In experimental animals, bradykinin type-1 receptors (BK-1Rs) are induced during inflammation and ischemia, and, by exerting either cardioprotective or cardiotoxic effects, they may contribute to the pathogenesis of heart failure. Nothing is known about the expression of BK-1Rs in human heart failure. Human heart tissue was obtained from excised hearts of patients undergoing cardiac transplantation (n = 13), due to idiopathic dilated cardiomyopathy (IDC; n = 7) or to coronary heart disease (CHD; n = 6), and from normal hearts (n = 6). The expression of BK-1Rs was analyzed by means of competitive RT-PCR, Western blot analysis, and immunohistochemistry. Expression of BK-1R mRNA was increased in both IDC (2.8-fold) and CHD (2.1-fold) hearts compared with normal hearts. The observed changes were verified at the protein level. Expression of BK-1Rs in failing hearts localized to the endothelium of intramyocardial coronary vessels and correlated with an increased expression of TNF-alpha in the vessel wall. Treatment of human coronary artery endothelial cells with TNF-alpha increases their BK-1R expression. These novel results show that BK-1Rs are induced in the endothelium of intramyocardial coronary vessels in failing human hearts and so may participate in the pathogenesis of heart failure.
Abstract The transcription factor SOX9 is a key regulator of multiple developmental processes, and is frequently re-expressed in non-small cell lung cancer (NSCLC). Its precise role in the progression of NSCLC histopathologies has however remained elusive. We show that SOX9 expression relates to poor outcome and invasive histopathology in human adenocarcinomas, and is absent in murine early minimally invasive and human in situ adenocarcinoma. Interestingly, despite wide SOX9 expression across advanced NSCLC histotypes, its genetic deletion in the murine Kras G12D ;Lkb1 -/- model selectively disrupted only the growth of papillary NSCLC, without affecting the initiation of precursor lesions or growth of mucinous or squamous tissue. Spatial tissue phenotyping indicated a requirement of SOX9 expression for the progression of surfactant protein C-expressing progenitor cells, which gave rise to papillary tumours. Intriguingly, while SOX9 expression was dispensable for squamous tissue formation, its loss in fact led to enhanced squamous tumour metastasis, which was associated with altered collagen IV deposition in the basement membrane. Our work therefore demonstrates histopathology-selective roles for SOX9 in NSCLC progression, namely a requirement for papillary adenocarcinoma progression, but opposing metastasis-suppressing function in squamous histotype tissue. This attests to a pleiotropic SOX9 function, linked to the cell of origin and microenvironmental tissue contexts.
Pleural mesothelioma (MPM) is an aggressive malignancy with an average patient survival of only 10 months. Interestingly, about 5%-10% of the patients survive remarkably longer. Prior studies have suggested that the tumor immune microenvironment (TIME) has potential prognostic value in MPM. We hypothesized that high-resolution single-cell spatial profiling of the TIME would make it possible to identify subpopulations of patients with long survival and identify immunophenotypes for the development of novel treatment strategies.We used multiplexed fluorescence immunohistochemistry (mfIHC) and cell-based image analysis to define spatial TIME immunophenotypes in 69 patients with epithelioid MPM (20 patients surviving ≥ 36 months). Five mfIHC panels (altogether 21 antibodies) were used to classify tumor-associated stromal cells and different immune cell populations. Prognostic associations were evaluated using univariate and multivariable Cox regression, as well as combination risk models with area under receiver operating characteristic curve (AUROC) analyses.We observed that type M2 pro-tumorigenic macrophages (CD163+pSTAT1-HLA-DRA1-) were independently associated with shorter survival, whereas granzyme B+ cells and CD11c+ cells were independently associated with longer survival. CD11c+ cells were the only immunophenotype increasing the AUROC (from 0.67 to 0.84) when added to clinical factors (age, gender, clinical stage, and grade).High-resolution, deep profiling of TIME in MPM defined subgroups associated with both poor (M2 macrophages) and favorable (granzyme B/CD11c positivity) patient survival. CD11c positivity stood out as the most potential prognostic cell subtype adding prediction power to the clinical factors. These findings help to understand the critical determinants of TIME for risk and therapeutic stratification purposes in MPM.
Currently available imaging techniques have limited specificity for the detection of active myocardial inflammation. Aluminum 18F-labeled 1,4,7-triazacyclononane-N,N′,N″-triacetic acid conjugated folate (18F-FOL) is a PET tracer targeting folate receptor β (FR-β), which is expressed on activated macrophages at sites of inflammation. We evaluated 18F-FOL PET for the detection of myocardial inflammation in rats with autoimmune myocarditis and studied the expression of FR-β in human cardiac sarcoidosis specimens. Methods: Myocarditis was induced by immunizing rats (n = 18) with porcine cardiac myosin in complete Freund adjuvant. Control rats (n = 6) were injected with Freund adjuvant alone. 18F-FOL was intravenously injected, followed by imaging with a small-animal PET/CT scanner and autoradiography. Contrast-enhanced high-resolution CT or 18F-FDG PET images were used for coregistration. Rat tissue sections and myocardial autopsy samples from 6 patients with cardiac sarcoidosis were studied for macrophages and FR-β. Results: The myocardium of 10 of 18 immunized rats showed focal macrophage-rich inflammatory lesions, with FR-β expression occurring mainly in M1-polarized macrophages. PET images showed focal myocardial 18F-FOL uptake colocalizing with inflammatory lesions (SUVmean, 2.1 ± 1.1), whereas uptake in the remote myocardium of immunized rats and controls was low (SUVmean, 0.4 ± 0.2 and 0.4 ± 0.1, respectively; P < 0.01). Ex vivo autoradiography of tissue sections confirmed uptake of 18F-FOL in myocardial inflammatory lesions. Uptake of 18F-FOL in inflamed myocardium was efficiently blocked by a nonlabeled FR-β ligand folate glucosamine in vivo. The myocardium of patients with cardiac sarcoidosis showed many FR-β–positive macrophages in inflammatory lesions. Conclusion: In a rat model of autoimmune myocarditis, 18F-FOL shows specific uptake in inflamed myocardium containing macrophages expressing FR-β, which were also present in human cardiac sarcoid lesions. Imaging of FR-β expression is a potential approach for the detection of active myocardial inflammation.
Article Figures and data Abstract eLife digest Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract Aging, obesity, hypertension, and physical inactivity are major risk factors for endothelial dysfunction and cardiovascular disease (CVD). We applied fluorescence-activated cell sorting (FACS), RNA sequencing, and bioinformatic methods to investigate the common effects of CVD risk factors in mouse cardiac endothelial cells (ECs). Aging, obesity, and pressure overload all upregulated pathways related to TGF-β signaling and mesenchymal gene expression, inflammation, vascular permeability, oxidative stress, collagen synthesis, and cellular senescence, whereas exercise training attenuated most of the same pathways. We identified collagen chaperone Serpinh1 (also called as Hsp47) to be significantly increased by aging and obesity and repressed by exercise training. Mechanistic studies demonstrated that increased SERPINH1 in human ECs induced mesenchymal properties, while its silencing inhibited collagen deposition. Our data demonstrate that CVD risk factors significantly remodel the transcriptomic landscape of cardiac ECs inducing inflammatory, senescence, and mesenchymal features. SERPINH1 was identified as a potential therapeutic target in ECs. eLife digest Cardiovascular diseases are the number one cause of death in the western world. Endothelial cells that line the blood vessels of the heart play a central role in the development of these diseases. In addition to helping transport blood, these cells support the normal running of the heart, and help it to grow and regenerate. Over time as the body ages and experiences stress, endothelial cells start to deteriorate. This can cause the cells to undergo senescence and stop dividing, and lay down scar-like tissue via a process called fibrosis. As a result, the blood vessels start to stiffen and become less susceptible to repair. Ageing, obesity, high blood pressure, and inactivity all increase the risk of developing cardiovascular diseases, whereas regular exercise has a protective effect. But it was unclear how these different factors affect endothelial cells. To investigate this, Hemanthakumar et al. compared the gene activity of different sets of mice: old vs young, obese vs lean, heart problems vs healthy, and fit vs sedentary. All these risk factors – age, weight, inactivity and heart defects – caused the mice’s endothelial cells to activate mechanisms that lead to stress, senescence and fibrosis. Whereas exercise training had the opposite effect, and turned off the same genes and pathways. All of the at-risk groups also had high levels of a gene called SerpinH1, which helps produce tissue fiber and collagen. Experiments increasing the levels of SerpinH1 in human endothelial cells grown in the laboratory recreated the effects seen in mice, and switched on markers of stress, senescence and fibrosis. According to the World Health Organization, cardiovascular disease now accounts for 10% of the disease burden worldwide. Revealing the affects it has on gene activity could help identify new targets for drug development, such as SerpinH1. Understanding the molecular effects of exercise on blood vessels could also aid in the design of treatments that mimic exercise. This could help people who are unable to follow training programs to reduce their risk of cardiovascular disease. Introduction According to WHO, cardiovascular diseases (CVDs) account for 10% of the global disease burden and constitute the number 1 cause of death in the western world. CVDs are mainly caused by behavioral (physical inactivity, unhealthy diet) and metabolic (obesity, hypertension, diabetes, high cholesterol) risk factors (Mendis et al., 2011). Aging, however, is by far the biggest contributor to CVD, and aging population is becoming an enormous challenge worldwide. The heart contains a dense vascular network, and endothelial cells (ECs) are indeed the most abundant cell population in the adult mouse heart (Pinto et al., 2016). In addition to their transport function, ECs are defined to control vasomotor tone, maintain vascular homeostasis, regulate angiogenesis, and establish bidirectional communication with other cell types and organs via paracrine signaling mechanisms (Aird, 2007; Aird, 2012; Kivelä et al., 2019; Talman and Kivelä, 2018; Hemanthakumar and Kivelä, 2020). ECs are found to be highly adaptive to physiological stimuli during normal growth and development (White et al., 1998; Bloor, 2005), and the diversity of ECs in different tissues has now been acknowledged. ECs are also maladaptive to a spectrum of pathological events involving, for example, inflammation or oxidative stress (Cines et al., 1998; Gimbrone and García-Cardeña, 2016), and the development of heart diseases is strongly linked to endothelial dysfunction and impaired vascular remodeling. However, the molecular cues, which cause maladaptation and dysfunction of ECs in the heart in response to pathological signals, remain elusive. Physical inactivity increases the incidence of several chronic diseases, whereas regular exercise training has positive effects on most of our tissues (Hawley et al., 2014). Because microcirculation is present in every organ in the body, ECs have a unique ability to influence the homeostasis and function of different tissues, and they are potentially a major cell type mediating the positive effects of exercise throughout the body. Although the cardiac benefits of exercise are clear and there have been major advances in unraveling the molecular mechanisms, the understanding of how the molecular effects are linked to health benefits is still lacking (Hawley et al., 2014). Especially, the effects of exercise on ECs have not been characterized. We hypothesized that the major CVD risk factors aging, obesity, and pressure overload will induce adverse remodeling of cardiac EC transcriptome (Gimbrone and García-Cardeña, 2016; Ungvari et al., 2018; Brandes, 2014), whereas exercise training would provide beneficial effects (White et al., 1998; Bloor, 2005). Both physiological and pathological stimuli significantly modified the cardiac EC transcriptome. Intriguingly, our results demonstrated that CVD risk factors promoted activation of transforming growth factor-β (TGF-β) signaling, inflammatory response, cellular senescence, and induced mesenchymal gene expression in cardiac EC, whereas exercise training promoted opposite protective effects. Results Exercise training and CVD risk factors modulate cardiac EC number, vascular density, and transcriptome To mimic the effect of the most common CVD risk factors (aging, obesity, pressure overload/hypertension, and physical inactivity), we used adult C57BL/6J wild-type mice in the following experimental groups: aged (18 months) vs. young (2 months) mice, high-fat diet (HFD) induced obesity (14 weeks HFD) vs. lean mice, transverse aortic constriction (TAC) vs. sham-operated mice, and exercise training (progressive treadmill running for 6 weeks) vs. sedentary mice (Figure 1—figure supplement 1A,B). Exercise trained mice showed improved ejection fraction compared to the sedentary mice, whereas aging, HFD, and TAC resulted in impaired heart function (Figure 1—figure supplement 1C–F and Figure 1—source data 2). HFD also induced marked weight gain, increased fat mass, and impaired glucose tolerance (Figure 1—figure supplement 1G–I). Left ventricular (LV) mass was increased in aged, HFD-treated, and TAC mice (Figure 1—source data 2). Exercise training also slightly increased LV mass, which reflects mild physiological hypertrophy often observed in endurance-trained athletes (Arbab-Zadeh et al., 2014; Figure 1—source data 2). Exercise training significantly increased, whereas aging, HFD, and TAC decreased the percentage, count, and mean fluorescence intensity of the cardiac ECs (CD31+CD140a-CD45-Ter119-DAPI-) compared to the controls, when analyzed by fluorescence-activated cell sorting (FACS; Figure 1A,B, Figure 1—figure supplement 2A–D). This was also demonstrated by immunohistochemistry for CD31-positive coronary vessels (Figure 1C,D). The cardiac ECs were gated and sorted by FACS (Figure 1—figure supplement 3A), and the isolated ECs were first analyzed by quantitative PCR analysis, which indicated significant enrichment of EC markers Cdh5 and Tie1 in the sorted fraction compared to whole heart or other cardiac mononuclear cells (Figure 1—figure supplement 3B). In addition, isolation resulted in 87.4 ± 1.9% cell viability and RNA purification strategy yielded intact and stable RNA with average RNA integrity number (RIN) of 8.7 (Figure 1—figure supplement 3C,D). RNA sequencing of isolated ECs was used to profile the expression pattern of cardiac EC transcripts in different experimental groups. Two-dimensional PCA of the EC transcriptomes exhibited significant proportion of variance in the gene expression pattern, which can be attributed to the treatment-induced changes in cardiac EC transcriptome (Figure 2—figure supplement 1A–E). Notably, unsupervised hierarchical clustering of EC data sets for all experimental interventions (sedentary, exercise trained, young, aged, sham, TAC) revealed consistent clustering and high degree of similarity in the gene expression pattern (Figure 2—figure supplement 1F–J). The analysis for differentially expressed genes (DEGs) showed a large number of up- and downregulated genes especially in aged, obese, and TAC-operated mice followed by a smaller number of affected genes in exercise trained mice. The number of significantly up- and downregulated genes with the false discovery rate (FDR) 0.05 for each treatment are shown in the MA plots and the top 50 DEGs for each treatment are presented by heat maps (Figure 2A–E,F–J). Figure 1 with 3 supplements see all Download asset Open asset Effects of exercise training, aging, obesity, and pressure overload on cardiac endothelial cell (EC) number and vascular density. (A and B) Fluorescence-activated cell sorting (FACS) analysis and quantification of mean fluorescence intensity (MFI) of the cardiac ECs (CD31+CD140a-CD45-Ter119-DAPI-) in various mouse models. (C and D) Representative immunofluorescence images and quantification of CD31+ blood vessel area (%) in the heart. Scale bar 100 μm. Data is presented as mean ± SEM. Student’s t-test was used, *p<0.05, **p<0.01, ***p<0.001. In panel (B), each color-coded circle indicates an individual biological sample. In panel (D), the number of mice in each experimental group is indicated in the respective graph, N = 3–5 male mice/group. Figure 1—source data 1 Source data for Figure 1B and D. https://cdn.elifesciences.org/articles/62678/elife-62678-fig1-data1-v2.xlsx Download elife-62678-fig1-data1-v2.xlsx Figure 1—source data 2 Echocardiography measurements of cardiac function and ventricular dimensions in the indicated experimental group. https://cdn.elifesciences.org/articles/62678/elife-62678-fig1-data2-v2.pdf Download elife-62678-fig1-data2-v2.pdf Figure 2 with 2 supplements see all Download asset Open asset Transcriptomic changes in cardiac endothelial cells (ECs) in exercise trained, aged, obese, and transverse aortic constriction (TAC)-treated mice. (A–E) MA-plots (log ratio over mean) showing the number of differentially expressed genes (DEGs) in cardiac ECs for each experiment. Number of significantly up- and downregulated genes with the false discovery rate (FDR; Benjamini–Hochberg adjusted p-value) threshold of 0.05 are indicated in the plots. (F–J) Top 50 DEGs in cardiac ECs of the indicated experimental groups. In the heatmap, each color-coded circle (red, green, and black) indicates an individual biological sample within each experimental group. N = 3–4 male mice/group. Figure 2—source data 1 Source data for Figure 2F, G, H, I and J. https://cdn.elifesciences.org/articles/62678/elife-62678-fig2-data1-v2.xlsx Download elife-62678-fig2-data1-v2.xlsx CVD risk factors induce senescence and TGF-β signaling together with mesenchymal gene expression in cardiac ECs To understand the biological functions of the DEGs, we used PANTHER classification analysis (Figure 3A). The analysis revealed that genes related to EC development, adherence junction organization, IGFR signaling, adrenomedullin receptor signaling, and mitochondria were upregulated by exercise training. Furthermore, exercise training downregulated pathways related to cellular aging, vascular membrane permeability, negative regulation of angiogenesis, TGF-β1 production, collagen activated tyrosine kinase signaling, and ossification. In contrast, pathways related to TGF-β, IFNα, TNFα, oxidative stress, EC differentiation, vascular permeability, cell aging, collagen synthesis, SMAD signaling, and mesenchymal cell development were highly enriched in cardiac EC from both aged and obese mice. Downregulated pathways in these mice included tissue and lipid homeostasis, ECM assembly, tube morphogenesis, cell adhesion, cell number maintenance, EC proliferation, vasculature development, artery development, and NOTCH signaling. Pressure overload activated pathways such as cellular response to TGF-βR2 activation of fibrotic pathways, inactivation of cell survival pathways Erk1/2 and MAPK, and ossification process, whereas cellular homeostasis and vasculature development were repressed. Figure 3 with 3 supplements see all Download asset Open asset Cardiovascular disease (CVD) risk factors activate mesenchymal gene expression in cardiac endothelial cells (ECs). (A) Gene ontology analysis of the up- and downregulated genes. Note the opposite changes induced by exercise training compared to the CVD risk factors. (B–F) Heatmaps showing the differential gene expression of endothelial and mesenchymal genes previously associated with endothelial-to-mesenchymal transition (EndMT). Genes are selected based on published data sets (references are found in Figure 3—source data 1). In all panels, the up- and downregulated genes with the false discovery rate (FDR; Benjamini–Hochberg adjusted p-value) threshold of 0.05 were considered. In the heatmap, each color-coded circle (red, green, and black) indicates an individual biological sample within each experimental group. N = 3–4 male mice/group. Figure 3—source data 1 Genes and reference list for endothelial and mesenchymal genes indicated in the Figure 3B–F heat map. (A) Reference list for endothelial and mesenchymal genes indicated in the Figure 3B (EXE vs. SED) heat map. (B) Reference list for endothelial and mesenchymal genes indicated in the Figure 3C (aged vs. young) heat map. (C) Reference list for endothelial and mesenchymal genes indicated in the Figure 3D (high-fat diet [HFD] vs. Chow) heat map. (D) Reference list for endothelial and mesenchymal genes indicated in the Figure 3E (transverse aortic constriction [TAC] [2] vs. Sham) heat map. (E) Reference list for endothelial and mesenchymal genes indicated in the Figure F (TAC [7] vs. Sham) heat map. https://cdn.elifesciences.org/articles/62678/elife-62678-fig3-data1-v2.docx Download elife-62678-fig3-data1-v2.docx Figure 3—source data 2 Source data for Figure 3B, C, D, E and F. https://cdn.elifesciences.org/articles/62678/elife-62678-fig3-data2-v2.xlsx Download elife-62678-fig3-data2-v2.xlsx Comparison of the GO biological terms, which were significantly affected by exercise training and the CVD risk factors, demonstrated clear opposite effects on the EC transcriptome. Aging and HFD promoted oxidative stress response, activation of inflammatory and fibrosis pathways (Figure 3—figure supplement 1A–E) and cellular aging, and inhibited pathways regulating cell number maintenance, proliferation, and lipid homeostasis. Exercise training, in turn, promoted EC homeostasis and vascular growth, and prevented vascular aging, inflammation, and pathological activation. In the cardiac ECs of HFD and TAC-treated mice, a significant upregulation of senescence-associated secretory phenotype (SASP) genes (Figure 3—figure supplement 2C–E) were observed. To validate the link between obesity and senescence in the heart, we performed SA-β-galactosidase staining in HFD- and chow-fed mouse heart sections. In HFD-fed mice, we observed several clusters of SA-β-galactosidase positive cells in the heart, which were not observed in the chow-fed animals (Figure 3—figure supplement 2A). Quantification showed significant increase in these cells (Figure 3—figure supplement 2B). Further studies are needed to identify these cells and their role in CVD development. Because the analyses indicated upregulation of genes and pathways associated with mesenchymal development and endothelial-to-mesenchymal transition (EndMT) by all of the CVD risk factors, we reviewed our DEG sets for the expression of selected endothelial and mesenchymal markers based on the previously published data sets (Figure 3—source data 1). We found significant upregulation of many mesenchymal markers and downregulation of EC genes in aged and obese mice (Figure 3C,D). After 2 weeks of TAC, we also observed upregulation of several mesenchymal markers, whereas after 7 weeks of TAC, there was both up- and downregulation of the EC and mesenchymal markers, indicating possible reversal of the process (Figure 3E,F). Strikingly, exercise training downregulated several EndMT genes (Fscn1, Cd93, Vwa1, Sparc, Tuba1a, Cd44, Trp53, Col4a2, Mest, Cnn2, Tnfaip1, Lamb1, Ltbp4, and Unc5b), the angiogenesis inhibitor gene Vash1, and the endothelial activation marker Apln and its receptor Aplnr, whereas it upregulated the expression of Malat1, Mgp, Krit1, and Calcrl (Figure 3B). We validated the results using an expanded set of samples by qPCR for Apln, Vim, Tgfbr2, Vash1, Sparc, and Tgfb1 (Figure 3—figure supplement 3A–F). Serpinh1 expression is increased by aging and obesity and repressed by exercise training To identify genes, which could mediate the negative effect of aging and obesity and the protective effects of exercise, we performed gene overlap analysis of DEGs from these three experimental interventions. We found four genes significantly affected by all treatments, of which two genes (Serpinh1 and Vwa1) were upregulated by aging and HFD and downregulated by exercise training. The other two genes (Mest and Fhl3) were upregulated by HFD and downregulated by exercise training and aging (Figure 4A–C). We performed an in silico secretome analysis to characterize the properties of the identified genes using MetaSecKB database (Figure 4D). Both Serpinh1 and Vwa1 contain a signal peptide for secretion, indicating they could act as angiocrines in autocrine and/or paracrine fashion. Figure 4 with 3 supplements see all Download asset Open asset Serpinh1 expression is increased by aging and obesity and repressed by exercise training. (A) A Venn diagram showing the overlap of differentially expressed genes between the experiments. Four genes were identified to be significantly affected by aging, obesity, and exercise (Serpinh1, Vwa1, Mest, and Fhl3). (B) Bar plot showing the expression pattern of these four genes. In panels (A and B), the up- and downregulated genes with the false discovery rate (FDR; Benjamini–Hochberg adjusted p-value) threshold of 0.05 were considered to be significant (N = 3–4 male mice/group). (C) qPCR validation of Serpinh1 and Vwa1 normalized to Hprt1 (N = 4–6 male mice/group). (D) In silico secretome analysis of the identified genes. (E–G) Representative immunofluorescent and immunohistochemistry images showing the expression of SERPINH1 in human endothelial cell (EC) and human heart samples. Red arrowhead in the bottom panel F indicates the expression in large vessels and ‘L’ indicates vessel lumen. White arrowheads in the panel G denote the co-expression of SERPINH1 and CDH5 in coronary vessels (yellow signal). (H–K) mRNA expression of Serpinh1, Vwa1, Vim, and Tgfbr2 in the cardiac ECs of sedentary and exercise trained aged mice (N = 4–5 female mice/group). Scale bar 100 μm. Data is presented as mean ± SEM. Student’s t-test was used, *p<0.05, **p<0.01, ***p<0.001. Figure 4—source data 1 Source data for Figure 4B, C, H, I, J and F. https://cdn.elifesciences.org/articles/62678/elife-62678-fig4-data1-v2.xlsx Download elife-62678-fig4-data1-v2.xlsx We focused on Serpinh1, as it has a known role as a collagen chaperone and has been linked to fibrosis (Ito and Nagata, 2019), making it an attractive candidate. We validated the endothelial Serpinh1 expression by qPCR (Figure 4C), and at single cell level using Tabula Muris database (Tabula Muris Consortium et al., 2018) and cardiac EC atlas from the Carmeliet lab (Kalucka et al., 2020). The scRNAseq analysis revealed that Serpinh1 is expressed in variety of cell types within the mouse heart, including fibroblasts, myofibroblasts, smooth muscle cells, ECs, endocardial cells, and to lesser extent in cardiomyocytes (Figure 4—figure supplement 1A–D). In ECs, Serpinh1 was found to be expressed throughout all EC clusters, with the highest expression in the apelin-high cluster marking activated ECs (Figure 4—figure supplement 2A–F). Interestingly, the expression of mesenchymal markers such as Tagln2, Vim, and Smtn was also high in this cluster. Next, we analyzed the expression of SERPINH1 (also called as HSP47) in healthy human heart and in human cardiac ECs. Immunohistochemistry demonstrated SERPINH1 to be highly expressed throughout the coronary vasculature and in fibroblasts in human heart, and weak staining was also detected in cardiomyocytes (Figure 4E–G, Figure 4—figure supplement 1D). Analysis using the EndoDB database (E-GEOD-43475) showed that the expression of SERPINH1 is highly similar in both veins and arteries and in different tissues (heart, lungs, liver, human cardiac arterial EC (HCAEC), and human umbilical venous EC [HUVEC]) (Figure 4—figure supplement 3A). In human cardiac ECs, SERPINH1 was localized perinuclearly, similar to what has been demonstrated in other cells types, and consistent with the ER retention motif in its N-terminus (Figure 4E; Masuda et al., 1994; Razzaque et al., 1998; Honzawa et al., 2014). We next tested, if exercise training can attenuate the expression of Serpinh1, Vwa1, and selected markers of TGF-β signaling/EndMT also in aged mice. Of the studied genes, mRNA expression of Serpinh1 and Vim were significantly repressed by exercise training, and there was a tendency also for Vwa1 (Figure 4H–K). Overexpression of SERPINH1 induces mesenchymal features in human ECs To study the effects of SERPINH1 in human ECs, we produced lentiviral vector encoding myc-tagged hSERPINH1. Both HUVECs and HCAECs were analyzed. SERPINH1 protein was localized similar to the native protein (Figure 5B), and the expression was verified by western blotting (Figure 5—figure supplement 1A). Overexpression of SERPINH1 altered the cellular morphology characterized by impaired or discontinuous vascular endothelial cadherin junctions, increased stress fiber formation, and larger cell size (Figure 5A,B). Furthermore, analysis of EC and mesenchymal cell-related transcripts demonstrated significant repression of EC markers (CD31, CDH5, TIE1, NRARP, and ID1) and induction of a proliferation gene CCND1, and mesenchymal/EndMT markers (TAGLN, aSMA, CD44, VIM, NOTCH3, ZEB2, SLUG, FN1, VCAM1, and ICAM1) (Figure 5C). VE-cadherin downregulation was also confirmed at protein level (Figure 5D) and increased aSMA expression by immunofluorescence staining (Figure 5E). We also analyzed the effect of SERPINH1 on cellular senescence. SA-β-galactosidase staining showed increased number of cells undergoing senescence and there was a clear upregulation of senescence-associated genes (Figure 5G,H). Figure 5 with 1 supplement see all Download asset Open asset Overexpression of SERPINH1 modifies the endothelial cell (EC) phenotype and induces mesenchymal gene expression in human cardiac ECs. (A) Representative phase-contrast images of live human cardiac arterial EC (HCAEC) transduced with LV-CTRL and LV-SERPINH1-Myc, and quantification of the aspect ratio (length to width ratio) of the cell. (B) Representative immunofluorescent images showing the expression of Myc-tagged SERPINH1 in green, F-Actin in gray, and CDH5/VE-Cadherin in red. The inset within the white box shows magnified view of VE-Cadherin junctions in HCAECs. (C) qPCR analysis of endothelial and mesenchymal markers in SERPINH1 overexpressing cells. (D) Western blot analysis and quantification of CDH5/VE-cadherin expression in the SERPINH1 overexpressing HCAECs (normalized to GAPDH). (E) Representative immunofluorescent images showing DAPI in blue, CDH5/VE-Cadherin in green, and α-smooth muscle actin (aSMA) in red. (F) qPCR analysis of SERPINH1 and EndMT markers in HCAECs stimulated with TGF-β1 (50 ng/ml) or H2O2 for 5 days. (G) Representative images and quantification of SA-β-gal+ senescent cells (in blue) normalized to total nuclei (%) in SERPINH1 overexpressing and control cells. (H) qPCR analysis of senescence-associated secretory phenotype (SASP) genes in HCAECs transduced with LV-CTRL and LV-SERPINH1-Myc. In panels A, C, D, F, G, and H, N = 3 biological replicates/group were analyzed. Scale bar 100 μm. Data is presented as mean ± SEM. Student’s t-test was used, *p<0.05, **p<0.01, ***p<0.001. Figure 5—source data 1 Source data for Figure 5A, C, D, F, G and H. https://cdn.elifesciences.org/articles/62678/elife-62678-fig5-data1-v2.xlsx Download elife-62678-fig5-data1-v2.xlsx Figure 5—source data 2 Source data for Figure 5D. https://cdn.elifesciences.org/articles/62678/elife-62678-fig5-data2-v2.pptx Download elife-62678-fig5-data2-v2.pptx Transcriptomic changes pointed toward activated TGF-β signaling and oxidative stress in response to all of the CVD risk factors. Both are known to contribute to EC dysfunction and EndMT, and thus we tested if they act as upstream regulators of SERPINH1. Indeed, our results show that TGF-β1-treatment of HCAECs significantly upregulated the expression of SERPINH1 together with other known EndMT markers, and there was also small but significant induction of SERPINH1 by hydrogen peroxide treatment (Figure 5F). SERPINH1 is needed for collagen 1 deposition by ECs To investigate the significance of SERPINH1 depletion in human cardiac ECs, HCAECs were transduced with four independent shSERPINH1 lentiviral constructs. The constructs induced approximately 80% deletion of SERPINH1 mRNA (Figure 6D). The cell morphology was not affected after 2 days (Figure 6A), but 10 days of silencing significantly changed EC morphology and decreased the cell density in culture (Figure 6B), suggesting that SERPINH1 might play a role in EC homeostasis and survival. SERPINH1 silencing significantly inhibited collagen fibril deposition, detected by immunohistochemistry for type 1 collagen (Figure 6B,C). Only the cells transduced with the construct #1 could produce some extracellular collagen 1, and these cells also survived better than the cells transduced with constructs #2, #3, or #4 (Figure 6B,C). Next, we treated the cells with TGF-β1 and hydrogen peroxide for 5 days to induce EndMT features, as described previously (Evrard et al., 2016; Magenta et al., 2011). We used the shSERPINH1 (#1) construct, because from the other silencing constructs not enough cells survived for the experiments. The results indicated that silencing of SERPINH1 prevented the appearance of Taglin-positive cells, a commonly used readout for EndMT, which were observed in the control cells (Figure 6E). Figure 6 Download asset Open asset SERPINH1 silencing in human cardiac endothelial cell (EC) inhibits collagen production and EndMT. (A) Representative phase contrast images of live human cardiac arterial ECs (HCAECs) transduced with LV-SCR and LV-shSERPINH1 (#1 and #2) and quantification of the aspect ratio (length to width ratio) of the cells 48 hr after transduction. (B) Representative CDH5/VE-Cadherin immunofluorescent images (green) showing the cell morphology and density after 10 days of SERPINH1 silencing. Collagen 1 staining is shown in red, and quantification of collagen 1 is shown in C. (D) qPCR analysis of SERPINH1 deletion levels using four independent constructs. (E) Representative immunofluorescent images showing TAGLN expression in the control and SERPINH1 silenced HCAECs treated with recombinant human TGF-β1 with and without H2O2 for 5 days. In the panels (A, C and D), N = 3 biological replicates/group were analyzed. Scale bar 100 μm. Data is presented as mean ± SEM. Student’s t-test was used, *p<0.05, **p<0.01, ***p<0.001. Figure 6—source data 1 Source data for Figure 6A, C and D. https://cdn.elifesciences.org/articles/62678/elife-62678-fig6-data1-v2.xlsx Download elife-62678-fig6-data1-v2.xlsx We also studied the effect of SERPINH1 on cell proliferation/migration. In the scratch wound assay, overexpression of SERPINH1 significantly promoted wound closure (Figure 7A,B), whereas silencing of SERPINH1 for 2 days significantly decreased EC proliferation/migration (Figure 7C,D). Cell proliferation was slightly increased by SERPINH1 overexpression, whereas silencing almost completely blocked proliferation, as determined by EdU incorporation (Figure 7E–G). Figure 7 Download asset Open asset SERPINH1 overexpression enhances and silencing inhibits wound closure in vitro. (A and B) Representative phase contrast images of scratch wound healing assay performed in human cardiac arterial ECs (HCAECs) treated with LV-CTRL and LV-SERPINH1, and quantification of the wound closure (%) with respect to time (hours). (C a
Loss of endothelial cells (ECs) with ensuing exposure of thrombogenic subendothelial surface is a common cause of thromboembolic complications in atherosclerotic arteries. Thus, endothelial denudation has emerged as a major contributor to the pathogenesis of atherosclerosis and its complications. Despite ongoing efforts in elucidating the pathogenesis of endothelial erosions in human atherosclerotic arteries, the mechanisms of erosion have remained enigmatic, partly due to lack of well-established methods for its identification. Here the authors point out plausible pitfalls in the current methodology and provide an improved immunohistochemical method for identifying endothelial erosion; i.e., immunofluorescence double staining with antibodies against CD42b and CD31/CD34. This method enables reliable detection of ECs and platelets in the same staining by allowing detection of “pseudoendothelium” caused by CD31 staining of a thin platelet layer covering sites of endothelial erosion. As erosion with a luminal platelet thrombus is likely to represent an in vivo erosion, and erosion without platelets an ex vivo artefact, the method makes it possible to exclude artefactual erosions resulting from sample processing. The novel immunostaining protocol presented here allows more reliable detection of endothelial erosions and so may facilitate studies on the mechanisms involved in the pathogenesis of plaque erosion and acute coronary syndromes.
Asbestos is a global occupational health hazard, and exposure to it by inhalation predisposes to interstitial as well as malignant pulmonary morbidity. Over time, asbestos fibers embedded in lung tissue can become coated with iron-rich proteins and mucopolysaccharides, after which they are called asbestos bodies (ABs) and can be detected in light microscopy (LM). Bronchoalveolar lavage, a cytological sample from the lower airways, is one of the methods for diagnosing lung asbestosis and related morbidity. Search for ABs in these samples is generally laborious and time-consuming. We describe a novel diagnostic method, which implements deep learning neural network technology for the detection of ABs in bronchoalveolar lavage samples (BALs).BALs with suspicion of asbestos exposure were scanned as whole slide images (WSIs) and uploaded to a cloud-based virtual microscopy platform with a neural network training interface. The images were used for training and testing a neural network model capable of recognizing ABs. To prioritize the model's sensitivity, we allowed it to also make false-positive suggestions. To test the model, we compared its performance to standard LM diagnostic data as well as the ground truth (GT) number of ABs, which we established by a thorough manual search of the WSIs.We were able to reach overall sensitivity of 93.4% (95% CI: 90.3-95.7%) in the detection of ABs in comparison to their GT number. Compared to standard LM diagnostic data, our model showed equal to or higher sensitivity in most cases.Our results indicate that deep learning neural network technology offers promising diagnostic tools for routine assessment of BALs. However, at this stage, a human expert is required to confirm the findings.
OSBP-related protein 8 (ORP8) encoded by Osbpl8 is an endoplasmic reticulum sterol sensor implicated in cellular lipid metabolism. We generated an Osbpl8−/− (KO) C57Bl/6 mouse strain. Wild-type and Osbpl8KO animals at the age of 13-weeks were fed for 5 weeks either chow or high-fat diet, and their plasma lipids/lipoproteins and hepatic lipids were analyzed. The chow-fed Osbpl8KO male mice showed a marked elevation of high-density lipoprotein (HDL) cholesterol (+79%) and phospholipids (+35%), while only minor increase of apolipoprotein A-I (apoA-I) was detected. In chow-fed female KO mice a less prominent increase of HDL cholesterol (+27%) was observed, while on western diet the HDL increment was prominent in both genders. The HDL increase was accompanied by an elevated level of HDL-associated apolipoprotein E in male, but not female KO animals. No differences between genotypes were observed in lecithin:cholesterol acyltransferase (LCAT) or hepatic lipase (HL) activity, or in the fractional catabolic rate of fluorescently labeled mouse HDL injected in chow-diet fed animals. The Osbpl8KO mice of both genders displayed reduced phospholipid transfer protein (PLTP) activity, but only on chow diet. These findings are consistent with a model in which Osbpl8 deficiency results in altered biosynthesis of HDL. Consistent with this hypothesis, ORP8 depleted mouse hepatocytes secreted an increased amount of nascent HDL into the culture medium. In addition to the HDL phenotype, distinct gender-specific alterations in lipid metabolism were detected: Female KO animals on chow diet showed reduced lipoprotein lipase (LPL) activity and increased plasma triglycerides, while the male KO mice displayed elevated plasma cholesterol biosynthetic markers cholestenol, desmosterol, and lathosterol. Moreover, modest gender-specific alterations in the hepatic expression of lipid homeostatic genes were observed. In conclusion, we report the first viable OsbplKO mouse model, demonstrating a HDL elevating effect of Osbpl8 knock-out and additional gender- and/or diet-dependent impacts on lipid metabolism.
