Effects of lipid-lowering drug on vancomycin-sensitive Corynebacterium kroppenstedtii

Abstract

Background

Granulomatous mastitis (GM) is a rare inflammatory breast disease primarily affecting reproductive-aged women with a history of breastfeeding. Within GM lesions, Corynebacterium kroppenstedtii has frequently been isolated, suggesting it may play a significant role in the development of GM. It is important to develop the novel therapeutic regimen to fight against the Corynebacterium kroppenstedtii infection.

Results

In our study, the combination of vancomycin and the lipid-lowering drug Tetrahydrolipstatin (THL) exhibited a synergistic effect against bacteria. Proteomic analysis revealed that potential targets of THL are involved in intermediary metabolism, cellular respiration, lipid metabolism, signaling pathways, virulence, and detoxification and adaptation processes.

Conclusions

The combination of THL and vancomycin could be a novel therapeutic regimen to enhance clinical outcomes and reduce the recurrence of GM.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12866-026-04844-9.

Introduction

Granulomatous mastitis (GM) is a chronic inflammatory condition characterized by the invasion of breast lobules, with the formation of granulomas in breast tissue serving as the primary pathological manifestation [1]. Clinically, it is also referred to as granulomatous lobular mastitis (GLM). GLM typically occurs in postpartum women of childbearing age and has a high incidence [2, 3]. It manifests as chronic focal inflammation, presenting diverse clinical features, often lacking typical characteristics on imaging examinations, which can lead to misdiagnosis and complicate clinical diagnosis [3]. The exact etiology and pathogenesis of GLM remain unclear, but potential contributing factors may include autoimmunity, infections, oral contraceptives using, pregnancy, childbirth, breastfeeding, hormonal imbalances (such as hyperprolactinemia), smoking, ethnic differences or α-1 antitrypsin deficiency [4]. Recently, microbial infection has emerged as a significant etiological hypothesis [5].

In 2002, Paviour et al.. reported the isolation of Corynebacterium species from cases of mastitis, including Corynebacterium kroppenstedtii (C. kroppenstedtii), Corynebacterium amycolatum, Corynebacterium tuberculostearicum [6, 7], suggesting a correlation between GLM and Corynebacteria infection. Many researchers subsequently identified C. kroppenstedtii as the most common pathogenic bacterium isolated from GLM cases [4, 7–11], indicating that its infection may be a key factor in the development of GLM.

Corynebacterium is classified as a distinct group of Gram-positive bacteria characterized by a complex cell envelope structure. The corynomycolates (short-chain α-split-chain and β-hydroxy fatty acids) are esterified into arabinogalactan polysaccharides, which covalently bind to the peptidoglycan core of the cell wall, forming mycolic acid-galactan-peptidoglycan complexes [12]. The corynomycolic acids linked to the cell membrane create a unique outer “mycomembrane” enriched with glycolipids, polysaccharides, and various proteins, including S-layer proteins and pili, formed based on trehalose. It was reported that the suprabacteric acid on the cell envelope is associated with stress resistance and pathogenicity in both Corynebacterium and Mycobacterium strains [13–16].

C. kroppenstedtii, belonging to the genus Corynebacterium, is a lipophilic bacterium with special growth requirements, exhibiting slow or negligible growth on traditional media and necessitating a high-lipid environment for optimal proliferation [17, 18]. Unlike other Corynebacteria, the outer layer of C. kroppenstedtii lacks the typical corynomycolic acid structure, which may contribute to its lipophilicity [19]. Lipophilicity is a significant pathogenic feature of C. kroppenstedtii and the high fat content in female breast tissue provides an ideal environment for its growth, making postpartum mastitis in women often associated with C. kroppenstedtii infection [18].

In clinic, treatment with wide local excision (WLE), steroids or other immunosuppressives has been advocated, but recurrence remains a major issue [1, 20, 21]. In most circumstances, the recurrence for blind lumpectomy is worse than the original presentation. Steroid usage is associated with a better outcome than lumpectomies, but has a poorer outcome than WLE with en-bloc total/partial duct excision, and is more unpredictable. The long-term side effects are often problematic for young women, although steroid use continues to have a role in predominantly lump-only lesions in the young age group. Treatment with methotrexate is another option for conservative management, but has limited evidence [1]. If Corynebacteria are the causative factor in the development of GLM, the latter might call for a first line of treatment by lipophilic antimicrobials [22]. Considering its unique molecular structure, C. kroppenstedtii shares certain morphological and taxonomic similarities with Mycobacterium tuberculosis (M. tuberculosis). This bacterium causes disease by surviving within lipid vacuoles in neutrophil-infiltrated granulation tissue rather than in inflammatory tissue. As a result, only drugs with high lipophilicity and high volume of distribution can achieve sufficient inhibitory concentrations in this environment [23]. Currently, conventional antibacterial agents, hormonal treatments, and surgical interventions for GLM patients often yield limited effects, as commonly used clinical drugs are typically water-soluble antibiotics, leading to a high recurrence rate of the disease and considerable distress for patients [24–26]. Therefore, there is an urgent need to identify new drugs targeting C. kroppenstedtii infections and develop novel treatment modalities. To our knowledge, no evidence-based guidelines have been proposed to date to support appropriate antibiotic therapy for C. kroppenstedtii infections. Some scholars advocate for a triple antibacterial regimen (isoniazid, rifampicin, ethambutol) to treat breast inflammation resulting from C. kroppenstedtii infection, as this approach can inhibit bacterial growth, reduce local breast inflammation and yield satisfactory outcomes [27]. However, attention must be paid to the complexities of this treatment process (long duration, high drug doses, and multidrug combinations), as well as potential side effects such as liver damage and failure.

Tetrahydrolipstatin (THL), a synthetic Lipstatin derivative, is a universal lipase inhibitor that prevents cell proliferation and leads to apoptosis by inhibiting the activity of fatty acid synthetase, and is clinically used as a diet drug [28]. In addition to its lipase inhibition properties, THL also acts as an effective esterase inhibitor, successfully targeting mycobacterial serine esterase and lipoesterase activities, including those of Ag85C, Rv3802, Pks13-TE, and other proteins [29, 30]. Our recent study has demonstrated that THL inhibits TesA, a crucial enzyme involved in the biosynthesis of phthiocerol dimycocerosate (PDIM), a key virulence factor for the mycobacterial cell wall (unpublished data). Interestingly, the combination of vancomycin and THL has been shown to effectively inhibit the growth of M. tuberculosis by disrupting cell wall synthesis and increasing the permeability of the cell wall [31]. Given the structural similarities between Mycobacterium and Corynebacterium, the combination of vancomycin and THL as a novel medication regimen for GM treatment caused by C. kroppenstedtii infection has potential clinical application.

In the present study, we aim to investigate the effects of combining THL and vancomycin on the growth of C. kroppenstedtii, analyze potential drug targets, evaluate the potential therapeutic value against C. kroppenstedtii infection, and provide an experimental basis for clinical application of this drug regimen.

Results

Drug susceptibility assay

In this study, we firstly evaluated the inhibition efficiency of THL and vancomycin against C. kroppenstedtii. The result indicated that the THL exhibited an inhibition range with a MIC of 250–500 µg/mL, while vancomycin showed a MIC of 1.6 µg/mL. According to the Clinical and Laboratory Standards Institute (CLSI) guideline M45 (https://clsi.org/standards/products/microbiology/documents/m45/), these susceptibility result indicate that the C. kroppenstedtii is sensitive to vancomycin. Interestingly, THL increased the susceptibility of C. kroppenstedtii to vancomycin by more than fourfold in the drug susceptibility assays.

To investigate whether this growth inhibition was due to a synergistic effect between THL and vancomycin, we employed the checkerboard method to calculate the fractional inhibitory concentration index (FICI) (Table 1). In the presence of vancomycin (0.2 µg/mL), the MIC of THL decreased from 250 to 500 µg/mL to 31.25–62.5 µg/mL. Conversely, the MIC of vancomycin reduced from 1.6 µg/mL to 0.08 µg/mL in the presence of THL (100 µg/mL), suggesting that the combination can inhibit the growth of C. kroppenstedtii in synergistic manner (FICI = 0.1125-0.3).

Table 1.

Vancomycin susceptibility obtained in macrodilution series and analysed by the checkerboard method for C. kroppenstedtii

Drug	MIC (µg/mL)/FICI*
Vancomycin	1.6 /-
THL	250–500/-
THL (100 µg/mL) +Vancomycin	0.08 /0.05
Vancomycin (0.2 µg/mL) + THL	31.25–62.5 / 0.0625-0.25

The data shown are from one representative of at least four independent experiments

* The synergy was defined as a FICI ≤ 0.5, indifference as a FICI between > 0.5 and 4, antagonism as a FICI > 4

Bacterial growth curve

To assess the bacterial growth state, we measured the optical density at 600 nm (OD₆₀₀) of C. kroppenstedtii at designated time points. It was showed that the bacterial growth displayed the similar growth manners during the first 6 h, after which the bacteria entered a logarithmic phase characterized by rapid growth. From the 32th hours, the bacteria reached a plateau phase. Notably, the bacterial growth rate was lower in the presence of vancomycin/THL combination than the control or THL alone. As the concentration of THL increased in the experimental conditions, bacterial growth decreased in a dose-dependent manner when compared to the control or THL alone (Fig. 1). These observations suggest that the THL enhances the susceptibility of C. kroppenstedtii to vancomycin.

Fig. 1.

Drug susceptibility assay on C. kroppenstedtii. Van (1): vancomycin 1 µg/mL; THL(100): THL 100 µg/mL; THL(200): THL 200 µg/mL; THL(400): THL 400 µg/mL

An absorbance experiment was performed to record the normal and delayed growth of C. kroppenstedtii with or without drugs. The data were obtained from three independent experiments.

Cell infection model

For further investigate the impact of THL/vancomycin or vancomycin alone, we established a C. kroppenstedtii infection model using murine macrophage Raw146.7 cell. The RAW 264.7 cell line is tumorigenic with macrophage differentiation and can be used in oxidative stress, inflammatory, and antibacterial activity studies. It was observed that the growth of Raw264.7 cells reached a logarithmic growth phase within 24 to 48 h in DMEM medium. Under microscopy, the Raw264.7 cells appear round or oval in shape, with diameters ranging from approximately 10 to 30 μm. After infection with the bacteria at a multiplicity of infection (MOI) of 1:10, the cell began to exhibit shrinkage and weak adhesion.

Firstly, the cell viability was assessed using the MTT colorimetric assay as previously described [32]. After treatment with THL, vancomycin or vancomycin/THL combination, 30µL MTT solution (80%, m/v) was added, followed by incubation in the dark for 30 min. No significant color changes were observed compared to the negative control across all drug concentrations tested, even at the highest dose.

Then, various drugs were added into the infected cells for 24 h incubation. Compared to the control group, the morphology of the infected cells treated with drugs gradually recovered, returning to a more typical round or oval shape, particularly in the presence of the combination of vancomycin and THL. Notably, the number of the surviving bacteria decreased progressively with increasing drug dose in a dose-dependent manner (Fig. 2).

Fig. 2.

The C. kroppenstedtii was analysed by enumerating the CFU counts at different time points post drug exposure. THL100: THL 100 µg/mL; THL200: THL 200 µg/mL; THL400: THL 400 µg/mL. Mean values from three biological experiments ± SD are shown. The p values were calculated by Students t-test, compared with the group vancomycin alone. Van: vancomycin, *: p < 0.05

Proteomic analysis

In the present study, we employed a label-free quantitative proteomic approach to investigate the differentially expressed proteins between THL-treated bacteria and the control group. A total of 1,749 proteins were identified in both groups, among which 74 differentially expressed proteins (fold change [FC] > 1.2 or FC < 0.83, and p < 0.05) were found in THL-treated bacteria. This included 43 up-regulated proteins and 31 down-regulated proteins (Fig. 3). For instance, the dolichyl-phosphate-mannose-protein mannosyltransferase, encoded by the gene HMPREF1861_01610, showed a 2.26-fold increase in expression compared to untreated bacteria. Similarly, phosphomethylpyrimidine synthase (Thic) was overexpressed by 1.99 times relative to the control bacteria. Conversely, the serine protease, encoded by the gene DI525_03295, showed reduced expression in bacteria without drug treatment. Additionally, pyridoxamine 5-phosphate oxidase family proteins, encoded by the genes HMPREF1861_01996 and CJ202_07450, respectively, were also down-regulated in THL treated bacteria. Overall, the majority of the proteins regulated following THL treatment were protease, including serine protease, hydrolase, esterase, dehydratases. Notably, more than 13 domains of unknown function (DUF) proteins, such as DUF-4288, DUF-4177, DUF-349, were also found to be regulated in THL-treated bacteria (Table S1).

Fig. 3.

Proteomic analysis the THL-treated C. kroppenstedtii and control. (A) The number of identified proteins in the THL-treated bacteria and control; (B) the number of proteins differently expressed in the THL-treated bacteria compared with controls; (C) the volcano plot of proteomic data in the THL-treated bacteria compared with controls; (D) the KEGG functional enrichment analysis of differentially expressed proteins (TOP 20) in the THL-treated bacteria

Gene ontology (GO) analysis revealed among the 114 differentially expressed proteins in biological process, 46 were associated with metabolic processes. Additionally, out of 126 proteins identified in molecular function, 77 were involved in catalytic activity (Fig. 4).

Fig. 4.

KEGG functional enrichment analysis the down-regulated proteins in THL-treated C. kroppenstedtii

Kyoto Encyclopedia of Genes and Genomes (KEGG) functional enrichment analysis further demonstrated that differentially down-regulated proteins were predominantly associated with nucleotide sugars biosynthesis, amino sugar and nucleotide sugar metabolism, fructose and mannose metabolism, O-antigen nucleotide sugar biosynthesis and the oxidative phosphorylation. In contrast, the up-regulated proteins were primarily involved in the sulfur metabolism, selenocompound metabolism, monobactam biosynthesis and purine metabolism (Fig. 5).

Fig. 5.

The GO analysis of differentially expressed proteins in THL-treated C. kroppenstedtii from proteomic analysis. BP, biological process; CC, pellular pomponent; MF, molecular function

Discussion

GM is a rare benign inflammatory disease of the breast that typically occurs in females of childbearing age and can clinically mimic breast cancer [33]. Some reporters demonstrated that human breast tissue is not sterile and harbors a diverse microbiome, including Corynebacteria, especially C. kroppenstedtii, and to a lesser extent Staphylococcal spp [34]. Corynebacterium species are characterized by a complex cell wall architecture which is composed of a peptidoglycan layer, which itself is covalently linked to arabinogalactan, an additional heteropolysaccharide meshwork in plasma membrane, an outer layer of mycolic acids (corynomycolic acids) functionally equivalent to the outer membrane of Gram-negative bacteria [35]. As top layer, outer surface material is composed of free polysaccharides, glycolipids, and proteins (including S-layer proteins, pili, and other surface proteins) [35]. Among these, the C. kroppenstedtii can serve triacylglycerols, producing di- or monoacylglycerol, glycerol, free fatty acid as precursors for the synthesis of new phospholipids or as carbon sources for growth. Due to the absence of corynomycolic acids in its cell envelope, the bacteria might reduce the need for exogenous fatty acids for growth. The absence of certain genes, such as pks, cmrA, fadD, and accD3, contributes to its unique lipophilic properties [36]. The lipid uptake systems, which facilitate the uptake glycerol (glpF) and glycerol-3-phosphonate (glpT and ugpAEBC), are linked to the lipophilic phenotype of C. kroppenstedtii [37]. Therefore, these bacteria can easily infect human adipose tissue, potentially lead to mastitis, colonization in fat lobules may help them escape neutrophils.

Currently, there is no consensus in the literature regarding the treatment for GM. Proposed clinical treatment regimens for GM caused by C. kroppenstedtii infections include steroids, immunosuppressants, antibiotic therapy, surgical debridement, and watchful waiting [38–40]. Considering the lipophilic properties of C. kroppenstedtii, only lipophilic antibiotic with a high volume of distribution, such as doxycycline, clarithromycin and rifampicin, may reach adequate therapeutic concentrations within the lipogranulomas of the lipid-rich breast [27]. However, the commonly clinical antibiotics are often water-soluble, which can lead to recurrent infection and significant patient discomfort [24–26]. Moreover, corynebacteria display sensitivity to vancomycin, doxycycline, and fusidic acid, while showing poor sensitivity to β-lactam antibiotics traditionally prescribed for breast infections, and many are multidrug resistant [41]. Moreover, a triple antibacterial therapy regimen involving isoniazid, rifampicin, and ethambutol is recommended for treating C. kroppenstedtii infections [27]. Additionally, it was also reported that the rifampicin is effective in the treating patients with idiopathic granulomatous lobular mastitis (IGLM) with complete clinical and ultrasonographic response after 6–9 months [42]. Combination antibiotics with corticosteroid may also help inhibit granulomatous disease development, for instance, the combination of a steroid (triamcinolone acetonide) and an antibiotic (cefuroxime) [27]. Therefore, it is necessary to find the novel regimen to treat the GM infection by C. kroppenstedtii.

The glycopeptide antibiotic vancomycin has been considered as the “last resort” treatment of severe infections caused by Gram-positive bacteria by blocking the cell wall syntheisis and impeding RNA synthesis, commonly used in clinical practice [43]. In the present study, we observed that the MIC of vancomycin on C. kroppenstedtii was 1.6 µg/mL, indicating the experimental bacteria was sensitive to vancomycin according to the CLSI guideline M45. THL, also called Orlistat, aderivative of lipstatin, acts as an inhibitor of gastric and pancreatic lipases. It is also a covalent inhibitor of many serine esterases, especilly involved into the cell wall lipid, in bacteria [29, 44], suggesting that THL also can affect the bacteria cell wall lipid.

In the present study, we evaluated the lower-lipid drug THL for its potential inhibitory effects against C. kroppenstedtii. From the proteomic analysis, it was observed that more than 140 proteins showed significant difference in expression in THL-treated C. kroppenstedtii compared to the control. The differentially expressed proteins identified through proteomic analysis provide a key molecular basis for the synergistic bacteriostatic effect of THL and vancomycin against C. kroppenstedtii by regulating the metabolic pathways, cell wall structural integrity, and stress compensation ability of the bacterium. The specific correlation mechanisms are as follows:

Glucose metabolism-related differentially expressed proteins-disrupting the supply of cell wall synthesis precursors and enhancing vancomycin permeability

Proteomic results showed that after THL treatment, proteins involved in nucleotide sugar synthesis, amino sugar and nucleotide sugar metabolism, and fructose and mannose metabolism pathways in bacteria were significantly down-regulated (e.g., changes in the expression of glucosamine-6-phosphate (GlcN-6-P) deaminase (NagB)). In the UDP-GlcNAc biosynthetic pathway of C. glutamicum, GlcN-6-P is converted to fructose-6-phosphate (Fru-6-P) by NagB [45]. We also observed an increasing in NagB level in the presence of THL, suggestion that the UDP-GlcNAc biosynthetic pathway in cell wall biosynthesis is also influenced. These pathways are core links in the synthesis of bacterial cell wall polysaccharides (such as arabinogalactan and peptidoglycan precursor modification groups). Nucleotide sugars (e.g.., UDP-glucose, UDP-N-acetylglucosamine) are direct synthetic precursors of cell wall polysaccharides. The down-regulation of their metabolic proteins leads to insufficient supply of polysaccharide synthesis precursors, directly impairing the structural integrity of the cell wall and increasing the porosity of the cell wall.

The mechanism of action of vancomycin is to bind to the d-alanyl-d-alanine (d-Ala-d-Ala) structure of peptidoglycan precursors, preventing peptidoglycan cross-linking [43]. The impairment of cell wall integrity significantly improves the efficiency of vancomycin penetrating the cell wall, enabling it to more easily reach the peptidoglycan synthesis site and enhance the inhibitory effect. Meanwhile, after THL treatment, the abnormal expression of glycosyltransferase family proteins (e.g., dolichyl-phosphate-mannose-protein mannosyltransferase, encoded by HMPREF1861_01610, with a 2.26-fold upregulation in expression) is a compensatory response of bacteria (attempting to repair polysaccharide synthesis defects) [46]. However, the glycosyl transfer reaction catalyzed by this enzyme depends on nucleotide sugar substrates. The insufficient substrates prevent it from effectively maintaining the cell wall polysaccharide structure, but instead exacerbate cell wall structural disorders and further amplify the effect of vancomycin.

Energy metabolism and amino acid synthesis-related proteins-weakening bacterial compensation ability and reducing tolerance to vancomycin

KEGG functional enrichment analysis further showed the down-regulation of oxidative phosphorylation pathway proteins. Oxidative phosphorylation is the main pathway for bacteria to produce ATP. The down-regulation of proteins in this pathway leads to insufficient energy supply for bacteria. Cell wall synthesis (such as peptidoglycan cross-linking and lipid bilayer repair) is a high-energy-consuming process. Energy deficiency prevents bacteria from timely repairing cell wall damage caused by vancomycin, accelerating bacterial lysis. Upregulation of sulfur metabolism and purine metabolism proteins is a stress compensatory response of bacteria to THL pressure (e.g., purine metabolism provides precursors for DNA synthesis, and sulfur metabolism is involved in the activity modification of key enzymes). However, this compensation is insufficient to offset the core defects of glucose metabolism disorders and energy deficiency. Although bacteria attempt to maintain basic life activities, they cannot repair the structural damage of the cell wall, resulting in a significant decrease in the tolerance threshold to vancomycin;

The L-serine is a key precursor for the synthesis of cell wall phospholipids (such as phosphatidylglycerol and phosphatidylinositol). The key enzymes (SerA, SerC, SerB) in L-serine synthesis pathway plays the functions [47]. The down-regulation of its synthetase leads to reduced phospholipid synthesis, disrupting the cell wall lipid-protein bilayer structure, further increasing cell membrane permeability, and creating conditions for vancomycin to enter the interior of bacteria. ThiC (phosphomethylpyrimidine synthase) is a key enzyme for the synthesis of thiamine pyrophosphate (TPP), which is a core coenzyme for glucose metabolism and energy metabolism [48, 49]. The down-regulation of ThiC probably lead to TPP deficiency, exacerbating glucose metabolism disorders and insufficient energy supply, forming a vicious cycle of “metabolic defects - energy deficiency - failure of cell wall repair” and enhancing the bacteriostatic effect of vancomycin.

Functional proteins such as esterases and proteases-destroying the cell wall lipid-protein structure and cooperating with vancomycin to target the cell wall

As a lipase and esterase inhibitor, THL treatment leads to the differential expression of serine proteases, hydrolases, esterases, and other proteins in bacteria (e.g.., down-regulated expression of the serine protease-encoding gene DI525_03295) [29]. These proteins probably are involved in the degradation and modification of cell wall lipids (such as exogenous lipids dependent on C. kroppenstedtii and the processing and maturation of cell wall proteins. C. kroppenstedtii is a lipophilic bacterium that relies on lipases to hydrolyze triglycerides in the environment to obtain fatty acids for cell membrane lipid synthesis [36]. The inhibition of lipases/esterases by THL (combined with the down-regulated expression of related enzymes in proteomics) directly reduces the available fatty acid precursors for bacteria, resulting in incomplete cell membrane lipid bilayer structure. The core target of vancomycin is cell wall peptidoglycan [43]. By regulating the expression of lipases, proteases, and other proteins to destroy the cell wall lipid-protein structure, THL complements vancomycin’s “peptidoglycan targeting”, disrupting the cell wall integrity from multiple dimensions (lipid layer-protein layer-peptidoglycan layer) and ultimately achieving a synergistic bacteriostatic effect.

ABC transporters and stress-related proteins-reflecting the bacterial stress state and confirming the effectiveness of the synergistic effect

After THL treatment, the expression of methionine ABC transporter ATP-binding protein (homologous protein of DI525_00290) and cation-proton antiporter (DI525_07205) in bacteria was up-regulated. ABC transporters are involved in the transport of amino acids, antibiotics, and other substances in bacteria. The MetP protein in Candida hansenii, likely a homolog of the methionine ABC transporter ATP-binding protein encoded by DI525_00290 in C. kroppenstedtii, consists of a large subunit with 12 predicted transmembrane segments and an additional subunit with one predicted transmembrane segment [50]. The bacterial ABC transporters mediate various cellular processes such as multidrug resistance (MDR), acquisition of essential nutrients, adhesion, formation of spores, conjugation, biofilm formation and toxin secretion [51]. The upregulation of these protein in THL-treated C. kroppenstedtii is a manifestation of bacteria attempting to cope with drug pressure by enhancing methionine uptake (to maintain protein synthesis) and vancomycin efflux.

Cation-proton antiporters are involved in bacterial pH tolerance and osmotic pressure regulation. The Na⁺/H⁺ antiporter MnhD subunit, encoded by the gene mnhD in C. kroppenstedtii DSM 44,385, has a homologous protein in the staphylococcal mnh operon (mnhABCDEFG) was previously reported to involved in pH tolerance and halotolerance [52]. Our finding also showed that a cation-proton antiporter, encoded by DI525_07205, is up-regulated in THL-treated bacteria, suggesting that bacteria are in a stress state (cell wall damage leads to intracellular osmotic pressure imbalance).

In addition, previous research has indicated that the abundance of glycine oxidase ThiO in Bradyrhizobium correlates with glyphosate degraded in soil, serving as an indicator to monitor glyphosate degradation [53]. In our study, the glycine oxidase ThiO, encoded by the thio gene, is probably involved in the glyphosate metabolism induced by THL-treatment. These observations indicated that THL also participate the bacterial metabolism in THL induced stressful condition. The up-regulation of these proteins indirectly confirms that the synergistic effect of THL and vancomycin has caused significant pressure on bacteria. Bacteria need to activate stress compensation mechanisms to maintain basic survival, further indicating that this drug combination can effectively inhibit bacterial growth, which is consistent with the result of “synergistic effect” detected by the checkerboard method.

In the present study, we just assess the combination of THL and vancomycin on bacteria C. kroppenstedtii in vitro. Some limitations are existed: e.g. reliance on In Vitro models with limited clinical translatability; lack of evaluation of drug safety and long-term efficacy; lack the animal model to verify the experimental work. Understanding how bacteria respond to antibiotics or drugs can provide insights into their tolerance or resistance mechanisms. As the innovative drug combination should be considered in C. kroppenstedtii infection therapy development, future studies will assess the efficacy, the mechanistic and the pharmacodynamics of the formulated Vancomycin/THL combination in an animal model to verify the experimental work.

Conclusion

In the present study, we demonstrated that C. kroppenstedtii is sensitive to vancomycin, more importantly, the combination of THL and vancomycin exhibited a potent synergistic effect in inhibiting bacteria growth. Our proteomic analysis revealed that THL disrups critical physiological processes in the bacterium, including the sugar metabolism, oxidative phosphorylation, monobactam biosynthesis. These disruption compromise the structural intergrity of the cell wall, thereby enhancing the efficacy of vancomycin. These findings hold significant clinical promise for the treatment of GM. The synergistic regimen of THL with vancomycin presents a novel, targeted therapeutic strategy that could directly address the unique lipophilic nature and pathogenesis of C. kroppenstedtii infections. By potentially increasing treatment efficacy, this combination could lead to improved clinical outcomes. Therefore, our work not only elucidates a compelling antimicrobial synergy but also paves the way for developing a more effective and targeted antibiotic therapy for GM, offering a tangible prospect for improving patient care.

Materials and methods

Materials

The C. kroppenstedtii strain was obtained from the clinical laboratory at shanxi provincial people’s hospital. Brain Heart Infusion Broth (BHI) was sourced from the Hopebio (Qingdao, China). Vancomycin and thiazolyl blue tetrazolium bromide (MTT) were purchased from the Solarbio (Beijing, China). Tetrahydrolipstatin (THL) was acquired from the MedChemExpree (Shanghai, China). The Raw264.7 cell line was gifted by the Prof. Hui Liao and Fang Zhang at Shanxi provincial people’s hospital. Dulbeccos Modified Eagles Medium (DMEM, high glucose) was purchased from the BOSTER (Wuhan, China) and fetal bovine serum (FBS) was purchased from the Thermo Fisher Scientific (USA). Molecular regents were purchased from Thermo Fisher Scientific (USA).

Antibacterial drug susceptibility assay

The C. kroppenstedtii preculture was grown in 25 cm² flasks at 37 °C with shaking in Brain Heart Infusion Broth medium supplemented with 1% (v/v) Tween 80. The precultures were diluted to obtain a turbidity of 0.5 McFarland, which corresponds to approximately 1.5 × 10⁸ colony forming unit (CFU) per mL as measured with a nephelometer. Subsequently, the diluted microbial suspensions were further diluted 100 times to achieve 1.5 × 10⁶ CFU/mL. A volume of 100 µL of bacteria were inoculated into 96-well plates containing 100 µL of diluted drugs (THL or vancomycin). The plates were incubated at 37 °C for 48 h. After a 48- hour incubation, 30 µL of an 80% (m/v) MTT solution were added to each well and the plate were incubated for an additional 30 min. The minimum inhibitory concentration (MIC) was obtained by visual observation and corresponded to the lowest drug concentration that inhibited 100% of microbial growth compared to control wells without drugs. The THL stock solution was prepared at a concentration of 50 mg/mL in DMSO, while vancomycin was prepared at 10 mg/mL in water.

The combined effect of vancomycin and THL were assessed using macrodilution method. The fractional inhibitory concentration index (FICI) was calculated according to the checkerboard method (Eq. 1) [54]. Here, MIC_A or MIC_B represent the MIC of the THL or vancomycin alone, respectively. MIC_AB is the MIC of the THL in combination with a fixed vancomycin concentration, MIC_BA is the MIC of the vancomycin in combination with a fixed THL concentration were obtained. The fixed concentrations were 4 and 20-fold lower than the MIC_A or MIC_B, respectively. In agreement with the checkerboard method, synergy is reached when the FICI is < 0.5 [31].

1

Bacteria growth curves

The C. kroppenstedtii were grown at 37 °C with orbital shaking at 100 rpm for 48 h, starting with an initial optimal density at 600 nm (OD₆₀₀) with 0.01. Bacterial growth was assessed in the presence of THL, vancomycin, a combination of THL and vancomycin, in the absence of drugs, all within BHI medium. Absorbance was record at 600 nm using a SynergyHT plate reader (BioTek) and analysed with Excel (Microsoft). All conditions were tested in triplicate and all experiments were repeated at least three times.

Macrophage infection with C. kroppenstedtii

For macrophage infection and CFU enumeration, murine macrophage cell line RAW 264.7 was used and performed according to a previously reported as the Mycobacteria infection model [55, 56]. The Raw264.7 cell were cultured in DMEM supplemented with 10% (v/v) FBS and 100 µg/mL penicillin/streptomycin (Solarbio) in the incubator maintaining 37 °C and 5% CO₂ to allow for overnight differentiation into a monolayer in 96-well plates. The C. kroppenstedtii preculture was grown in BHI medium containing 1% (v/v) Tween 80. Subsequently, cells were washed and infection was performed at appropriate multiplicity of infections (MOI) with C. kroppenstedtii. Briefly, corynebacterial cells with pre-determined CFU were suspended in DMEM medium and added to the Raw264.7 monolayer for 3 h. After infection, extracellular bacteria were removed by washing the cells and fresh DMEM medium was added, either containing the desired drugs or without drugs as a control. The cells were further incubated at 37 °C with 5% CO₂ for 24 h.

The following day, the supernatant was discarded and 100 µL of lysis buffer (BHI containing 0.1% SDS) was added. Bacterial CFUs were enumerated by adding serial dilutions of the lysed cell suspension to BHI agar plates supplemented with 1% (v/v) Tween 80. The plates were incubated at 37 °C to allow for bacterial growth and the colony counts were performed after 48 h. The experiment was conducted in triplicate across three independent assays.

Proteomic analysis

To further investigate the potential targets of THL against C. kroppenstedtii, we conducted a proteomic analysis in this study. Briefly, the bacteria were grown in BHI medium for 36 h in the presence of THL, with bacterial growth in the absence of drug serving as control group. The bacterial pellets were collected by centrifugation with 3000 g for 15 min and washed twice with sterilized PBS. The pellets were stored at -80℃ for further analysis.

The bacteria resuspension and protein extraction were performed using SDT (4% m/v, SDS, 100 mM Tris-HCl pH7.6, 1 mM DTT) buffer. The concentration of extracted protein estimation was estimated using a bicinchoninic acid (BCA) assay with a BCA Protein Assay Kit (Bio-Rad, USA). The extracted proteins were digested with trypsin according to filter-aided sample preparation (FASP) procedure. The resulting peptides were desalted using C18 Cartridges (Empore SPE cartridges, Sigma) and concentrated via vacuum centrifugation. Finally, the dried samples were reconstituted in 40 µL of 0.1% (v/v) formic acid.

LC-MS/MS analysis was performed on a time-of-flight (TOF) Pro mass spectrometer (Bruker) coupled with Nanoelute (Bruker Daltonics) over durations of 60/120/240 min. The peptides were loaded onto a reverse phase trap column (Thermo Scientific) connected to an analytical C18-reversed phase column (Thermo Scientific Easy Column, 10 cm long, 75 μm inner diameter, 3 μm resin). Mobile phase A consisted of 0.1% (v/v) formic acid, while mobile phase B contained 84% (v/v) acetonitrile containing 0.1% (v/v) formic acid. The mobile phase was delivered using a linear gradient program at a flow rate of 300 nL/min.

The mass spectrometry (MS) was performed in a positive ion mode, collecting ion mobility MS spectra over a mass range of m/z 100–1700 and 1/k0 of 0.6 to 1.6. The analysis included 10 cycles of parallel accumulation–serial fragmentation (PASEF) MS/MS with a target intensity of 1.5 k and a threshold of 2500. Active exclusion was enabled with a release time of 0.4 min. The MS raw data were combined and searched using the MaxQuant 1.5.3.17 software for identification and quantification analysis on a UniProt C. kroppenstedtii (strain DSM 44385/JCM 11950/CIP 105744/CCUG 35717) database (https://www.uniprot.org/). The p-values were determined based on two independent-sample t -tests. Differentially expressed proteins were defined as having a p-value < 0.05 and a fold change (FC) > 1.2 or < 0.83 between two groups. These proteins were subjected to further analysis.

Abbreviations

GM

Granulomatous mastitis

GLM

Granulomatous lobular mastitis

C. kroppenstedtii

Corynebacterium kroppenstedtii

THL

Tetrahydrolipstatin

PDIM

Phthiocerol dimycocerosate

CLSI

Clinical and laboratory standards institute

FICI

Fractional inhibitory concentration index

OD

Optimal density

MOI

Multiplicity of infection

FC

Fold change

GO

Gene ontology

KEGG

Kyoto encyclopedia of genes and genomes

Authors’ contributions

DY introduced the primary idea. DY, YFZ, HFX, YJZ were involved designing the study. Resources were provided by DY and YJZ. DY, YFZ and ISS had equal roles in designing and implementation of the actual steps of methods. Statistical analysis and interpretations were done by DY and SPZ. The primary version of the manuscript was prepared by DY and ISS, it was revised and completed by YJZ, HPL, CHH and RSL. The submitted version of this manuscript was also read and accepted by DY, YJZ and RSL.

Funding

This work was supported by the grants from Health Commission of Shanxi Province (NO. 2023110) and the Natural Science Foundation of Shanxi Province (NO. 20210302123356).

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.