oxide synthase; KO, knockout; LDHA, lactate dehydrogenase A; MCSF, macrophage colony-stimulating factor; WT, wild-type.

Protein ANalysis THrough Evolutionary Relationships gene ontology pathway analysis revealed variation in multiple pathways in IFN-γ–stimulated BMDMs lacking LDHA. The leukocyte activation pathway exhibited a significant downregulation in Il1b and Il18 genes transcripts. Adaptive immune response pathway showed upregulation of Il12b, myeloid cell differentiation, and affected cytokine–cytokine receptor interaction pathways (Table 2).

Table 2.

Top four significantly affected Protein ANalysis THrough Evolutionary Relationships pathways by LDHA deletion in IFN-γ macrophages

Pathways Affected by LDHA Deletion	Upregulated Genes		Downregulated Genes
Regulation of leukocyte activation	Il7r	Tnfrsf1b	Il1b	Il18
Adaptive immune response	Il12b	Tnfsf4	Il27	Tnfsf18
Myeloid cell differentiation	Itgb3	Cdkn1c	Hif1a	Irf8
Cytokine–cytokine receptor interaction	Cxcl3	Ccl22	Ccl5	Ccl6

The top four most significant pathways from Protein ANalysis THrough Evolutionary Relationships analysis in IFN-γ–stimulated macrophages between lactate dehydrogenase A wild-type and knockout with top two upregulated and downregulated genes per pathway. LDHA, lactate dehydrogenase A.

Using pathway analysis guidance, multiple genes were identified whose expression was significantly altered by the deletion of LDHA in IFN-γ–stimulated BMDMs (Figure 2A). Glut1, Hk1, Pfkfb3, Gapdh, and Pkm gene transcript numbers were upregulated in LDHA KO BMDMs compared with the WT counterparts. Cytokine Il1b, Il18, Il27, and chemokine Ccl (2–6; 8–9) gene transcript reads were in lower abundance in LDHA KO BMDMs. Itgb3, and Il12b transcripts were present in higher abundance. Ass1, Nox1, and Nos2 were significantly upregulated in LDHA KO. Nfkb and Nlrp3 were downregulated in LDHA KO BMDMs. A more comprehensive list of significantly differentially expressed genes, including the top 50 changes between LDHA KO and WT BMDMs after IFN-γ stimulation, is also provided (Supplemental Figure 1E).

Figure 2.

LDHA-deficient IFN-stimulated BMDMs display downregulated proinflammatory gene expression. (A) Heatmap of significantly altered (P < 0.01) genes between IFN-γ–stimulated LDHA WT and KO BMDMs that are involved in glycolysis, proinflammatory response, cytokine expression, chemokine expression, and NF-ƙB (n=3 animals). (B–E) qRT-PCR gene expression of Il1b, Il18 Nlrp3, and Pkm (n=3 animals). (F) Representative western blots and (G) quantification data of BMDMs from either LDHA WT or LDHA KO mice showed no difference in the p-p65/p65 ratio between the genotypes. IL-1β protein expression is significantly downregulated in LDHA KO macrophages (n=3 animals). Ponceaus S stain was used to confirm equal protein load. LDHB, lactate dehydrogenase B.

Next, Il1b, Il18, and Nlrp3 gene expressions were directly confirmed by PCR to be significantly downregulated in the IFN-γ–stimulated BMDMs lacking LDHA, whereas the glycolytic enzyme Pkm gene expression is upregulated, validating the results from the RNA sequencing analysis (Figure 2, B–E). IL-1ß protein expression was significantly decreased in the lysates of LDHA KO BMDMs at the protein level, whereas phosphorylated and total protein expression of RelA, gene product p65 was not significantly different between the genotypes (Figure 2, F and G, and Supplemental Figure 3B).

LDHA Deletion Induces Metabolic Changes in Macrophages

Untargeted metabolomics using LC-MS/MS was conducted in parallel with the RNA sequencing experiment on the same BMDMs (Figure 3A). Metabolites were separated into positively and negatively charged ions, and the resulting spectra were identified, aligned, and matched with the corresponding MS2 spectra allowing quantification. Top matches with P values < 0.05 showed a significant metabolic alteration in IFN-γ–stimulated macrophages lacking LDHA (Figure 3B).

Figure 3.

LDHA deletion in IFN-stimulated BMDMs results in a less proinflammatory metabolome. (A) Schematic of the in vitro metabolomic experimental design in which matured BMDMs were activated with IFN-γ (100 μg/ml) for 24 hours, resulting in M1 phenotype. Cell lysates were collected in ice-cold methanol and processed through liquid chromatography to separate positive and negative ions to be further processed by the mass spectrometer into MS/MS spectra. (B) Heatmap of identified and significantly altered (ANOVA; P < 0.05) metabolites with the matched MS2 spectra (n=3 animals). (C–I) Individual plots of metabolites: acetylcarnitine, pyridoxal 5-phosphate (vitamin B6), theophylline, l-carnitine, NAD, nicotinamide, creatinine. NAD, nicotinamide adenine nucleotide.

Metabolites acetylcarnitine, pyridoxal 5-phosphate (vitamin B6), l-carnitine, and choline, creatinine were all presented in significantly lower quantities in LDHA KO BMDMs when compared with LDHA WT (Figure 3, C, D, F, and I). By contrast, theophylline levels were higher in LDHA KO (Figure 3E). Nicotinamide adenine nucleotide (NAD) and nicotinamide levels were diminished by the LDHA deletion (Figure 3, G and H). LDHA KO BMDMs exhibited reduced lactate production compared with LDHA WT BMDMs after IFN-γ stimulation, whereas no difference was observed under baseline conditions without stimulation (Supplemental Figure 1C).

Quantitative enrichment analysis of significantly altered MS2-matched metabolites was conducted using MetaboAnalyst software and the top ten graphed pathways (Figure 4A). Glutamate, amino sugars, and pyruvate metabolism were most significantly altered by the LDHA deletion in BMDMs. MS2-matched metabolites from MS-DIAL software, along with significantly changed protein-coding genes (at least two-fold change in gene expression with q<0.05), were used in the MetaboAnalyst software Joint-Pathway Analysis module. The most significantly impacted pathways included the Rap1 signaling pathway, cytokine-cytokine receptor interactions, focal adhesion, and the mitogen-activated protein kinase signaling pathway (Figure 4B). Gene ontology pathway analysis of RNA sequencing was performed using the KEGG and Reactome databases, with results presented in Supplemental Figure 2.

Figure 4.

Joint-Pathway analysis of IFN-stimulated BMDMs lacking LDHA. (A) Quantitative enrichment analysis of metabolites performed in MetaboAnalyst with top ten affected metabolic pathways using the MS-DIAL–generated list of metabolites. (B) MetaboAnalyst software Joint-Pathway Analysis module output of integrated data of identified MS-DIAL MS2-matched metabolites with significant concentration change (ANOVA; P < 0.05) and significantly altered (P < 0.05 with at least two-fold change) transcript counts between IFN-stimulated WT and KO BMDMs. MAPK, mitogen-activated protein kinase.

Discussion

Macrophage functions are closely linked to their metabolic capabilities. Metabolic reprogramming allows macrophages to differentiate and adapt roles essential for an effective innate immune response in disease conditions.^5,24 Thus, macrophages, which are essential for innate immunity, hold significant potential for immunometabolic manipulation to regulate proinflammatory responses. Glycolysis is upregulated in macrophages that undergo proinflammatory activation and is critical for the macrophage response. In this study, we investigated the role of LDHA in the proinflammatory response of IFN-γ–stimulated BMDMs, providing important mechanistic insights into macrophage metabolic reprogramming during inflammation.

Our RNA sequencing data revealed significant metabolic and immune changes in LDHA KO BMDMs. Despite the loss of LDHA, we observed upregulation of key glycolytic genes, such as Glut1, Hk1, Pfkfb3, Gapdh, and Pkm, indicating a compensatory increase in glycolytic flux. This suggests that LDHA KO BMDMs are attempting to sustain energy production by enhancing glucose uptake and pushing intermediates through the glycolytic pathway. In addition to these metabolism-related gene changes, we found that Itgb3 and Il12b were more highly expressed in LDHA KO BMDMs. Itgb3 plays a role in cell adhesion and signaling, potentially influencing macrophage interactions with the extracellular matrix and other immune cells. Meanwhile, Il12b, which encodes part of the IL-12 cytokine, is involved in promoting T-cell responses, hinting at a shift toward adaptive immune signaling rather than a typical innate inflammatory response. Interestingly, the oxidative stress–related genes Ass1, Nox1, and Nos2 were also significantly upregulated in LDHA KO BMDMs, whereas the inflammatory mediators Nfkb and Nlrp3 were downregulated. This pattern suggests that, while LDHA KO induces increased oxidative stress, it simultaneously dampens the broader inflammatory response. The upregulation of oxidative stress pathways, alongside the attenuation of key inflammatory regulators, likely reflects a complex feedback mechanism linking cellular metabolism to immune regulation in the absence of LDHA, balancing metabolic stress with a restrained inflammatory profile.

IL-1β and IL-18 are crucial cytokines for inducing inflammatory responses, playing key roles in infection and cancer³¹; however, their excessive presence has been implicated in multiple pathologies.³² IL-1β and IL-18 are synthesized as precursors and require Nlrp3 inflammasome-mediated cleavage to become active.³³ Similar to the Nlrp3 inflammasome, we observed significant downregulation of Il1b and Il18 mRNA in BMDMs lacking LDHA, indicating a reduced proinflammatory profile. In addition, we confirmed a reduced protein level of IL-1β; however, no significant differences were observed in the protein expression levels of p-p65 and p65. This suggests that LDHA may play a critical role in supporting cytokine production through metabolic means, independent of major changes in NF-κB activation. LDHA-driven glycolysis could be essential for fueling the processes necessary for IL-1β synthesis and secretion. Furthermore, LDHA deletion may influence cellular redox states or other metabolic intermediates that directly or indirectly regulate IL-1β production. This decoupling of NF-κB activation from IL-1β production in LDHA KO macrophages underscores a metabolic layer of control over inflammation, indicating that metabolic reprogramming can selectively affect cytokine responses even when core inflammatory pathways such as NF-κB remain relatively intact. Investigating this crossregulation in future studies could offer valuable insights into how metabolism regulates immune responses, especially in macrophage-driven diseases, such as AKI.

Using untargeted metabolomics allowed us to observe the whole metabolome in contrast to concentrating on a specific pathway, which can introduce bias. We identified multiple metabolites that have been linked to the macrophage inflammatory response. Notably, acetylcarnitine, implicated in the activation of the inflammatory response,³⁴ was significantly downregulated in IFN-γ–activated LDHA KO macrophages. Similarly, pyridoxal 5-phosphate, linked to the suppression of IL-1β production through inhibition of NLRP3 inflammasome activation,^34,35 was also significantly downregulated in LDHA KO BMDMs. Furthermore, both theophylline and uridine 5-diphosphate N-acetylglucosamine (NAG) (UDP-GlcNAc), which are implicated in the anti-inflammatory response in macrophages,³⁶ were upregulated in LDHA KO BMDMs. Theophylline decreases the release of inflammatory mediators, such as TNF-α and interleukins, contributing to its effectiveness in reducing inflammation.³⁷ UDP-GlcNAc plays a key role in the synthesis of glycosaminoglycans and proteoglycans, essential components of the extracellular matrix involved in tissue repair and inflammation regulation.³⁸ In addition, NAG, a derivative of UDP-GlcNAc, has demonstrated anti-inflammatory effects by inhibiting the expression of inflammatory cytokines and reducing the activation of inflammatory pathways.³⁶ This upregulation suggests a lack of LDHA in macrophage suppression of the inflammatory response.

Creatine, an energy reserve molecule, promotes macrophage reprogramming to the M2 phenotype while inhibiting M1 functionality. Our RNA-seq data showed significant upregulation of the creatine transporter (Slc6a8) in LDHA-deficient macrophages, indicating increased creatine demand. Conversely, decreased creatinine levels from metabolomics study suggest these macrophages struggle to polarize toward the M1 phenotype because of the lack of LDHA.

Reduced NAD concentrations, NAD recycling, and the suppression of de novo NAD synthesis have been linked to decreased immune response and macrophage function.³⁹ Our multiomics analysis of LDHA KO IFN-γ–stimulated BMDMs identified nicotinate and nicotinamide metabolism as significantly affected, with notably lower concentrations of NAD and its precursor, nicotinamide. A recent study from our collaborators also demonstrated that LDHA KO BMDMs exhibit a reduced NAD⁺/NADH ratio, both with and without IFN-γ stimulation.²⁴ These data further reinforce that LDHA deletion results in a decreased ability of macrophages to express a proinflammatory profile.

Combining transcriptomic and metabolomics data reveals significant changes in metabolic pathways affecting macrophage function, polarization, and inflammation in LDHA-deficient BMDMs. LDHA absence impairs NLRP3 inflammasome activity and downstream cytokines (IL-1β and IL-18), disrupting metabolic programming and affecting the M1 phenotype. Our study shows that the top three metabolic changes in macrophages after LDHA deletion involve glutamate, amino sugars, and pyruvate metabolism.

Glutamate, an essential neurotransmitter, significantly influences inflammation and kidney function. In proinflammatory (M1) macrophages, glutamate supports the production of reactive oxygen species and nitric oxide, which are essential for killing pathogens, but also contribute to tissue damage and inflammation.⁴⁰ During AKI, dysregulated glutamate levels can lead to harmful effects. Modulation of renal receptors for glutamate, including metabotropic glutamate receptor 1 and AMPA receptors, contributes to the inflammatory response and exacerbates kidney damage.⁴¹

Amino sugars, such as glucosamine, are essential for cellular signaling and structural integrity. NAG, a component of hyaluronic acid, plays a role in inflammation and tissue repair.⁴² Although the role of amino sugars in macrophages in proinflammation is less studied, our research indicates that modulating macrophages' amino sugar metabolism through LDHA expression could pave the way for new investigations into the proinflammatory effects.

As we expected, we also observed that pyruvate metabolism is changed after LDHA deletion, as we know that in proinflammatory macrophages, glycolysis is upregulated and pyruvate is preferentially converted to lactate rather than being used in the citric acid cycle. This reprogramming supports rapid ATP production to assist the inflammatory response. This metabolic shift supports the biosynthetic and energetic demands of activated immune cells, contributing to inflammation.⁴³ Moreover, we confirmed that the deletion of LDHA significantly reduced lactate production in macrophages after IFN-γ stimulation. These results strongly advocate that LDHA may be important in promoting proinflammation response by driving glycolysis.

Interestingly, despite these significant shifts in macrophage polarization, we did not observe any changes in phagocytic activity (Supplemental Figure 1D). The dissociation between inflammatory signaling and phagocytic function points to a specific role of LDHA in modulating the inflammatory response, independent of the basic immunoregulatory function of macrophages. This implies that LDHA may be involved in the metabolic reprogramming of macrophages that drives the proinflammatory state, but is not necessary for phagocytosis. Future studies could focus on elucidating the exact metabolic pathways involved in LDHA-mediated inflammation and how these pathways influence macrophage behavior beyond cytokine production.

In AKI, the balance between IFN-γ–induced M1 macrophages and IL-4–induced M2 macrophages is crucial. An excessive M1 response can worsen tissue injury, whereas a shift toward M2 activation may aid in tissue recovery. Although previous studies have suggested that LDH may serve as a serum biomarker of injury in sepsis-induced AKI,^44,45 there is a lack of research on the modulation of LDHA and inflammation in AKI. The functional significance of these results relates to the potential to manipulate LDHA expression in macrophages to develop therapeutic approaches in the setting of AKI, as has been done in other nonrenal settings. Seth and colleagues have demonstrated the antitumor immune effects of myeloid-specific LDHA expression in an animal model of lung cancer.²⁵ Previous studies have targeted LDHA using stiripentol, an LDH inhibitor, which has shown promising antiepileptic effects and is now US Food and Drug Administration approved and used for pediatric epilepsy.⁴⁶ Our study focusing on modulating LDHA expression in macrophages to potentially reprogram inflammation by reducing proinflammatory profiles and enhancing anti-inflammatory responses is novel. Rather than using LDH as a cell injury marker, we aim to gain a deeper understanding of the LDHA-mediated metabolomics processes in inflammation. Given the significant role of inflammation in various types of AKIs, our work has the potential to lead to new insights into targeting LDHA in macrophages, leading to the development of cellular therapies for AKI.

A limitation of these studies pertains to the largely in vitro nature of this work. Future studies using in vivo models of AKI using myeloid-deficient LDHA mice will be important. In the follow-up study, we will use macrophage LDHA KO mice to investigate the role of macrophage LDHA in AKI progression and recovery, highlighting its significance in conditions with excessive inflammation, such as AKI and sepsis.

Supplementary Material

kidney360-6-197-s001.pdf ^{(1.5MB, pdf)}

kidney360-6-197-s002.pdf ^{(425KB, pdf)}

Disclosures

Disclosure forms, as provided by each author, are available with the online version of the article at http://links.lww.com/KN9/A776.

Funding

G. Osis: American Heart Association (20PRE35200054). A. Agarwal: National Institute of Diabetes and Digestive and Kidney Diseases (DK59600 and DK118932), National Institute of Environmental Health Sciences (U54 ES030246), and University of Alabama at Birmingham-University of California-San Diego O’Brien Center for Acute Kidney Injury (U54 DK137307). Y. Lu: University of Alabama at Birmingham (Frommeyer Fellowship Award). S. Barnes: National Institutes of Health (S10 RR027822).

Author Contributions

Data curation: Yan Lu, Gunars Osis.

Formal analysis: Stephen Barnes, Yan Lu, Gunars Osis, Landon Wilson.

Investigation: Yan Lu, Gunars Osis, Saakshi Thukral, Landon Wilson, Anna A. Zmijewska.

Methodology: Amie Traylor.

Project administration: Anna A. Zmijewska.

Software: Stephen Barnes, Gunars Osis, Landon Wilson.

Supervision: Anupam Agarwal, Stephen Barnes, James F. George.

Validation: Stephen Barnes, Anna A. Zmijewska.

Writing – original draft: Yan Lu, Gunars Osis.

Writing – review & editing: Anupam Agarwal, Stephen Barnes, James F. George, Yan Lu, Gunars Osis, Amie Traylor, Anna A. Zmijewska.

Data Sharing Statement

Data related to transcriptomic, proteomic, or metabolomic data. Original data created for the study are or will be available in a persistent repository upon publication. All data are included in the manuscript and/or supporting information. RNASeq data - Gene Expression Omnibus: GSE277304. Metabolomics data - National Metabolomics Data Repository (NMDR). Experimental Data. Figshare: Lippincott Data Repository; Gene Expression Omnibus. 10.6084/m9.figshare.27018541 https://figshare.com/s/f81efb256ca7d5acf4c1. https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?&acc=GSE277304 Access token: ktonusgafxifbwh. Experimental Data; Research Protocols; Published Material; Generated Data. Figshare: Lippincott Data Repository; Gene Expression Omnibus. 10.6084/m9.figshare.27018541 https://figshare.com/s/f81efb256ca7d5acf4c1 https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?&acc=GSE277304 Access token: ktonusgafxifbwh.

Supplemental Material

This article contains the following supplemental material online at http://links.lww.com/KN9/A775.

Supplemental Figure 1. (A) Western blot analysis of BMDMs from LDHA WT and KO mice, demonstrating significant LDHA deletion without compensatory upregulation of LDHB. No changes in LDHA or LDHB expression were observed after stimulation with IL-4 (20 μg/ml) or IFN-γ (100 μg/ml). Densitometric quantification is shown in (B). (C) Lactate production in LDHA KO BMDMs was reduced compared with WT after IFN-γ stimulation (100 μg/ml), whereas no significant difference was observed between the groups without stimulation. (D) Phagocytic activity was comparable between LDHA KO and WT BMDMs, both with and without IFN-γ stimulation (100 μg/ml). (E) Significantly differentially expressed genes (protein-coding genes only, P < 0.01, average expression of at least 50 reads) of IFN-γ–stimulated BMDMs because of the deletion of the LDHA. Genes upregulated in the LDHA KO BMDMs compared with the LDHA WT are shown on the left subpanel. The reverse is shown on the right panel.

Supplemental Figure 2. Gene ontology pathway analysis of the RNA seq. Genes with significantly altered expressions (P < 0.05) were analyzed using the g:GOSt Functional profiling software, as a part of the g:Profiler. (A) Dot plot representation of the pathway analysis hits using the Gene ontology biological processes database, with the y axis showing a negative log of the P value. (B) Highlight of the most statistically significant pathways. (C and D) In an identical format, the analysis was performed on the KEGG and REACTOME databases.

Supplemental Figure 3. IFN-γ–stimulated BMDMs deficient in LDHA exhibit reduced expression of proinflammatory genes. (A) Representative complete western blots of BMDMs stimulated with either IL-4 (20 μg/ml) or IFN-γ (100 μg/ml) for 24 hours. (B) Representative complete western blots of BMDMs from LDHA WT or LDHA KO mice treated with IFN-γ (100 μg/ml) for 24 hours.

Supplemental Method and Material document