Skip to main content
NCBI home page
Journal of Advanced Research logoLink to Journal of Advanced Research
. 2023 Jan 24;54:223–237. doi: 10.1016/j.jare.2023.01.018

Single-cell analysis reveals lysyl oxidase (Lox)+ fibroblast subset involved in cardiac fibrosis of diabetic mice

Heyangzi Li a,1, Xiaoqing Zhu b,1, Xi Cao a, Yicheng Lu a, Jianwei Zhou b,⁎, Xiaoming Zhang c,⁎
PMCID: PMC10703720  PMID: 36706988

Graphical abstract

graphic file with name ga1.jpg

Open in a new tab

Keywords: Single-cell RNA sequencing, diabetic cardiomyopathy, fibroblasts, lysyl oxidase

Research highlights

  • •

    Single-cell RNA sequence analysis revealed the heterogeneity of fibroblast of hearts in diabetic mouse for the first time.

  • •

    The number of cluster 4 fibroblasts were higher and we reveal the up-regulation of Lox in diabetic mouse group.

  • •

    We validate that the up-regulation of Lox is associated with the decreased cardiac function, increased interstitial fibrosis, higher expression of HYP and the amounts of collagen in the DM group.

  • •

    Upregulation of Lox might be involved in the development of diabetic myocardial fibrosis and cardiac dysfunction via activation of TGF-β1-Smad2/3 signaling pathway.

  • •

    Inhibition of Lox could improve cardiac function and fibrosis, which may indicate a potential target for the treatment of diabetic cardiomyopathy.

Abstract

Introduction

Myocardial fibrosis and cardiac dysfunction are the main characteristics of diabetic heart disease. However, the molecular mechanisms underlying diabetic myocardial fibrosis remain unclear.

Objectives

This study aimed to investigate the heterogeneity of cardiac fibroblasts in diabetic mice and its possible mechanism in the development of diabetic myocardial fibrosis.

Methods

We established a diabetic mouse model by injecting mice with streptozotocin. The overall cell profiles in diabetic hearts were analyzed using single-cell RNA transcriptomic techniques. Cardiac function was evaluated by echocardiography. Cardiac fibrosis was assessed by Masson’s trichrome and Sirius red staining. Protein expression was analyzed using Western blotting and immunofluorescence staining.

Results

A total of 11,585 cells were captured in control (Ctrl) and diabetic (DM) hearts. Twelve cell types were identified in this study. The number of fibroblasts was significantly higher in the DM hearts than in the Ctrl group. The fibroblasts were further re-clustered into nine subsets. Interestingly, cluster 4 fibroblasts were significantly increased in diabetic hearts compared with other fibroblast clusters. Lysyl oxidase (Lox) was highly expressed in DM fibroblasts (especially in cluster 4). Beta-aminopropionitrile, a Lox inhibitor, inhibited collagen expression and alleviated cardiac dysfunction in the diabetic group. Lysyl oxidase inhibition also reduced high glucose-induced collagen protein upregulation in primary fibroblasts. Moreover, a TGF-β receptor inhibitor not only prevented an increase in Lox and Col I but also inhibited the phosphorylation of Smad2/3 in fibroblasts.

Conclusions

This study revealed the heterogeneity of cardiac fibroblasts in diabetic mice for the first time. Fibroblasts with high expression of Lox (cluster 4 fibroblasts) were identified to play a crucial role in fibrosis in diabetic heart disease. The findings of this study may provide a possible therapeutic target for interstitial fibrosis.

Introduction

In 2021, the global diabetic incidence has reached by over 500 million and is estimated to exceed 700 million in 2045 [1]. Diabetes mellitus is accompanied by various complications, including diabetic nephropathy, diabetic retinopathy, and diabetic cardiomyopathy (DCM), which are the primary causes of morbidity and mortality [2]. Because of left ventricular dysfunction and interstitial fibrosis, DCM is independent of hypertension, valve disease, or coronary artery disease, which can be found in both human patients and animal models of diabetes [3], [4], [5], [6], [7]. Several studies have shown that hyperglycemia can induce cardiac fibrosis by activating multiple fibro genic pathways [5], [8], [9], [10], [11], [12]. However, the molecular mechanisms underlying DCM remain unclear. Approximately 20 % of all noncardiomyocytes are fibroblasts, which produce the most matrix in the heart. Their role in the structural and mechanical maintenance of the heart is crucial [13]. Although fibroblasts serve as scaffolds for cardiomyocytes under normal conditions, they are also the final effector cells of cardiac fibrosis [14], [15], [16], [17]. However, the overall profile of cardiac fibroblasts in diabetic myocardial fibrosis remains unknown.

Since fibroblasts participate in extracellular matrix (ECM) deposition, maintenance, and cardiac remodeling, it is important to determine the role of fibroblast heterogeneity in DCM. Because traditional bulk RNA sequences cannot detect changes in each cell type simultaneously, we performed single-cell RNA sequencing to systematically determine the characteristics of fibroblasts in diabetic heart tissue. Most importantly, our results revealed that one fibroblast subset with high expression of Lox may be crucial for cardiac fibrosis. Moreover, the emergence of Lox+ fibroblast subset is associated with fibrogenic signal transduction caused by an overactive transforming growth factor beta (TGF-β) response in the diabetic hearts. Our study proposes a novel understanding of the alternation in the phenotype of fibroblasts that contribute to cardiac fibrosis and indicates a potential goal for preventing and alleviating DCM.

Materials and Methods

Establishment of streptozotocin (STZ)-Induced diabetic model

Male C57BL/6 mice (22–25 g, 8 weeks old) were obtained from the Experimental Animal Center of Zhejiang University. The animal studies were approved according to the policies instituted by the Committee of Animal Experiment Center of Zhejiang University (Approval Number 20210192). For single-cell RNA sequencing, there were two groups of mice: control (Ctrl) and diabetic (DM) groups. For other in vivo studies, the mice were randomly divided into four groups: Ctrl, DM, Ctrl + Beta-aminopropionitrile (BAPN) (Ctrl + BAPN), and diabetic + BAPN (DM + BAPN) groups. The diabetic mice were intraperitoneally injected with 150 mg/kg of STZ, as previously described [18]. The Ctrl mice were received with similar volume of sodium citrate. The Lox inhibitor, BAPN (#A3134, Sigma-Aldrich, USA), was intraperitoneally injected (100 mg/kg/day, every-two days) into each mouse in the DM + BAPN and Ctrl + BAPN groups according to previous studies [19], [20], [21]. A glucose level over 16.7 mmol/L suggested that diabetic model was successfully completed. Weekly measurements of body weight and blood glucose levels were performed. Additionally, percentage of HbA1c levels of the different groups were obtained using the A1cNow self-check kit (Sinocare, China).

Single-Cell RNA dissociation

Mice from the Ctrl and DM groups (n = 4) were sacrificed eight weeks after diabetic conduction. According to the guidelines, all the mice were euthanized using isoflurane. Then the hearts were carefully cut off and saved in tissue storage solution (Miltenyi Biotec). The samples were then washed with phosphate-buffered saline (PBS), and cut into small pieces and digested at 37 °C with 150 U/mL collagenase II (Worthington) and 0.2 % collagenase IV (Worthington), 1.2 U/mL Dispase Ⅱ (Worthington), and 500 U/mL DNase I (Worthington) for 45 min. The digested cells were then filtered through a cell strainer (70 µm). The cells were resuspended in PBS, and we filtered them through a cell strainer (35 µm). The viability of isolated single cells was>90 % and cells were prepared for library preparation.

Single-Cell RNA sequencing and Statistical analysis

Following the manufacturer's instructions, scRNA-Seq libraries were acquired using the Chromium Controller Instrument and Chromium Single Cell 3′ V3 Reagent Kits (10X Genomics, Pleasanton, CA). First, 8,000 cells were loaded into each channel at a concentration of 1,000 cells/L to obtain single-cell Gel Bead-IN-Emulsions (GEMs), and then lead to the expected mRNA barcoding of about 5,000 single cells for each sample. Additionally, the GEMs were broken, and barcoded-cDNA was performed like purifying and amplifying after the RT step. The amplified barcoded cDNA was then fragmented, A-tailed, ligated with adaptors, and amplified by polymerase chain reaction (PCR). Finally, quantization of the libraries was performed by the Qubit High Sensitivity DNA assay (Thermo Fisher Scientific, USA) and the size distribution of the libraries were calculated by a High Sensitivity DNA chip on a Bioanalyzer 2200. An Illumina Nova-seq 6000 was used to sequence all libraries with a paired-end 150-base pair reading strategy (Illumina, San Diego, CA). The scRNA-Seq analysis was performed on the NovelBrain Cloud Analysis Platform. Briefly, low-quality reads were removed and we obtained clean data. Then the Cell Ranger supplied the feature-barcode matrices. Down sample analysis was sequenced subsequently and then we obtained the aggregated matrix. Cell normalization and regression were analyzed using the Seurat package (v3.2.4). The t-distributed stochastic neighbor embedding was established. Gene Ontology analysis was performed using Metascape [22], which was visualized using a bioinformatics platform (bar plot and enrichment dot bubble, http://www.bioinformatics.com.cn/).

Gene set enrichment analysis (GSEA) analysis

To further identify the characteristics of differentially expressed genes of two groups, GSEA software analyzed the difference of two groups based on the two gene sets [23].

Cell communication analysis

To analyze potential ligand-receptor interactions between cells, we performed cell communication analysis using CellPhoneDB [24].

Pseudo-time analysis

Single-cell trajectories analysis was analyzed by Monocle2. After selecting marker genes of raw expression counts of the cell passed filtering and the Seurat clustering result, Monocle analysis was calculated.

Echocardiogram

M−mode echocardiography was performed using a Vevo 2100 system (FUJIFILM Visual Sonics, Inc., Canada) at 8 and 14 weeks. As described in a previous study [25], mice were anesthetized with isoflurane, and LV function was measured using two parameters: ejection fraction (EF) and fractional shortening (FS).

Real time quantitative PCR (qRT-PCR) analysis

Samples were resuspended in RNAiso Plus ( #9019, Takara) according to our previous study [26]. First, the RNA was concentrated in chloroform. After centrifugation at 12,500 × g for 15 min, supernatants containing RNA were extracted with isopropanol and cleaned with ethanol. Further, cDNA was obtained according to the protocol (#RR047A, Takara) and qRT-PCR was conducted using the TB Green Premix Ex Taq TM Ⅱ kit (#RR820,Takara) on a LightCycler480 (#480II, Roche). Gene normalization was analyzed by comparison with the expression of its control gene (β-actin), and the fold changes were analyzed using the 2 − ΔΔCt method. In addition,the sequences of the primers (5′ to 3′) were as follows:

Lox forward primer: GTGCCCGACCCCTACTACAT.

Lox reverse primer: TGACATCCGCCCTATATGCT.

β-actin forward primer: GGCTGTATTCCCCTCCATCG.

β-actin reverse primer: CCAGTTGGTAACAATGCCATGT.

Bulk RNA sequencing

Total RNA of Ctrl group and DM group was extracted as previously described, and bulk RNA sequencing was performed. Finally, the mapped reads of the samples were assembled using the StringTie software. After generation of the final transcriptome, StringTie and Ballgown were used to display the expression level of mRNA, which is shown as FPKM. Differentially expressed mRNA (|log2(Fold change)| > 1and p-value < 0.05) were selected using the R package edgeR. OmicStudio tools were used to perform bioinformatics analysis at https://www.omicstudio.cn/tool.

Measurement of plasma concentrations of creatine kinase (CK) and lactate dehydrogenase (LDH)

Serum samples were aliquoted into tubes and frozen at −80℃. Concentrations of CK and LDH were measured using Cobas c311 (Roche, USA).

Transmission electron microscopy

Heart tissues were fixed overnight in glutaraldehyde (2.5 %) overnight. Fixed myocardium from different groups at 14 weeks were washed with PBS, osmicated in 1 % OsO4, and dehydrated in 70–90–100 % ethanol. Subsequently, the samples were processed separately using acetone, epoxy resin, and uranyl acetate. Images were taken using a Tecnai G2 spirit120KV.

Culture of cardiac primary fibroblast and treatment

Primary fibroblasts from to 1–3 days post-born C57BL/6 mice were cultured based on a previous study [27]. Briefly, the ventricles were rinsed in cold PBS, cut into 1 mm3 in size, and digested in a solution containing 1 mg/mL collagenase II (#LS004174, Worthington, USA), 1 mg/mL papain (#LS003119, Worthington, USA), and 20 U/µL DNase I (#88700, Thermo Fisher Scientific, USA) for 5 min and the digeation process was repeated 3-5 times. The supernatant was then collected, centrifuged at 400 × g and resuspended in low-glucose Dulbecco’s modified Eagle’s medium (supplemented with 10 % fetal bovine serum). After 1 h of culture, the supernatants containing cardiomyocytes were removed and attached cells were fibroblasts. For different treatments, fibroblasts were cultured in normal glucose medium (5.5 mmol/L), high glucose (HG) medium (33 mmol/L), or D-mannitol (33 mmol/L, an osmotic control) for seven days. Beta-aminopropionitrile (500 µmol/L) was used as an irreversible inhibitor of Lox. For TGF-β (#100–21, Peprotech, USA) stimulation, fibroblasts were co-cultured with 10 ng/mL TGF-β1 and HG for 48 h, while SB431542 (#s1067, Selleck, China) was used to inhibit the TGF-β receptors.

Immunofluorescence, Masson, and Sirius red staining

Hearts from different groups were harvested, perfused with 4 % paraformaldehyde, and fixed overnight. Subsequently, they were subjected to dehydration in 10–20–30 % sucrose individually and then cut into 12 µm thickness. For immunofluorescence, the slices were washed, permeabilized, and blocked. Sections were incubated with primary antibodies (Lox, #NB110-59729, Novus; Vimentin, #NB300-223, Novus; Phalloidin, #PHDG1-A, Cytoskeletin; Col I, #NB600-450, Novus) after blocking. After incubating the primary antibodies overnight, secondary antibodies was incubated after washing for three times. The sections were then washed and stained with 4′,6-diamidino-2-phenylindole (DAPI). Images were pictured using a confocal microscope (FV3000, Olympus). Masson’s trichrome staining and Sirius Red staining were performed according to the manufacturer’s protocols. Images were observed using Olympus microscope.

Western blotting analysis

Samples were lysed in RIPA Lysis Buffer (#R0010, Solarbio) supplemented with phosphatase and protease (#P1261, Solarbio) and vortexed on ice three times every 10 min. After being centrifuged at 12,000 × g for 15 min, samples were then boiled with loading buffer to harvest the supernatants. After preparation, the samples were loaded on gels, and the samples were transferred to a cold transfer buffer. After blocking, the membranes were incubated with the following primary antibodies: β-actin (#A5316, Sigma), Lox (#ab31238, Abcam), p-Smad2 (#18338, Cell Signaling Technology), Smad2 (#5339, Cell Signaling Technology), p-Smad3 (#ab52903, Abcam), Smad3 (#9523, Cell Signaling Technology), GAPDH (#60004, Proteintech), Col I (#ab260043, Abcam), Col III (#22734–1-AP, Proteintech), p-c-jun (#ab32385, Abcam), c-jun (#ab40766, Abcam), and TGF-β1 (#ab215715, Abcam). After incubating the primary antibodies overnight, the samples were washed three times by 1 % Tween 20-PBS (TBS-T), incubated with secondary antibodies for 60 min, and washed (three times for 10 min). The relative expression of proteins was detected using Odyssey Clx (Li-COR Biosciences, USA).

Lysyl oxidase activity assay

The Amplite Fluorimetric Lysyl Oxidase Assay Kit (#15255, AAT Bioquest, USA) was used to measure the Lox activity. First, supernatants from the hearts or cells from different groups were prepared in a solid black 96-well plate and then incubated with Lox working solution at 37℃ for 30 min. The amount of hydrogen peroxide production was measured using a fluorescence plate reader (Ex/Em = 540/590 nm, Varioskan Flash, Thermo Scientific, USA).

Measurement of hydroxyproline (HYP)

A HYP assay kit (#A030-2–1, Nanjing Jiancheng, China) was used to quantify HYP in the heart tissues and serum. Briefly, the heart tissues were hydrolyzed (for 20 min at 95 °C) and centrifuged (3,500 g for 10 min) after adjusting pH value between pH 6.0–6.8. The supernatants were then subjected to the manufacturer's protocols and measured using a microplate reader at 550 nm. The amount of HYP was analyzed, and the results are presented in µg/mg.

Results

Establishment of a diabetic mouse model

A DM mouse model was injected with 150 mg/kg STZ in male C57BL/6 mice. Weight of the mice decreased in the DM group, and the level of blood glucose increased in the DM group from 72 h to 8 weeks after induction of diabete model (p < 0.05, Figs. 1A-B). Moreover, at 8 weeks after the induction, the heart weight/body weight ratio and plasma HbA1c% concentrations were elevated in the DM group compared to those in the Ctrl group (p < 0.05, Figs. 1C-D).

Fig. 1.

Fig. 1

Open in a new tab

Single-cell RNA sequencing showing cell profiles in DM mouse hearts. (A) Workflow for single-cell RNA transcriptome experiments. (B) t-SNE plot showing distinction of overall cell profiles between Ctrl (n = 4, shown in red) and DM (n = 4, shown in blue) groups. (C-D) t-SNE showing 20 cell clusters (C) and 12 cell types (D) in the integrated single-cell transcriptomes acquired from the Ctrl and DM groups. (E) Bar graph representing the frequency of cells acquired in each cluster. (F) Dot plot showing cell-specific markers. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Clustering analysis of Single-Cell RNA sequencing data from diabetic mouse hearts

Single-cell RNA sequencing of heart samples was performed eight weeks after the STZ injection (Fig. 1A). Viability of digested cells from the hearts used for single-cell sequencing analysis was > 90 % in the Ctrl and DM groups (Figs. 1E-F). A total of 11,585 cells from all the mouse hearts were captured. Fig. 1B shows that the overall cell profiles of Ctrl and DM mouse hearts were different. According to cell-specific markers, 20 distinct cell clusters were identified, which belong to 12 cell types, including endothelial cells, cardiomyocytes, endocardial cells, fibroblasts, macrophages, proliferating cells, murals, smooth muscle cells, T cells, B cells, glial cells, and neutrophils (Fig. 1C-D). Surprisingly, fibroblasts and endothelial cells were the two major cell types in the Ctrl and DM hearts (Fig. 1E). Meanwhile, the number of fibroblasts was remarkably higher, whereas the number of endothelial cells was remarkably reduced in the DM group. Moreover, the high expression of other cell-specific markers in each cell type (Fig. 1F) confirms the accuracy of the clustering method based on the cell markers shown in Figs. 1G.

Enrichment analysis of differently expressed genes in DM hearts

Gene set enrichment analysis illustrated that differentially expressed genes (DEGs) between the Ctrl and DM hearts were mainly enriched in pathways, such as collagen fibril organization, inflammatory response, and response to oxidative stress (Fig. 2A). Upregulated DEGs were enriched in GO terms, including inflammatory response, extracellular matrix organization, and cellular response to interleukin 1. Downregulated DEGs were enriched in GO terms, including immune system processes and cellular responses to IFN-β (Fig. 2B). To further determine the characteristic variation between the Ctrl and DM groups, ligand-receptor interactions among various cell types were analyzed (Fig. 2C). Compared with the Ctrl group, fibroblasts in the DM hearts had stronger connections to their neighboring cells (indicated by an arrowhead in Fig. 2C), suggesting that fibroblasts might be key regulators in the DM heart.

Fig. 2.

Fig. 2

Open in a new tab

Enrichment analysis of differently expressed genes in DM hearts. (A) Gene Set Enrichment Analysis of differently expressed genes between Ctrl and DM hearts from a single-cell RNA sequencing. (B) Top 10 GO enrichment pathways for up-regulated (red bars) and down-regulated (blue bars) genes. (C) Circos plot showing ligand-receptor interactions among various cell types. The thickness of ribbons is proportional to the number of interactions between two cells. (D) Fibroblasts were highlighted in orange color according to the expression of cell-specific marker (Col1a1, Col1a2, and Dcn). (E) Volcano plot showing up-regulated (red) and down-regulated (blue) differently expressed genes in DM hearts compared with Ctrl hearts. (F) Top 20 GO enrichment pathways of differently expressed fibroblasts genes. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Because fibroblasts were also the most remarkably increased cell type in DM hearts (Fig. 1E), we focused our study on fibroblasts (Fig. 2D). A total of 278 DEGs in fibroblasts between the Ctrl and DM groups were identified (p < 0.05, Fig. 2E), of which 162 (such as S100a4, Fbln5, Lox, and Lepr) were up-regulated and 116 (such as Ccnd1, Ifit3, and Stat1) were downregulated. These DEGs were enriched in GO terms, including collagen fibril organization and extracellular matrix organization (Fig. 2F).

Re-Clustering of fibroblasts and Up-Regulation of Lox in cluster 4 of fibroblasts in DM hearts

Fibroblasts were further re-clustered into nine subsets (Fig. 3A) based on the re-clustering strategies shown in Fig. 3B-C. Surprisingly, compared to the Ctrl group, the number of fibroblasts in clusters 2 and 4 was significantly higher in the DM group (Fig. 3D). GO enrichment analysis of clusters 2 and 4 was then performed. In cluster 2, upregulated DEGs were enriched in inflammatory responses (such as cellular response to interleukin 1 and 6) (Fig. 3E). Upregulated DEGs in cluster 4 were enriched in organization, positive regulation of fibroblast proliferation and cellular response to TGF-β stimulus (Fig. 3F). Since Lox mRNA is a marker gene of cluster 4 fibroblasts (Fig. 3B) and the Lox gene is mainly expressed in fibroblasts (Fig. 3G), we further explored the role of Lox. As shown in Fig. 3H-I, Lox was not only upregulated in DM hearts, but also highly expressed in fibroblasts (especially in cluster 4) compared with other cell types in the heart (Fig. 3I). Immunofluorescence staining confirmed the increased expression of Lox protein in the DM hearts (Fig. 3J). Moreover, qRT-PCR analysis verified higher Lox mRNA expression in the DM hearts than in the Ctrl hearts (p < 0.01, Fig. 3K).

Fig. 3.

Fig. 3

Open in a new tab

Re-clustering of fibroblasts and up-regulation of Lox in cluster 4 of fibroblasts in DM hearts. (A) Fibroblasts were re-clustered into nine subsets. (B) Specific markers used for fibroblast re-clustering. (C) Heatmaps showing distinct expression profiles of each fibroblast subset. (D) Bar graph representing the number of fibroblasts in each subset. (E-F) Top eight GO enrichment pathways for up-regulated (red bars) and down-regulated (blue bars) genes in cluster 2 (E) and cluster 4 (F). (G) t-SNE plot showing the difference in Lox mRNA expression between the Ctrl and DM groups. (H-I) Violin plot showing highly expressed Lox mRNA in fibroblasts (H), especially in cluster 4 (I). (J) Representative immunofluorescence image of Lox protein (red) in the Ctrl and DM hearts, stained with Col Ⅰ (green, fibroblast marker) and phalloidin (grey); scale bar = 10 µm. (K) Validation of Lox mRNA expression using qRT-PCR analysis (n = 6, **p < 0.001 vs Ctrl group). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

In addition, we also performed bulk RNA sequencing to reveal the expression of genes in the hearts at 8 weeks (Figs. 2A). Principal component analysis illustrated the distribution of the Ctrl and DM groups (Figs. 2B). Additionally, Lox expression was elevated in the DM group, which was consistent with our single-cell sequence data (Figs. 2C). GO enrichment pathway analysis revealed the DGEs involved in collagen fibril organization, cell-matrix adhesion, response to hypoxia, etc. (Figs. 2D).

Inhibition of Lox alleviated STZ-Induced myocardial dysfunction

To further determine the function of Lox in the DCM, mice were treated with BAPN to suppress Lox expression (Fig. 4A). Hearts from different groups were harvested at 8 and 14 weeks after the STZ injection. Fig. 4B and D showed lower body weights in the DM group than in the Ctrl group (p < 0.01). Moreover, the body weight in the Ctrl + BAPN group slightly decreased. Blood glucose levels in the DM group increased to > 17.6 mmol/L and it showed higher glucose levels in the DM group than in the Ctrl group (p < 0.01) (Fig. 4C and E).

Fig. 4.

Fig. 4

Open in a new tab

Inhibition of Lox alleviated STZ-induced myocardial dysfunction. (A) Workflow of BAPN (an inhibitor of Lox) treatment in DM mice. (B-E) Effect of BAPN on body weight and glucose level in DM mice (**p < 0.01, DM vs Ctrl group). (F) Representative echocardiography M−mode images of mice hearts at 8 and 14 weeks after STZ injection. (G-H) Statistical analysis of echocardiogram. n = 7–10, **p < 0.01, *p < 0.05. (I-J) Measurement of serum levels of CK and LDH, n = 4, **p < 0.01, *p < 0.05. (K) Transmission electron microscopic analysis of hearts tissue at 14 weeks after STZ injunction. The Ctrl group image shows a well-organized structure and complete mitochondria. The DM group image shows damaged and ruptured mitochondria (red arrowheads). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Echocardiography was performed to explore myocardial function. In the early term of diabetes (eight weeks), there was no significant difference in the EF% and FS% between the Ctrl and DM groups (Fig. 4F-H). However, EF% and FS% remarkably decreased at 14 weeks in the DM group (p < 0.01 vs Ctrl group, Fig. 4F-H), which could be prevented by treatment with BAPN. Similar to the echocardiogram results, serum CK and LDH levels were significantly elevated in the 14 weeks DM group, which could be partly abolished by BAPN treatment. Transmission electron microscopic analysis showed that the mitochondria in the Ctrl hearts were complete and well-organized, whereas the mitochondria in the DM hearts were damaged or ruptured (Fig. 4K). Additionally, the injection of BAPN alleviated the degree of mitochondrial damage in DM hearts. These data demonstrate that Lox inhibition alleviates myocardial dysfunction.

Inhibition of Lox decreased collagen fibril production in DM hearts

Increased expression of Lox was displayed in fibroblasts in the DM hearts compared to that in the Ctrl hearts via immunofluorescence staining (Fig. 5A). Lysyl oxidase activity in the hearts was also higher in the DM group (p < 0.01 vs Ctrl group), which could be inhibited after BAPN treatment (p < 0.01 vs DM group; Fig. 5B). Hydroxyproline, an amino acid, exists only in collagen, and the HYP level in the heart or serum represents the degree of fibrosis of a tissue. Cardiac or serum HYP levels in the DM group were elevated (p < 0.01, Fig. 5C-D). Additionally, injection of BAPN resulted in a reduction in HYP levels in the DM group (p < 0.05, Fig. 5C). Western blotting results illustrated the elevated content of Col I and III proteins in myocardial tissues in the DM group at 14 weeks (p < 0.05 vs Ctrl group, Fig. 5E-G).

Fig. 5.

Fig. 5

Open in a new tab

Inhibition of Lox decreased collagen fibril production in DM hearts. (A) Representative immunofluorescence staining image displaying expression of Lox (red) stained with Col Ⅰ (green, fibroblast marker) and phalloidin (grey); scale bar = 10 µm. (B) Measurement of Lox activity of heart homogenates (n = 8, **p < 0.01). (C-D) Quantification of hydroxyproline (HYP) in heart tissues (n = 5) and serum (n = 8) (*p < 0.05, **p < 0.01). (E) Protein levels of Col Ⅰ and III by Western blotting. (F-G) Quantification of Col Ⅰ and III expression (n = 3, *p < 0.05, **p < 0.01). (H-K) Representative Masson and Sirius red staining images. In the Masson staining (H), the collagen area is blue, and the collagen area is red in the Sirius red staining (J). Quantification of results from Masson (I) and Sirius red staining (K) (n = 9, **p < 0.01). (L-P) Scatterplots showing correlation of Lox expression with HYP content, content of Col Ⅰ, content of Col III, and fibrosis area from the Masson and Sirius red staining. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

The severity of myocardial fibrosis was evaluated using Masson and Sirius red staining. Masson staining results showed an increase in the area of collagen deposition in the DM hearts compared with that in the Ctrl hearts (p < 0.01, Fig. 5H-I). Meanwhile, hearts in the DM + BAPN group displayed a lower area of collagen (p < 0.01 vs DM group, Fig. 5H-I). Sirius red staining revealed an increased collagen-based area in the DM group compared with the Ctrl group (p < 0.01, Fig. 5J-K). Sirius red staining showed that BAPN treatment reduced the collagen-based area in diabetic mouse hearts (p < 0.01, Fig. 5J-K). Additionally, as shown in Fig. 5L-P, Lox expression significantly correlated with HYP content, Col I and III content, and fibrosis area.

Inhibition of Lox declined HG-Induced collagen protein upregulation in primary fibroblasts

To further investigate the relevant signaling Lox involved ,we cultured primary fibroblasts and stimulated fibroblasts with high glucose (HG) (Fig. 6A). Immunofluorescence staining and western blotting analysis displayed that HG resulted in an increase in Lox expression in fibroblasts (p < 0.01 vs NG group, Fig. 6B-D). Additionally, HG also enhanced Lox activity in fibroblasts (p < 0.01 vs NG group, Fig. 6E). Beta-aminopropionitrile treatment inhibited the HG-induced increase in Lox expression and activity (p < 0.01, Fig. 6B-E). Moreover, the increase in collagen I and III proteins induced by HG in fibroblasts could also be inhibited by BAPN treatment (Fig. 6F-H). As shown in Fig. 6I-J, Lox expression was correlated with the content of Col I or Col III in cultured primary fibroblasts (Fig. 6I-J).

Fig. 6.

Fig. 6

Open in a new tab

Inhibition of Lox declined HG-induced collagen protein expression in primary fibroblasts.(A) Workflow of primary fibroblast culture. (B) Immunofluorescence staining showing the expression of Lox protein (Red), stained with Vimentin (green) and DAPI (blue); scale bar = 10 µm. (C) Western blot image showing the expression of Lox in fibroblasts. (D) Statistical analysis of Western blot results for Lox protein expression (n = 5, **p < 0.01). (E) Measurement of Lox activity in fibroblasts (n = 8, **p < 0.01). (F) Western blot image showing the expression of Col Ⅰ and III in the fibroblasts. (G-H) Quantification analysis of the expression of Col Ⅰ and III in fibroblasts (n = 5, *p < 0.05, **p < 0.01). (I-J) Scatterplots showing the correlation of Lox expression with Col Ⅰ or Col III protein content. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

High Glucose-Induced Lox upregulation via activation of TGF-β1 signaling pathway in fibroblasts

Single-cell RNA analysis showed that the TGF-β-related GO pathway was enriched in cluster 4 of fibroblasts (cellular response to TGF-β stimulus, Fig. 3F). Next, we explored the relationship between TGF-β and HG-induced Lox upregulation in fibroblasts. Transforming growth factor-β1 expression was higher in the DM hearts than in the Ctrl hearts (Figs. 2E-F). The expression of TGF-β1 increased in fibroblasts under HG conditions (p < 0.01), but BAPN did not inhibit the HG-induced upregulation of TGF-β1 (Figs. 2G-H). Recent studies have shown that TGF-β1, one of the most important regulators of fibrosis, plays a crucial role in regulating Lox expression [28], [29], [30], [31]. Our results illustrated that the content of Lox and Col I increased in the HG + TGF-β1 group (p < 0.05 vs HG group), which could be inhibited by the TGF-β receptor inhibitor SB431542 (p < 0.05, Fig. 7A-C). Additionally, TGF-β1 increased the ratios of p-Smad3/Smad3, p-Smad2/Smad2, and p-c-jun/c-jun in fibroblasts under HG conditions, which was prevented by SB431542 (p < 0.05, Fig. 7D-H). The above conclusions were verified by immunofluorescence staining (Fig. 7I-L).

Fig. 7.

Fig. 7

Open in a new tab

HG induced Lox upregulation via TGF-β1 signaling pathway activation in fibroblasts. (A) Western blotting for Lox and Col Ⅰ protein. (B-C) Quantification analysis of Lox and Col Ⅰ protein (n = 4, *p < 0.05, **p < 0.01). (D) Western blotting for p-Smad3, Smad3, p-Smad2, and Smad2. (E-F) Statistical analysis of ratios of p-Smad3/smad3 and p-Smad2/smad2 (n = 4, **p < 0.01). (G) Western blotting for c-jun and p-c-jun. (H) Quantification analysis of p-c-jun/c-jun ratio (n = 4, *p < 0.05, **p < 0.01). (I-L) Representative immunofluorescence staining illustrated the expression of Lox, p-Smad3, p-Smad2, and p-c-jun (red) stained with phalloidin (green), vimentin (pink, I and J), Col Ⅰ (pink, K and L), and DAPI (blue); scale bar = 10 µm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

The cellular source and difference of the Lox+ fibroblasts compared to other fibroblast subsets

To further explore the function of Lox+ fibroblasts, we analyzed the differential expressed genes (DEGs) of fibroblasts (Lox+ fibroblasts versus other fibroblast subsets) and displayed them in the volcano map (Figs. 3A). As for the cellular source of Lox+ fibroblasts, in general, we believe that Lox+ fibroblasts originate from resting fibroblasts in cardiac tissue after they are activated during DCM. Methodologically, we analyzed the cellular trajectory of fibroblasts by pseudo-time analysis (Figs. 3B). Fibroblasts were segregated into 3 states which named state1, state2, and state 3. State 1 represents the start of the differentiation (Figs. 3C). Distribution of each subset of fibroblasts in each state were displayed in Figs. 3D. As can be seen, state 1 were the origin of the differentiation directions of fibroblasts, Lox+ fibroblasts (cluster 4) were mainly in states 2 and 3.

Moreover, we analyzed the GO enrichment analysis and KEGG enrichment analysis of DEGs (Lox+ fibroblasts versus other fibroblast subsets). GO enrichment analysis illustrated that DEGs participated in processes such as response to hypoxia, collagen fibril organization, angiogenesis, and so on (Figs. 3E). Moreover, KEGG enrichment analysis illustrated that up-regulated DEGs in Lox+ fibroblasts were enriched in HIF-1 signaling pathway, TGF-beta signaling pathway and so on. Down-regulated DEGs were enriched in focal adhesion, protein digestion and absorption, and tight junction (Figs. 3F).

Discussion

In this study, single-cell RNA sequencing analysis was conducted to determine the molecular mechanisms underlying cardiac dysfunction in STZ-induced diabetic mice. We captured 11,585 cells and characterized 12 cell types. Interestingly, the number of fibroblasts in the heart samples increased significantly in diabetic mice, and fibroblasts had stronger connections to their neighboring cells, suggesting that fibroblasts might be key regulators of cardiac dysfunction in diabetic mice. Previous studies have reported that fibrosis plays a crucial role in DCM [32], [33] and fibroblasts contribute to the increased synthesis of collagen [34]. Briefly, collagen I is the most abundant type (approximately 85 % of fibrillar collagen), and cardiac stiffness is mainly regulated by the content of collagen I [35], [36]. Collagen III constitutes the remaining 15 % of fibrillar collagen and is related to cardiac compliance [35], [36]. The results of our study illustrated the increased expression of collagen I and III in diabetic mice.

To further explore the role of fibroblasts in diabetic hearts, fibroblasts were re-clustered into nine subsets. Among these subsets, clusters 2 and 4 were the two most remarkably increased clusters among the diabetic mouse hearts. According to results of the GO enrichment analysis, cluster 4 might be more likely to be involved in ECM organization and fibroblast proliferation than cluster 2. Among the DEGs of cluster 4, Lox is involved in the regulation of collagen crosslinking, which affects remodeling of the ECM [37]. The other DEGs were Fbln5, Fbln1, and S100A6. Fbln5 is involved in the formation of elastic fibers in the ECM [38], Fbln1 stabilizes ECM integrity and reacts with other ECM proteins [39], and S100A6 is involved in apoptosis, proliferation, and interstitial fibrosis [40], [41]. Therefore, changes in cluster 4 fibroblasts suggest that cluster 4 may be the crucial fibroblasts that participate in the process of diabetic myocardial fibrosis.

Our study illustrated that the expression of Lox protein was not only upregulated in diabetic mouse hearts, but was also distinctively higher in fibroblasts (especially in cluster 4). Therefore, we hypothesized that Lox is one of key genes involved in the regulation of myocardial fibrosis related to diabetes. Many studies have shown that elevated Lox expression is related with multiple diseases, such as heart failure [29], myocardial infarction [28], chronic kidney disease [42], and systemic sclerosis [43], [44]. Meanwhile, the increase in Lox expression exerts an indispensable role in several diabetic complications related to fibrosis of nephropathy [45] and retinopathy [46]. However, the role of Lox upregulation in DCM remains more exploration. Our present results have shown that BAPN, a Lox inhibitor [20], could not only reduce the HG-induced upregulation of collagen expression in primary cultures but also inhibit the overexpression of collagen I and III in diabetic mouse hearts. BAPN alleviated myocardial fibrosis and prevented cardiac dysfunction. These findings imply that Lox may be a crucial gene in the promotion of diabetic myocardial fibrosis.

The role and cellular mechanisms of Lox in the regulation of fibrosis have been explored in many studies. It has been reported that Lox and its family members may related to the process of remodeling the extracellular matrix [37]. As the only member of the Lox family that takes collagen as a substrate, Lox is the most abundant form of the Lox family [47]. Lysyl oxidase is first synthesized as an inactive proenzyme that can be cleaved into an active form and a propeptide fragment (Lox-pp) under external stimuli [48], [49]. Studies have shown that activated Lox manipulates the crosslinking of elastin and collagen [49], [50]. Other studies have reported that the active form of Lox can increase collagen fibrils by resisting matrix metalloprotease degradation and promoting collagen maturation [19]. The mechanism of Lox in diabetic myocardial fibrosis is unclear and requires further investigation.

The exact mechanism underlying Lox upregulation in diabetic mice remains unknown. Previous studies have demonstrated the role of TGF-β1 in the regulation of Lox expression. For example, exogenous (TNF-α) stimulation of cultured fibroblasts can increase the expression of TGF-β1 and lead to increased Lox [34]. In vitro primary fibroblasts, such as TGF-β1, can mediate Lox increase through PI3K/Akt, Smad3, and MAPK [51]. Transforming growth factor-β1 is secreted by multiple cells, including epithelial cells, mesangial cells, cardiomyocytes, macrophages, endothelial cells, fibroblasts, and vascular smooth muscle cells. For example, HG conditions stimulate collagen synthesis by increasing TGF-β production in tubular epithelial cells [52], [53]. Transforming growth factor-β activation in mesangial cells is stimulated by Ang II and hyperglycemia [54] and is also secreted by cardiomyocytes [55] and Treg cells [56]. Consequently, other cell types may secrete transforming growth factor-β as well as fibroblasts.

However, whether TGF-β1 regulates Lox production by fibroblasts in diabetic myocardia remain unclear. Therefore, in order to better simulate the HG environment in vivo, we stimulated fibroblasts with HG and observed whether the increase in TGF-β stimulated the increase of Lox expression in primary cultured fibroblasts. As TGF-β1 is the most important subtype of TGF-β isoforms [57],we then verified the increased expression of TGF-β1 under HG conditions both in vivo and in vitro. It shows that increased TGF-β1 can induce the expression of Lox and Col I in fibroblasts under HG conditions. Moreover, a TGF-β receptor inhibitor could not only prevent the increase in Lox and Col I but also inhibit the phosphorylation of Smad2/3 and c-jun (main component of transcription factor AP-1 [28]). Activation of TGF-β1/Smad2/3 signaling has been shown to be involved in isoproterenol-induced myocardial fibrosis [58]. Sethi et al. have also demonstrated that TGF-β binds to its receptor and phosphorylates the Smad2/3 complex in human trabecular meshwork cells. The phosphorylated Smad2/3 complex then translocates into the nucleus and binds to the transcription factor, AP-1, which in turn induces Lox expression [59]. The results of our study also suggest that the upregulation of Lox in diabetic myocardial fibrosis might be due to the TGF-β1/Smad2/3 pathway activation.

Moreover, we further analyzed the DEGs between Lox+ fibroblasts and other fibroblasts by GO and KEGG enrichment analysis. Results illustrated that HIF-1 signaling pathway may exert an important role in Lox+ fibroblasts. Therefore, combined with above bioinformatic analyses and previous study which illustrated that Lox could be regulated by hypoxia, together with TGF-β [28] we speculated that hypoxia might also be a possible regulator of Lox+ fibroblasts and it will be further investigated in our lab.

However, this study has some limitations. The Lox inhibitor BAPN, used in the present study is a non-selective inhibitor of Lox and Lox knockout model (Lox (-/-)) is lethal [60]. Small molecule inhibitors and lentivirus or adenovirus inhibition targeting Lox should be further explored. In addition, the role of the TGF-β1-Smad2/3 signaling pathway in the upregulation of Lox in the development of diabetic myocardial fibrosis should be further explored in vivo. Moreover, as SB431542 is a TGF-β signaling pathway inhibitor and an effective ALK5 inhibitor, the expression level of ALK5 (also named as TGFβR1) remained further exploration. Finally, the role of cluster 2 fibroblasts, which might be responsible for inflammation-related processes in diabetic mouse hearts needs further exploration.

Conclusions

In our study, single-cell RNA sequencing analysis revealed fibroblast profiles of the hearts of diabetic mice for the first time, and Lox+ fibroblasts have been found to play a crucial role in fibrosis in diabetic heart disease. Overexpression of Lox may be involved in the development of diabetic myocardial fibrosis and cardiac dysfunction via the activation of TGF-β1-Smad2/3 signaling pathway. Inhibition of Lox could be a promising target for the treatment of DCM.

Compliance with Ethics Requirements

All Institutional and National Guidelines for the care and use of animals (fisheries) were followed.

CRediT authorship contribution statement

Heyangzi Li: Conceptualization, Methodology, Investigation, Formal analysis, Visualization, Writing – original draft. Xiaoqing Zhu: Investigation, Writing – review & editing, Validation. Xi Cao: Methodology, Formal analysis, Resources. Yicheng Lu: Validation. Jianwei Zhou: Supervision, Funding acquisition, Writing – review & editing. Xiaoming Zhang: Conceptualization, Project administration, Supervision, Funding acquisition, Writing – review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

Thanks for the technical support provided by the core facilities, Zhejiang University School of Medicine, especially the assistance with confocal microscope provided by Junli Xuan. Thanks for the English language editing provided by Yingying Chen and Linlin Wang. We also thank NovelBioinformatics Ltd., Co. for the support of bioinformatics analysis with their NovelBrain Cloud Analysis Platform. Graphical abstract was created by Biorender.com. This study was supported by the National Natural Science Foundation of China (81472149).

Footnotes

Peer review under responsibility of Cairo University.

Appendix A

Supplementary data to this article can be found online at https://doi.org/10.1016/j.jare.2023.01.018.

Contributor Information

Jianwei Zhou, Email: 2195045@zju.edu.cn.

Xiaoming Zhang, Email: zxm@zju.edu.cn.

Appendix A. Supplementary material

The following are the Supplementary data to this article:

Supplementary data 1
mmc1.docx (1.4MB, docx)

Data availability

The accession number for the raw data files of the single-cell sequencing reported in this paper are uploaded on NCBI GEO dataset (GSE 213337). The accession number of Bulk RNA sequencing number is: GSE 215979. All data are available.

References

  • 1.Sun H., Saeedi P., Karuranga S., Pinkepank M., Ogurtsova K., Duncan B.B., et al. IDF Diabetes Atlas: Global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045. Diabetes Res Clin Pract. 2022;183 doi: 10.1016/j.diabres.2021.109119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Seferović PM, Paulus WJ. Clinical diabetic cardiomyopathy: a two-faced disease with restrictive and dilated phenotypes. Eur Heart J. 2015;36(27):1718-27, 27a-27c.10.1093/eurheartj/ehv134. [DOI] [PubMed]
  • 3.Evangelista I., Nuti R., Picchioni T., Dotta F., Palazzuoli A. Molecular Dysfunction and Phenotypic Derangement in Diabetic Cardiomyopathy. Int J Mol Sci. 2019;20(13) doi: 10.3390/ijms20133264. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Palomer X., Pizarro-Delgado J., Vázquez-Carrera M. Emerging Actors in Diabetic Cardiomyopathy: Heartbreaker Biomarkers or Therapeutic Targets? Trends Pharmacol Sci. 2018;39(5):452–467. doi: 10.1016/j.tips.2018.02.010. [DOI] [PubMed] [Google Scholar]
  • 5.Frati G., Schirone L., Chimenti I., Yee D., Biondi-Zoccai G., Volpe M., et al. An overview of the inflammatory signalling mechanisms in the myocardium underlying the development of diabetic cardiomyopathy. Cardiovasc Res. 2017;113(4):378–388. doi: 10.1093/cvr/cvx011. [DOI] [PubMed] [Google Scholar]
  • 6.Wilson A.J., Gill E.K., Abudalo R.A., Edgar K.S., Watson C.J., Grieve D.J. Reactive oxygen species signalling in the diabetic heart: emerging prospect for therapeutic targeting. Heart. 2018;104(4):293–299. doi: 10.1136/heartjnl-2017-311448. [DOI] [PubMed] [Google Scholar]
  • 7.Tate M., Grieve D.J., Ritchie R.H. Are targeted therapies for diabetic cardiomyopathy on the horizon? Clin Sci (Lond) 2017;131(10):897–915. doi: 10.1042/cs20160491. [DOI] [PubMed] [Google Scholar]
  • 8.Asbun J., Manso A.M., Villarreal F.J. Profibrotic influence of high glucose concentration on cardiac fibroblast functions: effects of losartan and vitamin E. Am J Physiol Heart Circ Physiol. 2005;288(1):H227–H234. doi: 10.1152/ajpheart.00340.2004. [DOI] [PubMed] [Google Scholar]
  • 9.Singh V.P., Le B., Khode R., Baker K.M., Kumar R. Intracellular angiotensin II production in diabetic rats is correlated with cardiomyocyte apoptosis, oxidative stress, and cardiac fibrosis. Diabetes. 2008;57(12):3297–3306. doi: 10.2337/db08-0805. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Aragno M., Mastrocola R., Alloatti G., Vercellinatto I., Bardini P., Geuna S., et al. Oxidative stress triggers cardiac fibrosis in the heart of diabetic rats. Endocrinology. 2008;149(1):380–388. doi: 10.1210/en.2007-0877. [DOI] [PubMed] [Google Scholar]
  • 11.Yue Y., Meng K., Pu Y., Zhang X. Transforming growth factor beta (TGF-β) mediates cardiac fibrosis and induces diabetic cardiomyopathy. Diabetes Res Clin Pract. 2017;133:124–130. doi: 10.1016/j.diabres.2017.08.018. [DOI] [PubMed] [Google Scholar]
  • 12.Ihm S.H., Chang K., Kim H.Y., Baek S.H., Youn H.J., Seung K.B., et al. Peroxisome proliferator-activated receptor-gamma activation attenuates cardiac fibrosis in type 2 diabetic rats: the effect of rosiglitazone on myocardial expression of receptor for advanced glycation end products and of connective tissue growth factor. Basic Res Cardiol. 2010;105(3):399–407. doi: 10.1007/s00395-009-0071-x. [DOI] [PubMed] [Google Scholar]
  • 13.Pinto A.R., Ilinykh A., Ivey M.J., Kuwabara J.T., D'Antoni M.L., Debuque R., et al. Revisiting Cardiac Cellular Composition. Circ Res. 2016;118(3):400–409. doi: 10.1161/circresaha.115.307778. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Miragoli M., Gaudesius G., Rohr S. Electrotonic modulation of cardiac impulse conduction by myofibroblasts. Circ Res. 2006;98(6):801–810. doi: 10.1161/01.RES.0000214537.44195.a3. [DOI] [PubMed] [Google Scholar]
  • 15.Souders C.A., Bowers S.L., Baudino T.A. Cardiac fibroblast: the renaissance cell. Circ Res. 2009;105(12):1164–1176. doi: 10.1161/circresaha.109.209809. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Shinde A.V., Frangogiannis N.G. Fibroblasts in myocardial infarction: a role in inflammation and repair. J Mol Cell Cardiol. 2014;70:74–82. doi: 10.1016/j.yjmcc.2013.11.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Levick S.P., Widiapradja A. The Diabetic Cardiac Fibroblast: Mechanisms Underlying Phenotype and Function. Int J Mol Sci. 2020;21(3) doi: 10.3390/ijms21030970. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Song L., Yuan J., Liu Y., Zhang D., Zhang C., Lin Q., et al. Ghrelin system is involved in improvements in glucose metabolism mediated by hyperbaric oxygen treatment in a streptozotocin-induced type 1 diabetes mouse model. Mol Med Rep. 2020;22(5):3767–3776. doi: 10.3892/mmr.2020.11481. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.El Hajj E.C., El Hajj M.C., Ninh V.K., Bradley J.M., Claudino M.A., Gardner J.D. Detrimental role of lysyl oxidase in cardiac remodeling. J Mol Cell Cardiol. 2017;109:17–26. doi: 10.1016/j.yjmcc.2017.06.013. [DOI] [PubMed] [Google Scholar]
  • 20.El Hajj E.C., El Hajj M.C., Ninh V.K., Gardner J.D. Cardioprotective effects of lysyl oxidase inhibition against volume overload-induced extracellular matrix remodeling. Exp Biol Med (Maywood) 2016;241(5):539–549. doi: 10.1177/1535370215616511. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.El Hajj E.C., El Hajj M.C., Ninh V.K., Gardner J.D. Inhibitor of lysyl oxidase improves cardiac function and the collagen/MMP profile in response to volume overload. Am J Physiol Heart Circ Physiol. 2018;315(3):H463–H473. doi: 10.1152/ajpheart.00086.2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Zhou Y., Zhou B., Pache L., Chang M., Khodabakhshi A.H., Tanaseichuk O., et al. Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat Commun. 2019;10(1):1523. doi: 10.1038/s41467-019-09234-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Hänzelmann S., Castelo R., Guinney J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinf. 2013;14:7. doi: 10.1186/1471-2105-14-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Vento-Tormo R., Efremova M., Botting R.A., Turco M.Y., Vento-Tormo M., Meyer K.B., et al. Single-cell reconstruction of the early maternal-fetal interface in humans. Nature. 2018;563(7731):347–353. doi: 10.1038/s41586-018-0698-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Rong J., Tao X., Lin Y., Zheng H., Ning L., Lu H.S., et al. Loss of Hepatic Angiotensinogen Attenuates Sepsis-Induced Myocardial Dysfunction. Circ Res. 2021;129(5):547–564. doi: 10.1161/circresaha.120.318075. [DOI] [PubMed] [Google Scholar]
  • 26.Chen J., Fu C.Y., Shen G., Wang J., Xu L., Li H., et al. Macrophages induce cardiomyocyte ferroptosis via mitochondrial transfer. Free Radic Biol Med. 2022;190:1–14. doi: 10.1016/j.freeradbiomed.2022.07.015. [DOI] [PubMed] [Google Scholar]
  • 27.Frangogiannis N.G., Dewald O., Xia Y., Ren G., Haudek S., Leucker T., et al. Critical role of monocyte chemoattractant protein-1/CC chemokine ligand 2 in the pathogenesis of ischemic cardiomyopathy. Circulation. 2007;115(5):584–592. doi: 10.1161/circulationaha.106.646091. [DOI] [PubMed] [Google Scholar]
  • 28.González-Santamaría J., Villalba M., Busnadiego O., López-Olañeta M.M., Sandoval P., Snabel J., et al. Matrix cross-linking lysyl oxidases are induced in response to myocardial infarction and promote cardiac dysfunction. Cardiovasc Res. 2016;109(1):67–78. doi: 10.1093/cvr/cvv214. [DOI] [PubMed] [Google Scholar]
  • 29.Lu M., Qin Q., Yao J., Sun L., Qin X. Induction of LOX by TGF-β1/Smad/AP-1 signaling aggravates rat myocardial fibrosis and heart failure. IUBMB Life. 2019;71(11):1729–1739. doi: 10.1002/iub.2112. [DOI] [PubMed] [Google Scholar]
  • 30.Remst D.F., Blom A.B., Vitters E.L., Bank R.A., van den Berg W.B., Blaney Davidson E.N., et al. Gene expression analysis of murine and human osteoarthritis synovium reveals elevation of transforming growth factor β-responsive genes in osteoarthritis-related fibrosis. Arthritis Rheumatol. 2014;66(3):647–656. doi: 10.1002/art.38266. [DOI] [PubMed] [Google Scholar]
  • 31.Belangero P.S., Leal M.F., Cohen C., Figueiredo E.A., Smith M.C., Andreoli C.V., et al. Expression analysis of genes involved in collagen cross-linking and its regulation in traumatic anterior shoulder instability. J Orthop Res. 2016;34(3):510–517. doi: 10.1002/jor.22984. [DOI] [PubMed] [Google Scholar]
  • 32.Russo I., Frangogiannis N.G. Diabetes-associated cardiac fibrosis: Cellular effectors, molecular mechanisms and therapeutic opportunities. J Mol Cell Cardiol. 2016;90:84–93. doi: 10.1016/j.yjmcc.2015.12.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Bugger H., Abel E.D. Molecular mechanisms of diabetic cardiomyopathy. Diabetologia. 2014;57(4):660–671. doi: 10.1007/s00125-014-3171-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Voloshenyuk T.G., Hart A.D., Khoutorova E., Gardner J.D. TNF-α increases cardiac fibroblast lysyl oxidase expression through TGF-β and PI3Kinase signaling pathways. Biochem Biophys Res Commun. 2011;413(2):370–375. doi: 10.1016/j.bbrc.2011.08.109. [DOI] [PubMed] [Google Scholar]
  • 35.Tayebjee M.H., MacFadyen R.J., Lip G.Y. Extracellular matrix biology: a new frontier in linking the pathology and therapy of hypertension? J Hypertens. 2003;21(12):2211–2218. doi: 10.1097/01.hjh.0000098178.36890.81. [DOI] [PubMed] [Google Scholar]
  • 36.Collier P., Watson C.J., van Es M.H., Phelan D., McGorrian C., Tolan M., et al. Getting to the heart of cardiac remodeling; how collagen subtypes may contribute to phenotype. J Mol Cell Cardiol. 2012;52(1):148–153. doi: 10.1016/j.yjmcc.2011.10.002. [DOI] [PubMed] [Google Scholar]
  • 37.Al-U'datt D., Allen B.G., Nattel S. Role of the lysyl oxidase enzyme family in cardiac function and disease. Cardiovasc Res. 2019;115(13):1820–1837. doi: 10.1093/cvr/cvz176. [DOI] [PubMed] [Google Scholar]
  • 38.Nakasaki M., Hwang Y., Xie Y., Kataria S., Gund R., Hajam E.Y., et al. The matrix protein Fibulin-5 is at the interface of tissue stiffness and inflammation in fibrosis. Nat Commun. 2015;6:8574. doi: 10.1038/ncomms9574. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Liu G., Cooley M.A., Jarnicki A.G., Hsu A.C., Nair P.M., Haw T.J., et al. Fibulin-1 regulates the pathogenesis of tissue remodeling in respiratory diseases. JCI Insight. 2016;1(9) doi: 10.1172/jci.insight.86380. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Donato R., Sorci G., Giambanco I. S100A6 protein: functional roles. Cell Mol Life Sci. 2017;74(15):2749–2760. doi: 10.1007/s00018-017-2526-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Mofid A., Newman N.S., Lee P.J., Abbasi C., Matkar P.N., Rudenko D., et al. Cardiac Overexpression of S100A6 Attenuates Cardiomyocyte Apoptosis and Reduces Infarct Size After Myocardial Ischemia-Reperfusion. J Am Heart Assoc. 2017;6(2) doi: 10.1161/jaha.116.004738. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Niu Y.Y., Zhang Y.Y., Zhu Z., Zhang X.Q., Liu X., Zhu S.Y., et al. Elevated intracellular copper contributes a unique role to kidney fibrosis by lysyl oxidase mediated matrix crosslinking. Cell Death Dis. 2020;11(3):211. doi: 10.1038/s41419-020-2404-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Nguyen X.X., Nishimoto T., Takihara T., Mlakar L., Bradshaw A.D., Feghali-Bostwick C. Lysyl oxidase directly contributes to extracellular matrix production and fibrosis in systemic sclerosis. Am J Physiol Lung Cell Mol Physiol. 2021;320(1):L29–L40. doi: 10.1152/ajplung.00173.2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Yao Y., Findlay A., Stolp J., Rayner B., Ask K., Jarolimek W. Pan-Lysyl Oxidase Inhibitor PXS-5505 Ameliorates Multiple-Organ Fibrosis by Inhibiting Collagen Crosslinks in Rodent Models of Systemic Sclerosis. Int J Mol Sci. 2022;23(10) doi: 10.3390/ijms23105533. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Chen J., Ren J., Loo W.T.Y., Hao L., Wang M. Lysyl oxidases expression and histopathological changes of the diabetic rat nephron. Mol Med Rep. 2018;17(2):2431–2441. doi: 10.3892/mmr.2017.8182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Kim D., Mecham R.P., Trackman P.C., Roy S. Downregulation of Lysyl Oxidase Protects Retinal Endothelial Cells From High Glucose-Induced Apoptosis. Invest Ophthalmol Vis Sci. 2017;58(5):2725–2731. doi: 10.1167/iovs.16-21340. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.López B., González A., Hermida N., Valencia F., de Teresa E., Díez J. Role of lysyl oxidase in myocardial fibrosis: from basic science to clinical aspects. Am J Physiol Heart Circ Physiol. 2010;299(1):H1–H9. doi: 10.1152/ajpheart.00335.2010. [DOI] [PubMed] [Google Scholar]
  • 48.Sethi A., Wordinger R.J., Clark A.F. Focus on molecules: lysyl oxidase. Exp Eye Res. 2012;104:97–98. doi: 10.1016/j.exer.2012.02.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Smith-Mungo L.I., Kagan H.M. Lysyl oxidase: properties, regulation and multiple functions in biology. Matrix Biol. 1998;16(7):387–398. doi: 10.1016/s0945-053x(98)90012-9. [DOI] [PubMed] [Google Scholar]
  • 50.Mäki JM. Lysyl oxidases in mammalian development and certain pathological conditions. Histol Histopathol. 2009;24(5):651-60.10.14670/hh-24.651. [DOI] [PubMed]
  • 51.Voloshenyuk T.G., Landesman E.S., Khoutorova E., Hart A.D., Gardner J.D. Induction of cardiac fibroblast lysyl oxidase by TGF-β1 requires PI3K/Akt, Smad3, and MAPK signaling. Cytokine. 2011;55(1):90–97. doi: 10.1016/j.cyto.2011.03.024. [DOI] [PubMed] [Google Scholar]
  • 52.Li J.H., Huang X.R., Zhu H.J., Johnson R., Lan H.Y. Role of TGF-beta signaling in extracellular matrix production under high glucose conditions. Kidney Int. 2003;63(6):2010–2019. doi: 10.1046/j.1523-1755.2003.00016.x. [DOI] [PubMed] [Google Scholar]
  • 53.Han D.C., Isono M., Hoffman B.B., Ziyadeh F.N. High glucose stimulates proliferation and collagen type I synthesis in renal cortical fibroblasts: mediation by autocrine activation of TGF-beta. J Am Soc Nephrol. 1999;10(9):1891–1899. doi: 10.1681/asn.V1091891. [DOI] [PubMed] [Google Scholar]
  • 54.Ruiz-Ortega M., Rodríguez-Vita J., Sanchez-Lopez E., Carvajal G., Egido J. TGF-beta signaling in vascular fibrosis. Cardiovasc Res. 2007;74(2):196–206. doi: 10.1016/j.cardiores.2007.02.008. [DOI] [PubMed] [Google Scholar]
  • 55.Green L.C., Anthony S.R., Slone S., Lanzillotta L., Nieman M.L., Wu X., et al. Human antigen R as a therapeutic target in pathological cardiac hypertrophy. JCI Insight. 2019;4(4) doi: 10.1172/jci.insight.121541. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Juneja T., Kazmi M., Mellace M., Saidi R.F. Utilization of Treg Cells in Solid Organ Transplantation. Front Immunol. 2022;13 doi: 10.3389/fimmu.2022.746889. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Annes J.P., Munger J.S., Rifkin D.B. Making sense of latent TGFbeta activation. J Cell Sci. 2003;116(Pt 2):217–224. doi: 10.1242/jcs.00229. [DOI] [PubMed] [Google Scholar]
  • 58.Wang J., Shen W., Zhang J.Y., Jia C.H., Xie M.L. Stevioside attenuates isoproterenol-induced mouse myocardial fibrosis through inhibition of the myocardial NF-κB/TGF-β1/Smad signaling pathway. Food Funct. 2019;10(2):1179–1190. doi: 10.1039/c8fo01663a. [DOI] [PubMed] [Google Scholar]
  • 59.Sethi A., Mao W., Wordinger R.J., Clark A.F. Transforming growth factor-beta induces extracellular matrix protein cross-linking lysyl oxidase (LOX) genes in human trabecular meshwork cells. Invest Ophthalmol Vis Sci. 2011;52(8):5240–5250. doi: 10.1167/iovs.11-7287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Hornstra I.K., Birge S., Starcher B., Bailey A.J., Mecham R.P., Shapiro S.D. Lysyl oxidase is required for vascular and diaphragmatic development in mice. J Biol Chem. 2003;278(16):14387–14393. doi: 10.1074/jbc.M210144200. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary data 1
mmc1.docx (1.4MB, docx)

Data Availability Statement

The accession number for the raw data files of the single-cell sequencing reported in this paper are uploaded on NCBI GEO dataset (GSE 213337). The accession number of Bulk RNA sequencing number is: GSE 215979. All data are available.

Articles from Journal of Advanced Research are provided here courtesy of Elsevier

RESOURCES