Overexpression of mitofusin 2 ameliorates inflammation and oxidative stress in lipopolysaccharide‐induced mastitis model by regulating phosphofurin acidic cluster sorting protein 2

Abstract

Background

Mastitis seriously affects the mammary health of humans and animals. Studies have found that inflammation and oxidative stress play key roles in the occurrence and development of mastitis. Therefore, in‐depth research on related molecular mechanisms is of great significance.

Methods

Postpartum mice were anesthetized with pentobarbital and administered lipopolysaccharide to develop the mouse mastitis model. Proteomic analysis was performed to compare protein expression in mitochondria‐associated endoplasmic reticulum membranes (MAM) from two mouse mammary gland groups. Western blot was used to detect the expression of MAM‐related proteins in mitochondria. AlphaFold3 was used to predict the molecular structures of phosphofurin acidic cluster sorting protein 2 (PACS2) and mitofusin 2 (MFN2) and their interaction levels. The MFN2–PACS2 interaction was investigated using co‐immunoprecipitation and small interfering RNA.

Results

The results showed that the inflammation level in the mammary gland tissue of mice with mastitis significantly increased, the total antioxidant capacity decreased, and the expression of MAM‐related proteins MFN2 and PACS2 was significantly downregulated. In cell experiments, overexpression of MFN2 can inhibit inflammation and oxidative stress responses, and promote the interaction between MFN2 and PACS2 to affect the formation of MAMs.

Conclusion

In summary, this study suggests that mastitis can alter the expression of MAM–related proteins in mouse breast tissue. The interaction between MFN2 and PACS2 regulates the formation of MAMs. Overexpression of MFN2 can promote the formation of MAMs and inhibit inflammation and oxidative stress response in mammary epithelial cells. Our results provided a new theoretical basis and potential therapeutic targets for the prevention and treatment of mastitis.

The interaction between mitofusin 2 (MFN2) and phosphofurin acidic cluster sorting protein 2 regulates the formation of mitochondria‐associated endoplasmic reticulum membranes (MAM). Overexpression of MFN2 can promote the formation of MAMs and inhibit inflammation and oxidative stress response in mammary epithelial cells.

Mechanism of mitofusin 2/phosphofurin acidic cluster sorting protein 2 regulating mitochondria‐associated endoplasmic reticulum membranes to relieve mammary oxidative stress.

1. INTRODUCTION

Mastitis, a common breast disease during lactation, causes serious problems to the health of lactating women and newborns globally. ¹ , ² Based on the different causes, it can be divided into acute suppurative mastitis, areola fistula, and plasma cell mastitis. Acute suppurative mastitis often occurs in the lactation period. ³ The incidence rate of acute mastitis in primipara is as high as 2%–4%, which is twice as much as that in multipara mastitis. Milk stasis is often accompanied by bacterial infection, which leads to inflammatory responses and oxidative stress. ⁴ Numerous research studies have reported that inflammatory responses and oxidative stress play an important role in the occurrence and development of mastitis. ⁵ , ⁶ , ⁷ An in‐depth understanding of the pathogenesis of mastitis has always been a key research area in the prevention and treatment of mastitis. Screening for new and efficient molecular targets in host cells is expected to provide new strategies for the prevention and treatment of mastitis.

Mastitis may lead to serious complications such as breast abscesses or even septicemia. And inflammation and oxidative stress can be regarded as a “double‐edged sword” in mastitis. They are essential for eliminating pathogens. If they are dysregulated, they will also cause collateral damage to the host tissue. ⁸ During mastitis, the body exhibits a strong inflammatory response, manifested as an increase in the levels of pro‐inflammatory cytokines and the infiltration of immune cells. ⁹ , ¹⁰ , ¹¹ This inflammatory response helps to fight against infections, but excessive inflammation will damage breast epithelial cells and disrupt the blood–milk barrier, exacerbating tissue damage and impeding lactation function. ¹² Meanwhile, immune cells such as neutrophils will produce reactive oxygen species (ROS) during the defense process. ¹³ , ¹⁴ Mastitis is usually accompanied by an elevated level of oxidative stress. ¹⁵ Although ROS helps to kill pathogens, excessive ROS will damage cellular components and further damage breast tissue. If the oxidative stress exceeds the body's antioxidant defense capacity (e.g., Nrf2 [nuclear factor erythroid 2‐related factor 2] pathway), it will amplify the inflammatory response and tissue damage through a vicious cycle. ¹⁶ To alleviate such damage, the body will mobilize regulatory mechanisms such as antioxidant Nrf2 pathways and autophagy to neutralize ROS and clear damaged cell debris. However, in cases of severe mastitis, these protective mechanisms may still be insufficient to avoid further damage. Therefore, an in‐depth understanding of the regulatory pathways of inflammation and oxidative stress in mastitis is essential for developing intervention measures to protect breast tissue.

The inflammatory and oxidative stress microenvironment developed by these excessive pro‐inflammatory mediators and ROS in the mammary gland will further promote the occurrence and development of mastitis. ¹⁷ This is mainly because inflammation and oxidative stress can severely damage the functions of the mitochondria and endoplasmic reticulum (ER) in mammary epithelial cells, leading to the destruction of the structure of mitochondria‐associated ER membranes (MAM), thus exacerbating the occurrence and development of mastitis. ¹⁸ MAMs play a crucial role in life activities. MAMs are special connection regions between the ER and mitochondria. ¹⁹ They not only play a key role in processes such as the regulation of intracellular calcium homeostasis, lipid synthesis, and transport but also participate in important physiological and pathological processes such as apoptosis and autophagy. The ER and mitochondria mainly rely on MAMs to achieve efficient material exchange and signal transmission to maintain the normal metabolism and function of cells. It has been reported that MAMs are closely related to the occurrence and development of inflammatory response and oxidative stress. ²⁰ When cells are stimulated by external factors to trigger inflammation and oxidative stress, the structure and function of MAMs will change. ²¹

In addition, the expression of some protein molecules on MAMs, such as mitofusin 2 (MFN2), will change significantly under inflammatory and oxidative stress conditions, thereby participating in the regulation of mitochondrial function in cells. ²² MFN2, as a key protein on the outer mitochondrial membrane, not only mediates mitochondrial fusion but also plays an important role in the formation and function of MAMs and participate in the regulation of ER–mitochondria communication. ²³ Abnormal expression of MFN2 will directly affect the integrity and function of MAMs. ²⁴

It is unclear how MFN2 regulates the function of MAMs, thereby affecting the occurrence and development of inflammatory responses and oxidative stress. As an important subcellular structure, MAMs are closely related to inflammation and oxidative stress. Therefore, in‐depth research on the molecular mechanisms of MAMs in the occurrence and development of mastitis is expected to provide theoretical basis for the prevention and treatment of mastitis, and provide new targets for screening potential small‐molecule compounds for prevention and treatment of mastitis.

2. MATERIALS AND METHODS

2.1. Animals

C57BL/6 mice (8‐week‐old males, n = 10; females, n = 20) were obtained from Liaoning Changsheng Biotechnology Co., Ltd. (China) and acclimatized for 1 week in a specific pathogen‐free facility prior to experiments. All animal experiments were conducted in accordance with guidelines approved by the Animal Ethics and Welfare Committee of Jilin University (approval no. SY202505020).

2.2. Cell culture

Mouse mammary epithelial cells (mMEC, EpH4‐Ev, ATCC CRL‐3063) were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS, Shanghai Bairui Biotechnology, China) at 37℃ under 5% CO₂. Cell passaging was performed using 0.25% trypsin (Merck Life Science, Germany), and a standard freezing medium containing FBS, DMEM, and dimethyl sulfoxide (DMSO, Merck Life Science) in the ratio 5:4:1 was used for cryopreservation.

2.3. Mouse mastitis model

Postpartum mice (days 1, 3, 5, and 7 after delivery) were anesthetized with 0.3% pentobarbital and administered 50 μL of lipopolysaccharide (LPS, 0.2 mg/mL, Merck Life Science) via intraductal injection into the fourth mammary gland pair. Surgical procedures were performed under aseptic conditions.

2.4. Hydrogen peroxide content

Mammary tissues were homogenized in hydrogen peroxide (H₂O₂) lysis buffer (1:10 [w/v], Beyotime Biotechnology, China). After centrifugation (12 000 g, 4℃, 5 min), supernatants were mixed with H₂O₂ detection reagent (Beyotime Biotechnology) and incubated for 30 min. Absorbance at 450 nm was measured using a microplate reader.

2.5. Total antioxidant capacity (FRAP assay)

Tissue homogenates were prepared in ice‐cold phosphate‐buffered saline (1:5, w/v). ‌Ferric ion reducing antioxidant power‌ (FRAP) working solution containing TPTZ 2,4,6‐Tri(2‐pyridyl)‐s‐triazine (Beyotime Biotechnology) was incubated with samples at 37℃ for 5 min. Absorbance at 593 nm was quantified.

2.6. Malondialdehyde measurement

Samples were homogenized in malondialdehyde (MDA) lysis buffer (Beyotime Biotechnology) and reacted with thiobarbituric acid (0.37%, Beyotime Biotechnology) at 100℃ for 15 min, and absorbance at 532 nm was recorded.

2.7. Catalase activity assay

Tissue lysates were reacted with 250 mmol/L H₂O₂ for 5 min. Residual H₂O₂ was quantified colorimetrically at 520 nm after reaction termination.

2.8. Structure prediction

The tertiary structures of MFN2 and phosphofurin acidic cluster sorting protein 2 (PACS2) were predicted using AlphaFold3 (DeepMind, UK). The highest confidence models were visualized and analyzed using ChimeraX, version 1.8 (UCSF, USA).

2.9. Co‐immunoprecipitation

Cell lysates containing 500 μg of total protein (quantified by bicinchoninic acid assay [BCA], Beyotime Biotechnology) were incubated with 2 μg of affinity‐purified primary antibodies at 4℃ overnight with rotation. Pierce Protein A/G Magnetic Beads (25 μL, 0.25 mg, Thermo Fisher Scientific, USA) were washed thrice with Tris‐buffered saline with Tween‐20 (TBST, 20 mmol/L Tris–HCl, 150 mmol/L NaCl, 0.1% Tween‐20) and incubated with antigen–antibody complexes at room temperature for 1 h. After three washes with TBST and one wash with distilled water, bound proteins were eluted using 2× SDS (sodium dodecyl sulfate) loading buffer (100℃, 10 min) or mild elution buffer (room temperature, 10 min) for subsequent Western blot analysis.

2.10. Western blotting

Proteins were extracted using radioimmunoprecipitation assay buffer (Beyotime Biotechnology), quantified via BCA assay (Beyotime Biotechnology), and separated on 12% SDS‐PAGE (polyacrylamide gel electrophoresis) gels. After the sample was transferred to polyvinylidene fluoride membranes (Millipore, USA), blots were probed with primary antibodies overnight at 4℃ and then with horseradish peroxidase–conjugated secondary antibodies (1:5000, Abcam, UK). Details of primary antibodies are presented in Table 1. Signals were detected using Enhanced Chemiluminescence (ECL) reagent (New Cell and Molecular Biotech, China).

TABLE 1.

Details of primary antibodies.

Antibody name	Catalog number	Dilution factor	Company
Cyclooxygenase‐2	27308‐1‐AP	1:1500	Proteintech
Inducible nitric oxide synthase	22226‐1‐AP	1:1000	Proteintech
Mitochondrial fusion protein 2	12186‐1‐AP	1:2500	Proteintech
Mitochondrial fusion protein 1	13798‐1‐AP	1:2500	Proteintech
Mitochondrial fission protein 1	10956‐1‐AP	1:4000	Proteintech
Dynamin‐related protein 1	8570	1:2000	Cell Signaling Technology
Optic atrophy 1	80471	1:2000	Cell Signaling Technology
Eukaryotic translation initiation factor 2 alpha (eIF‐2α)	9722	1:1000	Cell Signaling Technology
Phosphorylated‐eIF‐2α (P‐eIF‐2α)	3398	1:1000	Cell Signaling Technology
Phosphacinosin‐membrane‐associated protein 2	30172‐1‐AP	1:1000	Proteintech
β‐Actin	66009‐1‐Ig	1:5000	Proteintech

2.11. MAM isolation

Mammary tissues were homogenized in ice‐cold IB‐1 buffer (containing 0.5% bovine serum albumin, 0.5 mmol/L ethylene glycol‐bis(β‐aminoethyl ether)‐N,N,N′,N′‐tetraacetic acid, Merck Life Science). Crude mitochondria were isolated using differential centrifugation (740–10 000 g) and purified using Percoll density gradient centrifugation (95 000 g, 30 min). MAM fractions were collected and resuspended in Membrane Binding Buffer (MRB) buffer. The detailed steps are as follows.

An amount of 4 mL of Percoll culture medium was added to a 7‐mL glass centrifuge tube paired with a high‐speed centrifuge. The crude mitochondrial suspension collected was added to the 4‐mL Percoll medium; then 1.75 mL of MRB solution was gently added to the top layer, to avoid mixing the liquid in the centrifuge tube. The mixture was centrifuged at 0.4℃ and 95 000 g for 30 min. After centrifugation, two dense zones of liquid distribution were observed inside the centrifuge tube, with the upper layer being the MAM layer and the lower layer being the mitochondrial layer. MAM components were extracted and diluted 10‐fold with MRB. The mixture was centrifuged at 4℃ and 6300 g for 10 min. The supernatant was collected in a 7‐mL glass centrifuge tube equipped with a high‐speed centrifuge. The supernatant was centrifuged at 4℃ and 100 000 g for 1 h. The supernatant was discarded, and MAM precipitate was resuspended with 100–200 μL of MRB.

2.12. Proteomic analysis

MAM proteins were digested with trypsin and analyzed using liquid chromatography–tandem mass spectrometry in data‐independent acquisition mode. Raw data were processed using Spectronaut X (Biognosys AG, Switzerland). Proteins with fold change >1.2 or <0.83 (p < 0.05) were considered differentially expressed.

2.13. Quantitative reverse transcription polymerase chain reaction

Total RNA was extracted using TRIzol reagent (Thermo Fisher Scientific). Complementary DNA synthesis employed the ABScript III RT Kit (ABclonal, China). Quantitative polymerase chain reaction (qPCR) was performed using SYBR Green Master Mix (Sangon Biotech, China) using primer sequences presented in Table 2. Relative expression was calculated using the 2^−ΔΔCt method.

TABLE 2.

Primer sequences.

Primer	Forward (5′‐3′)	Reverse (5′‐3′)
NM_012589.2 interleukin‐6	GACAACTTTGGCATTGTGG	ATGCAGGGATGATGTTCTG
NM_008361.4 interleukin‐1β	TGGCATTGATCTGGTTCATC	TGGTGAGGTTGGAAGGTGAT
NM_001278601.1 tumor necrosis factor‐alpha	CAGGCTGTCTGGGAACTC	TGAGGGGTGAAAGTGGAGA
NM_010295.2 glutamate‐cysteine ligase catalytic subunit	CGAGGATGAGCAATGACCTT	TCAGCCTGCTCCAACTTGAC
NM_008129.4 glutamate‐cysteine ligase modifier subunit	GACCTGGAACTGGATGATGA	TTGCTGGTGCTTCTCGTTCT
NM_008706.5 NAD(P)H quinone dehydrogenase 1	GAGCTGTCTGGCTCTTGTCG	TCTGGAAGCTGCTGTTTGCT
NM_010442.2 heme oxygenase‐1	ACGATGATGCTGGTGGACTA	TTCTGGGCTGTCAGGAGTTC

2.14. Immunofluorescence and mitochondrial staining

Cells grown on coverslips were fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X‐100 (Merck Life Science), and blocked with 5% donkey serum (Jiangsu Kewei Biotechnology, China). Primary antibodies were incubated overnight at 4℃ and then with Alexa Fluor‐conjugated secondary antibodies (1:1000, Thermo Fisher Scientific). Mitochondria were labeled with MitoTracker Deep Red FM (200 nmol/L, Beyotime Biotechnology). Images were obtained using a Zeiss LSM 880 confocal microscope (Germany).

2.15. Statistical analysis

Data are presented as mean ± standard deviation (SD). Normality and homogeneity of variance were assessed using Shapiro–Wilk and Brown–Forsythe tests, respectively. One‐way analysis of variance (ANOVA) with Tukey's post hoc test was performed using GraphPad Prism 9.5. p < 0.05 was considered statistically significant.

3. RESULTS

3.1. Effects of LPS stimulation on mammary inflammation, oxidative stress in mice, and isolation and identification of MAMs in the mammary gland

A mouse mastitis model was developed using LPS treatment. In the LPS group, protein levels of the inflammatory factors inducible nitric oxide synthase (iNOS) and cyclooxygenase‐2 (COX‐2) (Figure 1A–C) and the gene expression of interleukin‐1β (IL‐1β), interleukin‐6 (IL‐6), and tumor necrosis factor‐alpha (TNF‐α) also increased significantly (Figure 1D–F), indicating a successful LPS‐induced mouse mastitis model. To assess redox levels in the mastitis mouse mammary gland, H₂O₂ levels, total antioxidant capacity, and MDA levels were measured using assay kits. Results showed that H₂O₂ levels in the mammary gland of mastitis mice significantly increased (Figure 1G), total antioxidant capacity significantly decreased (Figure 1H), and MDA levels significantly upregulated (Figure 1I), indicating oxidative stress due to redox imbalance in the mammary tissue of mastitis mice. Due to the key role of MAMs in many physiological processes, we explored the changes in MAMs of mastitis mouse mammary tissue. MAMs were successfully extracted from mammary tissue using gradient density centrifugation and verified (Figure 1J).

FIGURE 1.

Establishment of LPS (lipopolysaccharide)–induced mammary inflammation model and characterization of oxidative stress and MAMs (mitochondria‐associated endoplasmic reticulum membranes). (A–C) Protein expression and quantification of iNOS (inducible nitric oxide synthase) and COX‐2 (cyclooxygenase‐2) in mouse mammary glands. (D–F) mRNA (messenger RNA) expression of IL‐1β (interleukin‐1β), IL‐6 (interleukin‐6), and TNF‐α (tumor necrosis factor‐alpha) in mouse mammary glands. Mammary tissues from mastitis mice (n = 5) and control mice (n = 5) were collected to detect (G) H₂O₂, (H) total antioxidant capacity, and (I) MDA (malondialdehyde) using assay kits. (J) Purity analysis of extracted MAMs from mastitis and control mouse mammary tissues. n = 3; data are presented as mean ± SD (standard deviation).

3.2. Proteomic quality control and analysis of MAMs in mouse mammary glands

Proteomic analysis was performed to compare protein expression in MAMs from two mouse mammary gland groups. Proteomic quality control showed high numbers of identified proteins, good coverage, high reproducibility, stability, and accurate quantification (Figure 2A–E). Principal component analysis completely separated proteins from control and mastitis groups (Figure 2F). Volcano plots from the proteomic analysis of control and mastitis mouse mammary glands revealed differential protein expression, with 74 upregulated and 2039 downregulated proteins (Figure 2H), including significantly altered MFN2 and PACS2. The top 50 upregulated and downregulated proteins ranked by |log₂FC| are shown in Figure 2G. Column charts in Figure 2I–K show the second‐level Gene Ontology (GO) functional protein counts for differentially expressed proteins between the two groups, along with their GO and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment. NES indicates normalized enrichment scores.

FIGURE 2.

Proteomic profiling of mammary MAMs (mitochondria‐associated endoplasmic reticulum membranes) reveals LPS (lipopolysaccharide)–induced differential protein expression and functional enrichment. (A) Peptide length distribution. (B) Unique peptide count distribution. (C) Protein coverage distribution. (D) Molecular weight and isoelectric point distribution of proteins. (E) Sample correlation bubble plot. (F) PCA (principal component analysis) score plot of samples. (G) Differential protein heat map. (H) Volcano plot of differential gene expression (DEPs). (I) Bar chart of second‐level GO (Gene Ontology) functional protein counts. (J) GO enrichment analysis. (K) KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway enrichment analysis.

Proteomic analysis was designed to explore the activation of cellular oxidative stress and organelle membrane functions and the nuclear factor‐κB pathway in the mammary glands of mastitic dairy cows. Subsequently, the gene set enrichment analysis of MAM proteins in the control group and the inflammatory group was performed based on the GO and KEGG databases. As a result, the GO and KEGG gene sets were enriched, and their ES values were normalized and permutated. The GO analysis found that differentially expressed proteins were significantly enriched in membrane structures, such as mitochondrial membranes, ER membranes, and Golgi membranes (Figure 3A). The KEGG enrichment analysis identified 55 tertiary differential pathways (p < 0.05), including those related to MAMs and oxidative stress (Figure 3B). The butterfly plot revealed that most differentially expressed genes were downregulated (Figure 3C), and the histogram indicated the overall level of activated pathways (Figure 3D). The proteomic results suggested that MFN2 and PACS2 were significantly downregulated in the model group (Figure 3E,F), indicating that MFN2 and PACS2 may play crucial roles in LPS‐induced mammary inflammation.

FIGURE 3.

GSEA (gene set enrichment analysis) uncovers LPS (lipopolysaccharide)–induced membrane structure dysfunction and downregulation of MFN2 (mitofusin 2)/PACS2 (phosphofurin acidic cluster sorting protein 2) in MAMs (mitochondria‐associated endoplasmic reticulum membranes). (A) GSEA GO (Gene Ontology) enrichment plot. (B) GSEA KEGG (Kyoto Encyclopedia of Genes and Genomes) enrichment plot. (C) Butterfly plot of key genes. (D) Comparison of GSEA GO p‐values and NES (normalized enrichment scores). (E and F) Protein expression of MFN2 and PACS2. Data are presented as mean ± SD (standard deviation).

3.3. Effects of LPS stimulation on inflammation, oxidative stress, and MFN2 protein expression in mouse mammary epithelial cells

Previous studies have indicated that intraductal infusion of LPS into mouse teats triggers mastitis and oxidative stress in mice. To verify whether these effects also occur at the cellular level, we established an inflammatory model of mMECs using LPS. Western blot analysis revealed that protein levels of the inflammatory factors iNOS and COX‐2 increased in a time‐dependent manner after LPS stimulation (Figure 4A–C). Consistently, quantitative reverse transcription polymerase chain reaction (qRT‐PCR) results demonstrated significant upregulation in the gene expression of IL‐1β and IL‐6 (Figure 4D,E), confirming that LPS stimulation induces an inflammatory response in mMECs. Subsequently, the levels of catalase (CAT) and total antioxidant capacity in the cells were measured using assay kits. Figure 4F,G shows that both parameters decreased significantly with prolonged LPS stimulation. Further qRT‐PCR analysis of the expression of glutamate‐cysteine ligase catalytic subunit (GCLC), glutamate‐cysteine ligase modifier subunit (GCLM), and heme oxygenase‐1 (HO‐1) revealed that their gene expression levels increased with extended LPS stimulation (Figure 4H–J), indicating redox imbalance induced by LPS. To validate the proteomic findings, the protein expression of MFN2 in mammary tissue was examined. Consistent with the proteomic data, LPS stimulation resulted in a significant decrease in MFN2 levels within the mammary tissue (Figure 4K,L). Additionally, to evaluate the applicability of these findings in vitro, MFN2 expression in mMECs stimulated with LPS for various durations was assessed. Results indicated that MFN2 protein levels decreased in a time‐dependent manner after LPS stimulation (Figure 5A,B).

FIGURE 4.

LPS (lipopolysaccharide) triggers inflammation, oxidative imbalance, and time‐dependent MFN2 (mitofusin 2) suppression in mammary epithelial cells. (A–C) Protein expression and quantification of iNOS (inducible nitric oxide synthase) and COX‐2 (cyclooxygenase‐2) in mouse mammary epithelial cells (mMEC). (D and E) mRNA (messenger RNA) expression of IL‐1β (interleukin‐1β) and IL‐6 (interleukin‐6) in mMECs. (F) CAT (catalase) levels detected using an assay kit. (G) Total antioxidant capacity detected using an assay kit. (H–J) mRNA expression of GCLC (glutamate‐cysteine ligase catalytic subunit), GCLM (glutamate‐cysteine ligase modifier subunit), and HO‐1 (heme oxygenase‐1) in mammary tissues detected via qRT‐PCR (quantitative reverse transcription polymerase chain reaction). (K and L) Protein expression and quantification of MFN2 in mammary tissues from mastitis mice (n = 3) and control mice (n = 3) detected using Western blot. Data are presented as mean ± SD (standard deviation).

FIGURE 5.

LPS (lipopolysaccharide) modulates mitochondrial‐ER (endoplasmic reticulum) membrane protein expression and disrupts MAM (mitochondria‐associated endoplasmic reticulum membrane) integrity. (A and B) Protein expression and quantification of MFN2 (mitofusin 2) in mouse mammary epithelial cells (mMEC) detected using Western blot. (C–F) Protein expression and quantification of FIS‐1 (fission protein 1), DRP‐1 (dynamin‐related protein 1), and OPA‐1 (optic atrophy 1) in mMECs detected using Western blot. (G–I) Protein expression and quantification of eIF‐2α (eukaryotic translation initiation factor 2 alpha), p‐eIF‐2α, and PACS2 (phosphofurin acidic cluster sorting protein 2) in mMECs. (J) Protein expression of MFN2 and PACS2 in mMECs after LPS stimulation detected using immunofluorescence. (K) Mitochondrial and ER marker protein PDI (protein disulfide isomerase) expression detected using laser confocal microscopy. Data are presented as mean ± SD (standard deviation).

3.4. Effects of LPS on MAMs in mMECs

MAMs, formed by the outer mitochondrial membrane and the ER membrane, are essential for material exchange and communication. In this study, Western blot was first used to detect the expression of MAM‐related proteins in mitochondria, namely fission protein 1 (FIS‐1), dynamin‐related protein 1 (DRP‐1), and optic atrophy 1 (OPA‐1). The results indicated that FIS‐1 and DRP‐1 expression increased with LPS stimulation, whereas OPA‐1 expression decreased over time (Figure 5C–F). Subsequently, MAM‐related proteins in the ER were examined, revealing that LPS promoted eukaryotic translation initiation factor (eIF) phosphorylation and reduced PACS2 protein levels (Figure 5G–I). Additionally, mitochondrial‐protein disulfide isomerase (PDI) co‐staining was employed to assess MAM content. The results showed that whereas the ER structure remained largely unchanged, mitochondria were significantly reduced after LPS stimulation, leading to a significant decrease in MAM structures (Figure 5J–K).

3.5. Effects of MFN2 overexpression on LPS‐treated mMECs

We first overexpressed MFN2 and assessed cellular inflammation levels. Western blot analysis revealed that LPS stimulation upregulated the protein levels of the inflammatory factors iNOS and COX‐2, but this effect was reversed by MFN2 overexpression (Figure 6A–C). Subsequent qRT‐PCR analysis of pro‐inflammatory cytokines IL‐1β, IL‐6, and TNF‐α showed that LPS‐induced upregulation of these genes was also mitigated by MFN2 overexpression (Figure 6D–F). This indicates that overexpressing MFN2 can suppress LPS‐driven increases in pro‐inflammatory factors. Additionally, using antioxidant assay kits, we found that LPS reduced cellular CAT levels and total antioxidant capacity while upregulating antioxidant‐related enzymes. However, these changes were reversed by MFN2 overexpression (Figure 6G–L), suggesting that overexpressing MFN2 can restore redox balance in mMECs disrupted by LPS.

FIGURE 6.

MFN2 (mitofusin 2) overexpression attenuates LPS (lipopolysaccharide)–induced inflammation and restores redox homeostasis. (A–C) Protein expression and quantification of iNOS (inducible nitric oxide synthase) and COX‐2 (cyclooxygenase‐2) in mouse mammary epithelial cells (mMEC). (D–F) mRNA (messenger RNA) expression of IL‐1β (interleukin‐1β), IL‐6 (interleukin‐6), and TNF‐α (tumor necrosis factor‐alpha) in mMECs. (G) CAT levels detected using an assay kit. (H) Total antioxidant capacity detected using an assay kit. (I–L) mRNA expression of NQO1 (quinone dehydrogenase 1), GCLC (glutamate‐cysteine ligase catalytic subunit), GCLM (glutamate‐cysteine ligase modifier subunit), and HO‐1 (heme oxygenase‐1) in cells detected using qRT‐PCR (quantitative reverse transcription polymerase chain reaction). OE denotes MFN2 overexpression. (M) Yellow represents the predicted MFN2 protein structure; blue represents the predicted PACS2 (phosphofurin acidic cluster sorting protein 2) protein structure; purple represents amino acids on MFN2 close to PACS2; green represents amino acids on PACS2 close to MFN2; dashed lines indicate predicted intermolecular interactions. Data are presented as mean ± SD (standard deviation).

3.6. Research on the interaction mechanism between MFN2 and PACS2

PACS2 and MFN2, both MAM proteins, exhibit reduced expression during oxidative stress in mammary tissue, suggesting a potential interaction between them. Initially, we utilized AlphaFold3 to predict the molecular structures of PACS2 and MFN2 and their interaction levels, with results visualized using ChimeraX, version 1.8 (Figure 6M). The prediction identified several intermolecular forces between the two proteins. Immunofluorescence double staining for PACS2 and MFN2 also revealed decreased colocalization after LPS stimulation (Figure 5J). These findings imply a possible interaction between PACS2 and MFN2, which may be linked to MFN2's mechanism in alleviating oxidative stress. To further explore this interaction, co‐immunoprecipitation (CO‐IP) experiments were conducted. Results demonstrated that PACS2 was co‐precipitated with MFN2 and vice versa (Figure 7A,D), confirming an interaction between PACS2 and MFN2.

FIGURE 7.

MFN2‐PACS2 (mitofusin 2–phosphofurin acidic cluster sorting protein 2) interaction regulates MAM (mitochondria‐associated endoplasmic reticulum membrane) dynamics and mitigates LPS (lipopolysaccharide)–induced mitochondrial dysfunction. (A) CO‐IP (co‐immunoprecipitation) of MFN2 in mouse mammary epithelial cell (mMEC) lysates. (B) Immunoprecipitation of MFN2 in mMEC lysates. OE‐MFN2 denotes MFN2 overexpression. (C) Immunoprecipitation of MFN2 in mMEC lysates. (D) CO‐IP of PACS2 in mMEC lysates. (E) mRNA (messenger RNA) expression of MFN2 in mammary tissues detected using qRT‐PCR (quantitative reverse transcription polymerase chain reaction). (F) Immunoprecipitation of MFN2 in mMEC lysates. (G) Protein expression and quantification of DRP‐1 (dynamin‐related protein 1), OPA‐1 (optic atrophy 1), FIS‐1 (fission protein 1), and MFN1 in mMECs detected using Western blot. (H and I) Mitochondrial and ER marker protein PDI (protein disulfide isomerase) expression detected using laser confocal microscopy.

3.7. Impact of MFN2–PACS2 interaction on MAMs

We delved into the roles of PACS2 and MFN2 in MAM formation by assessing changes in their interaction upon MFN2 overexpression (Figure 7E). CO‐IP results showed that MFN2 overexpression strengthened MFN2–PACS2 interaction (Figure 7B). Conversely, PACS2 knockdown reduced this interaction (Figure 7C). This indicates that the expression levels of MFN2 and PACS2 influence their interaction. Immunofluorescence analysis further revealed that PACS2 knockdown significantly reduced MAM levels (Figure 7H).

3.8. MFN2 overexpression reverses LPS‐induced MAM reduction

To explore MFN2's role in MAM regulation, we investigated the interaction between MFN2 and PACS2 after MFN2 overexpression. Results showed that whereas LPS weakened their interaction, MFN2 overexpression reversed this effect (Figure 7F). Western blot analysis of mitochondrial MAM‐related proteins revealed that LPS increased DRP‐1 and FIS‐1, and decreased OPA‐1 and MFN1. However, MFN2 overexpression counteracted these changes (Figure 7G). Additionally, mitochondrial and ER marker protein PDI colocalization was assessed to examine MAM morphology. Consistent with Western blot results, MAMs were reduced in the LPS group but preserved with MFN2 overexpression (Figure 7H,I). In summary, MFN2 overexpression can reverse LPS‐induced MAM reduction.

4. DISCUSSION

In this study, we found that the inflammatory level increased significantly, whereas the total antioxidant capacity decreased significantly in the LPS‐treated mouse model of mastitis and mMECs. And the expression of MAM‐related proteins MFN2 and PACS2 was significantly downregulated. In cell experiments, overexpression of MFN2 can inhibit inflammation and oxidative stress responses, and promote the interaction between MFN2 and PACS2 to affect the formation of MAMs. These results further indicate that MFN2 and PACS2 may be key targets of LPS‐induced mastitis.

MDA, the final product of lipid peroxidation, is an important marker of oxidative stress. ²⁵ CAT is one of the important antioxidant enzymes in cells. CAT could reduce the damage of oxidative stress in cells. Our results showed that LPS stimulation led to a significant decrease in the level of CAT in cells, indicating that the antioxidant defense mechanism of cells was damaged. GCLC and GCLM are key enzymes for glutathione (GSH) synthesis, and GSH is one of the most important antioxidant molecules in cells. ²⁶ , ²⁷ In our study, the gene expression levels of GCLC and GCLM were significantly upregulated with the prolongation of LPS stimulation time, indicating that cells increase GSH synthesis to cope with oxidative stress. HO‐1, an important antioxidant enzyme, exerts antioxidant and anti‐inflammatory effects by degrading heme to produce carbon monoxide, biliverdin, and iron ions. ²⁸ , ²⁹ Our results showed that the gene expression level of HO‐1 increased significantly with the prolongation of LPS stimulation time. However, excessive expression of HO‐1 may also lead to the accumulation of iron ions to exacerbate oxidative stress. Therefore, the dual role of HO‐1 in oxidative stress still needs further study.

MAMs, dynamic connection structures between the ER and mitochondria, play an irreplaceable role in the physiological processes of cells. ³⁰ , ³¹ , ³² However, their role in oxidative stress in the mammary glands of LPS‐induced mastitis mice remains unclear. Among the differentially expressed proteins screened based on MAM proteomic analysis, the differential expression of MFN2 and PACS2 was statistically significant. As a key regulatory factor for mitochondrial fusion, a downregulated expression of MFN2 has been proven to lead to mitochondrial fragmentation and an increase in ROS production. ³³ , ³⁴ PACS2, a protein that maintains the MAM structure, may affect the cellular oxidative stress state through disrupting calcium homeostasis and mitochondrial autophagy pathways. ³⁵ , ³⁶ This mechanism has not been systematically reported in mastitis. And protein–protein interaction network analysis showed that these were at the core nodes of the MAM function regulation network and had a significant co‐expression relationship with key molecules in antioxidant pathways such as Nrf2 and superoxide dismutase.

MFN2 is a key regulatory protein for mitochondrial fusion. ³⁷ It not only participates in the regulation of mitochondrial dynamics but also plays important roles in cell metabolism, apoptosis, and immune responses. ³⁸ In this study, LPS stimulation decreased the protein level of MFN2 significantly. It has been reported that the protein level of MFN2 decreased significantly in an LPS‐induced inflammation model related to mitochondrial dysfunction and exacerbated oxidative stress. ²² However, we found that the gene expression level of MFN2 did not decrease after LPS stimulation but increased significantly after 24 h, suggesting that the decrease in MFN2 is not regulated at the transcriptional level but is achieved through posttranscriptional mechanisms. Posttranscriptional regulation may include the regulation of protein degradation and translation inhibition. Nag et al. ³⁹ found that phosphoglycerate mutase 5 can dephosphorylate MFN2 to promote the formation of the mitochondrial network. Therefore, future research can further explore the posttranscriptional regulation mechanism of MFN2 under LPS stimulation, especially the role of phosphorylation modification. In addition, the gene expression level of MFN2 increased after LPS stimulation for 24 h, suggesting that the gene expression of MFN2 regulates mitochondrial function. The results showed that LPS stimulation downregulated the protein level of PACS2 and upregulated the protein level of iNOS. The mitochondria in cells were stained with a mitochondrial dye, and the ER marker PDI was immunofluorescently labeled. The colocalization of the two fluorescence signals can reflect the integrity of the MAM structure. The results showed that the ER structure of mMECs after LPS stimulation did not change significantly, but the mitochondria were excessively fragmented, and the colocalized fluorescence decreased accordingly. We found that there may be a protein–protein interaction between PACS2 and MFN2, indicating that they may participate in the same biological process in cells. It has been reported that PACS2 knockout can lead to the destruction of the MAM structure and exacerbate the cellular stress response. ⁴⁰ , ⁴¹ We found that overexpression of MFN2 can effectively alleviate the MAM tissue damage and oxidative stress level caused by LPS stimulation. As a key regulatory protein for mitochondrial fusion, the overexpression of MFN2 may enhance the fusion ability of mitochondria to improve mitochondrial function and to stabilize the MAM structure. ⁴² It has been reported that MFN2 regulates the contact between mitochondria and the ER through its interaction with ER‐related proteins to maintain the structural integrity of MAMs. ⁴³ , ⁴⁴ , ⁴⁵ This dual‐regulatory mechanism enables MFN2 to exhibit a synergistic protective effect in response to oxidative stress. It ensures the efficient cooperation between mitochondria and the ER by maintaining the integrity of MAMs. And it blocks the vicious cycle of oxidative damage by reducing ROS accumulation and enhancing antioxidant capacity. Therefore, MFN2 is not only a core regulatory factor of mitochondrial dynamics but also a key node in cellular oxidative stress.

5. CONCLUSION

In summary, we found that the expression of MAM‐related proteins regulated the occurrence and development of inflammation and oxidative stress in mouse mammary tissue. The interaction between PACS2 and MFN2 affects the formation of MAMs. Overexpression of MFN2 can promote the formation of MAMs via regulating PACS to inhibit inflammation and oxidative stress responses in mMECs.

AUTHOR CONTRIBUTIONS

Bingxu Huang: Conceptualization; data curation; formal analysis; methodology; supervision; writing – original draft; writing – review and editing. Xiechen Zhou: Data curation; formal analysis; investigation; methodology; software. Yufei Zhang: Data curation; investigation; methodology; software. He Ma: Data curation; investigation; methodology; software. Shoupeng Fu: Data curation; methodology; project administration; supervision. Juxiong Liu: Data curation; formal analysis; investigation; methodology; supervision; validation. Wenjin Guo: Conceptualization; formal analysis; methodology; project administration; supervision; visualization. Xiaofeng Tian: Investigation; software; supervision.

FUNDING INFORMATION

This work was supported by the National Natural Science Foundation of China (32302826, 32372961) and Jilin Provincial Special Project for Health Research Talents (project no. 2020SCZ40), and the China Postdoctoral Science Foundation (2023M740623).

CONFLICT OF INTEREST STATEMENT

The authors declare that they have no conflict of interest.

ETHICS STATEMENT

All animal experiments were conducted in accordance with guidelines approved by the Animal Ethics and Welfare Committee of Jilin University (approval no. SY202505020).

Zhou X, Zhang Y, Ma H, et al. Overexpression of mitofusin 2 ameliorates inflammation and oxidative stress in lipopolysaccharide‐induced mastitis model by regulating phosphofurin acidic cluster sorting protein 2. Anim Models Exp Med. 2026;9:154‐167. doi: 10.1002/ame2.70110