Skip main navigation

Circulating Extracellular Vesicle-Propagated microRNA Signature as a Vascular Calcification Factor in Chronic Kidney Disease

Originally publishedhttps://doi.org/10.1161/CIRCRESAHA.122.321939Circulation Research. 2023;132:415–431

Abstract

Background:

Chronic kidney disease (CKD) accelerates vascular calcification via phenotypic switching of vascular smooth muscle cells (VSMCs). We investigated the roles of circulating small extracellular vesicles (sEVs) between the kidneys and VSMCs and uncovered relevant sEV-propagated microRNAs (miRNAs) and their biological signaling pathways.

Methods and Results:

We established CKD models in rats and mice by adenine-induced tubulointerstitial fibrosis. Cultures of A10 embryonic rat VSMCs showed increased calcification and transcription of osterix (Sp7), osteocalcin (Bglap), and osteopontin (Spp1) when treated with rat CKD serum. sEVs, but not sEV-depleted serum, accelerated calcification in VSMCs. Intraperitoneal administration of a neutral sphingomyelinase and biogenesis/release inhibitor of sEVs, GW4869 (2.5 mg/kg per 2 days), inhibited thoracic aortic calcification in CKD mice under a high-phosphorus diet. GW4869 induced a nearly full recovery of calcification and transcription of osteogenic marker genes. In CKD, the miRNA transcriptome of sEVs revealed a depletion of 4 miRNAs, miR-16-5p, miR-17~92 cluster-originated miR-17-5p/miR-20a-5p, and miR-106b-5p. Their expression decreased in sEVs from CKD patients as kidney function deteriorated. Transfection of VSMCs with each miRNA-mimic mitigated calcification. In silico analyses revealed VEGFA (vascular endothelial growth factor A) as a convergent target of these miRNAs. We found a 16-fold increase in VEGFA transcription in the thoracic aorta of CKD mice under a high-phosphorus diet, which GW4869 reversed. Inhibition of VEGFA-VEGFR2 signaling with sorafenib, fruquintinib, sunitinib, or VEGFR2-targeted siRNA mitigated calcification in VSMCs. Orally administered fruquintinib (2.5 mg/kg per day) for 4 weeks suppressed the transcription of osteogenic marker genes in the mouse aorta. The area under the curve of miR-16-5p, miR-17-5p, 20a-5p, and miR-106b-5p for the prediction of abdominal aortic calcification was 0.7630, 0.7704, 0.7407, and 0.7704, respectively.

Conclusions:

The miRNA transcriptomic signature of circulating sEVs uncovered their pathologic role, devoid of the calcification-protective miRNAs that target VEGFA signaling in CKD-driven vascular calcification. These sEV-propagated miRNAs are potential biomarkers and therapeutic targets for vascular calcification.

Novelty and Significance

What Is Known?

  • Chronic kidney disease (CKD) accelerates vascular calcification (VC) in part by promoting the phenotypic switching of vascular smooth muscle cells (VSMCs) and the VSMC-driven formation of extracellular calcium phosphate crystals in vessel walls.

  • Circulating small extracellular vesicles (sEVs) carry and propagate signaling molecules, including proteins, microRNAs (miRNAs), and DNA between cells. However, the composition, biological functions, and pathophysiological roles of sEVs in CKD-associated VC remain unclear.

What New Information Does This Article Contribute?

  • The miRNA transcriptomic signature of CKD-derived circulating sEVs uncovered a decrease in the calcification-protective miRNAs that target VEGFA signaling in CKD-driven VC.

  • sEV-derived miRNAs are strong candidates for biomarkers and therapeutic targets for VC.

  • Therapeutically blocking the VEGFA-VEGFR2 via inhibitors such as sunitinib, sorafenib, and fruquintinib may ameliorate vascular calcification in CKD.

In this study, we showed the following: (1) CKD-derived sEVs facilitate the phenotypic switching of VSMCs and calcification both in VSMCs in vitro and the mouse aorta in vivo; (2) the levels of miR-16-5p, miR-17-5p, miR-20a-5p, and miR-106b-5p, 4 miRNAs were decreased from both the CKD rodent and human sEVs; (3) this reduction in miRNA transcriptomic signatures was linked to an increased protein level of VEGFA; and (4) pharmacological and genetic inhibition of VEGFA ameliorated VC. Overall findings suggest that the VC and calcium phosphate crystallization processes are controlled by VEGFA signaling, which is activated by CKD-derived sEVs.

In This Issue, see p 397

Meet the First Author, see p 398

Over the past few decades, the global population affected by cardiovascular disease (CVD) has increased. CVD is associated with premature death, morbidity, and a high economic burden.1,2 It remains a leading cause of death for most ethnic groups. An estimated 17.9 million people died from CVD in 2019, accounting for 32% of global deaths.3 Vascular calcification (VC) plays a critical role in CVD pathogenesis and is characterized by calcium phosphate crystal formation in blood vessels.4 Particularly, the phenotypic switching of vascular smooth muscle cells (VSMCs), that is, their osteogenic differentiation into noncontractile VSMCs, is essential for extracellular matrix deposition and mineralization of the matrix.5 However, the pathogenesis and key signaling molecules involved in this process are not fully understood. Furthermore, vascular or valvular calcification is usually progressive and irreversible despite recent advances in antihypertensive, anticoagulant, or antilipidemic agents.4,5 Therefore, a global challenge is to noninvasively predict and treat phenotypic switching of VSMCs and VC.

Chronic kidney disease (CKD) is a major public health problem, affecting 700 million people and nearly one-third of the elderly.6 CKD is defined as a persistent glomerular filtration rate <60 mL/min per 1.73 meter squared or a urinary albumin to creatinine ratio ≥30 mg/g for 3 months or longer. A lower glomerular filtration rate is associated with a higher risk of CVD. A substantial proportion of CKD patients suffer from CVD or CVD-related death before reaching dialysis-dependent end-stage kidney disease.7,8 VC is a typical pathologic feature underlying cardiovascular events in CKD. Aortic or valvular calcification, rather than atherosclerosis, is a strong predictor of future cardiovascular events in CKD patients.9 An imbalance in calcium (Ca) and phosphate (Pi) metabolism with Klotho and fibroblast growth factor 23 (FGF23), called CKD-mineral bone disorder, is an important pathogenesis underlying VC.8,10 Treating CKD-mineral bone disorder is an essential treatment strategy for VC, yet it does not completely prevent or resolve VC despite emerging therapeutic options of oral phosphate binders or calcimimetics.10–12 Furthermore, little is known about the biological signaling facilitating interorgan communication between the kidneys and remote VSMCs.

Small extracellular vesicles (sEVs) are membrane-bound vesicles of endocytic origin naturally secreted by all cells at concentrations of >109 vesicles/mL.13,14 Emerging studies have reported the potential applications of sEVs as carriers in drug-delivery systems, particularly those originating from stem cells.13–15 Circulating sEVs carry and propagate signaling molecules, including proteins, small RNAs, and DNA between cells. However, the composition, biological functions, and pathophysiological role of sEVs in CKD remain unclear. We hypothesized that circulating sEVs propagate uncharacterized biological signals and play an essential role in VC in CKD. Based on the initial investigation in this study, sEVs in a CKD rodent model, but not sEV-depleted serum, promote pathogenic calcium deposition in VSMCs.

This study identified a transcriptomic signature of microRNAs (miRNAs) propagated in circulating sEVs from CKD model rodents and humans and found that a deficiency of 4 miRNAs is essential for the VC underlying CKD. In silico target analyses revealed VEGFA (vascular endothelial growth factor A)-VEGFR2 signaling as a convergent target of the miRNA quartet and identified these miRNAs as potential biomarkers for prediction and therapeutic targets for VC.

Methods

Data Availability

Detailed methods are provided in the Supplemental Materials. All animal studies complied with the guidelines for animal research of Tokyo Medical and Dental University. The Animal Care and Use Committee of Tokyo Medical and Dental University approved the study protocol. The ethics committee also approved this study at all participating institutions, and this research complies with the ethical principles of the Declaration of Helsinki. All participants provided written consent. Please see the Major Resources Table in the Supplemental Material.

Results

Circulating sEVs From CKD Rodent Models Facilitate VC in Aortic VSMCs

To examine the effect of CKD serum on VC, we evaluated the formation of calcium phosphate crystals in the A10 clonal embryonic rat aortic smooth muscle cell line (A10 cells; ATCC)16 under treatment with the serum of CKD model rats. We established this CKD model by giving rats a diet containing 0.75% adenine for 4 weeks17 and collected the serum after euthanizing the rats (Figure 1A). Masson trichrome staining of rat kidneys showed global fibrosis in the kidney interstitium of CKD rats (Figure 1B). Additionally, the urea nitrogen and creatinine levels indicated elevated uremic solutes in CKD serum (Figure 1C). Calcium staining with Alizarin red18 was performed 48 hours,19 after cell treatment with calcifying medium (CM) containing 4.0 mM Pi20 and serum derived from control or CKD serum (Figure 1A). Cell variability was not affected by this induction of calcification at 24 and 48 hours (Figure S1C). As shown in Figure 1D and 1E, calcified deposits and calcium levels increased in the cells treated with CKD serum. Furthermore, transcription of the osteogenic marker genes OSX (osterix), OCN (osteocalcin), OPN (osteopontin), and MGP (matrix gla protein) was upregulated in the CKD group (Figure 1F). The spontaneous precipitation of Ca and Pi in the medium was unlikely, given the absence of Alizarin red staining 1 hour post culture in CM (Figure S1D). Although serum Pi level was higher in CKD rats compared to control rats (Figure S1A), neither control nor CKD serum led to calcifiying deposits at 48 hours after the culture without CM (Figure S1E).

Figure 1.

Figure 1. Circulating sEVs from CKD rodent models facilitate vascular calcification in aortic VSMCs. A, Establishment of a CKD rodent model by giving wild-type Wister rats a diet containing 0.75% adenine for 4 weeks, and an outline of calcification assay using A10 clonal embryonic rat aortic smooth muscle cells. Calcification of VSMCs was assayed with Alizarin red staining 48 hours after culture in CM containing 5% rodent serum and 4 mM-inorganic phosphate. B, Masson Trichrome staining showed tubulointerstitial fibrosis and extensive tubular dilation in the kidneys of adenine-treated rats. C, Serum urea nitrogen and creatinine indicating uremic solutes were elevated in CKD serum (n=8 per group). D, Calcium staining with Alizarin red was performed 48 hours, after cell treatment with CM containing 4.0 mM Pi and serum derived from control (Ctrl) or CKD serum. E, Calcium contents after the lysis of calcified deposits were increased in the cells treated with CKD serum (n=6 per group). F, transcription of the osteogenic marker genes OSX (osterix), OCN (osteocalcin), osteopontin (OPN), and MGP (matrix gla protein) was upregulated in the CKD group (n=4 in the OSX and OCN transcription groups; n=3 in the OPN and MGP groups). G, Polymer-based purification of circulating was primarily used for subsequent functional assays of sEVs and sEVs-depleted serum. H, The representative size distributions of sEVs and electron microscopy of a single sEV from control or CKD rats. I, Particle counts of circulating sEVs were not different between control and CKD rats (n=11 in the Ctrl group and n=12 in the CKD group). J, Immunoblots for tetraspanins CD9, CD63, and CD81 and β-actin that were contained by the sEVs and were completely depleted after the polymer-based purification. CKD did not alter the expression of CD9 and CD63 in sEVs (n=8 per group). K, Confocal microscopy illustrated that the control and CKD sEVs labeled with 1 μmol/L of the fluorescent lipophilic tracer DiR were distributed in the cytoplasm of A10 cells 3 hours after administration to the medium. L and M, Calcification assays in A10 cells were performed using isolated sEVs and sEV-depleted serum collected from control or CKD rats. The sEVs of CKD rats, but not control sEVs, increased the calcified deposits (M) and calcium contents (N) (n=14, 14, 14, and 13 in each group, respectively). The Shapiro–Wilk test was used to test the normality of variables. Unpaired t tests were used for the comparison of variables with normal distribution and homogeneity of variance. For nonparametric variables or groups with n<6, we used Mann–Whitney U tests for 2-group comparisons and Wilcoxon tests for multiple comparisons. All data were presented as mean±SD. CKD indicates chronic kidney disease; CM, calcifying medium; Pi, phosphate; sEVs, small extracellular vesicles; and VSMCs, vascular smooth muscle cells.

Next, to clarify whether sEVs propagate biological signaling and promote VC in CKD, we performed calcification assays using isolated sEVs and sEV-depleted serum collected from control or CKD rats (Figure 1G and Figure S1A). Polymer-based purification was primarily used for these experiments to achieve >90% recovery of sEVs, subsequent functional assays of sEVs and sEVs-depleted serum, and translation to human diagnostic workups.14,21,22 First, we characterized similar size distributions between control rat sEVs and CKD sEVs, showing a mean diameter of 85 nm and 90 nm, respectively (Figure 1H). We also demonstrated electron microscopy of a single sEV from control or CKD rats, respectively. Particle counts of circulating sEVs were not different between control and CKD rats (Figure 1I). Following the depletion of sEVs, the serum completely lacked the tetraspanins CD9, CD63, and CD81 and β-actin that were initially contained by the sEVs, showing a nearly complete depletion of sEVs (Figure 1J). CKD did not alter the expression of CD9 and CD63 in sEVs. Confocal microscopy illustrated that the control sEVs labeled with 1 μmol/L of the fluorescent lipophilic tracer DiR were distributed in the cytoplasm of A10 cells 3 hours after administration to the medium (Figure 1K). As shown in Figure 1L and 1M, the sEVs of CKD rats, but not control sEVs, increased calcified deposits and calcium levels. These findings suggest that CKD sEVs propagate undetermined cargos and promote the differentiation of VSMCs and VC in vitro. The Ca and Pi levels in sEVs were not different between control and CKD rats and were much lower than those in serum (Figure S1B). Ca deposition by sEVs themselves was less likely due to the lack of formation after short-term exposure to CM and serum (Figure S1D).

Inhibition of Systemic Biogenesis of sEVs With GW4869 Ameliorated Aortic VC in CKD Model Mice

To verify if sEVs play an essential role in interorgan communications linking CKD and VC in vivo, we evaluated the effects of GW4869, a systemic inhibitor of sEV biogenesis,23,24 on aortic VC. CKD C57BL/6JJcl mice fed with adenine and a high-Pi–containing diet received an intraperitoneal injection of 2.5 mg/kg per 2 days of GW486924 or its vehicle (7.5% DMSO in saline) for 4 months (Figure 2A). Masson trichrome staining showed that global fibrosis in the kidney interstitium similarly affected mice receiving the vehicle or GW4869 (Figure 2B). Figure 2C and Figure S2A illustrate biochemical data of the model mice. The level of serum creatinine and the kidney fibrotic marker α-SMA increased in CKD mice with and without GW4869 (Figure 2C and 2D). We measured the particle counts and size distributions of sEVs using Nanosight (Figure 2E). Circulating sEVs decreased in CKD mice treated with GW4869 to about half of those in the vehicle-treated mice (Figure 2F). Imaging with Alizarin red staining revealed that CKD and a high-Pi diet induced Ca deposition in the mouse thoracic and abdominal aorta when treated with the vehicle but with GW4869 (Figure 2G). To quantify the calcification, we also performed Alizarin staining of thoracic aortic sections of mice (Figure 2H). The calcified area increased by CKD and high-Pi diet compared to mice under the standard diet, whereas the area was markedly decreased by GW4869 treatment (Figure 2I). Consistently, the Ca content in the thoracic aorta was increased under CKD and a high-Pi diet when treated with the vehicle, and GW4869 reversed this increase (Figure 2J). The mRNA expression levels of the osteogenic marker genes Mgp, Sparc, and Ogn markedly increased in the thoracic aorta of the vehicle-treated CKD mice. However, GW4869 administration almost entirely inhibited these increases in transcription (Figure 2K).

Figure 2.

Figure 2. Inhibition of systemic biogenesis of small extracellular vesicles (sEVs) with GW4869 mitigated aortic vascular calcification in CKD model mice. A, We evaluated the effects of GW4869, a systemic inhibitor of sEV biogenesis, on aortic vascular calcification in mice. CKD C57BL/6JJcl mice fed with adenine and a high phosphate (Pi) containing diet received an intraperitoneal injection of 2.5 mg/kg per 2 days of GW4869 or its vehicle (7.5% DMSO in saline) for 16 weeks. B, Masson trichrome staining showed that global fibrosis in the kidney interstitium similarly affected mice receiving the vehicle or GW4869. C and D, Serum creatinine levels (C), and kidney fibrotic marker α-SMA (D) increased in CKD mice with and without GW4869 (n=6 per group). E, The representative size distributions of sEVs using Nanosight and microscopy of a single sEV in mice. sEVs were purified using polymer-based precipitation (Figure 1G). F, Circulating sEVs decreased in CKD mice treated with GW4869 to about half of those in the vehicle-treated mice (n=6, 9, and 7 in each group, respectively). G, Imaging with Alizarin red staining revealed that CKD and a high-Pi diet induced calcium (Ca) deposition in the mouse thoracic and abdominal aorta when treated with the vehicle but with GW4869. H, Alizarin staining of thoracic aortic sections of mice. I, The calcified area increased by CKD and high-Pi diet compared to mice under the standard diet, whereas the area was markedly decreased by GW4869 treatment (n=6 per group). J, The Ca content relative to dry mass of the thoracic aorta was increased under CKD and a high-Pi diet when treated with the vehicle, and GW4869 reversed this increase (n=6 per group). K, The mRNA expression levels of the osteogenic marker genes Mgp, Sparc, and Ogn markedly increased in the thoracic aorta of the vehicle-treated CKD mice (n=6 per group). GW4869 administration almost entirely inhibited these increases in transcription. The Shapiro–Wilk test was used to test the normality of variables. For the comparison of variables with normal distributions and homogeneity of variance, unpaired t tests were used for 2 groups and 1-way ANOVA followed by Turkey post hoc test was used for >2 groups. For the nonparametric variables, we used nonparametric analyses; Mann–Whitney U tests were used for 2 groups and Wilcoxon tests for multiple comparisons. All data were presented as mean±SD. Ca indicates calcium; CKD, chronic kidney disease; and Pi, phosphate.

Uncovering the miRNA Signature of sEVs and Functional miRNAs Related to VC in CKD Rodents and Humans

To investigate the miRNA transcriptomic profile of peripheral circulating sEVs in CKD, we performed a comprehensive miRNA transcriptome analysis of sEVs collected from CKD model rats using the Affymetrix GeneChip miRNA 4.0 (Affymetrix, Santa Clara, CA).25,26Figure 3A displays the expression changes of each miRNA in CKD versus control rats. The horizontal line indicates that the fold change (FC) threshold of miRNA expression is ≥1.5, and the vertical line indicates that the threshold P-value of the t test is 0.1. Blue spots represent 8 downregulated miRNAs (rno-miR-16-5p, rno-miR-17-5p, rno-miR-20a-5p, rno-miR-106b-5p, rno-miR-107-3p, rno-let-7a-5p, rno-let-7c-5p, and rno-let-7d-5p) and the red spot represents the single-upregulated miRNA (rno-miR-206-3p). Downregulation of 4 miRNAs, miR-16-5p (FC, 0.18), miR-17-5p (FC, 0.51), miR-20a-5p (FC, 0.33), and miR-106b-5p (FC, 0.45), was confirmed with quantitative RT-PCR relative to miR-139-5p. We also performed miRNA profiling in the sEVs of rats following its purification via differential ultracentrifugation (Figure S3A and S3B).27 We found that the sEVs isolated with this method from rats with CKD lacked rno-miR-16-5p, rno-miR-17-5p, rno-miR-20a-5p, rno-miR-106b-5p, and rno-miR-107-3p (Figure S3C). The decrease in these miRNAs was replicated in the sEVs of CKD mice (Figure S2B and S2C).28 The expression levels of the internal control miRNAs miR-139-5p and miR-423-3p were not altered between control and CKD rats or mice (Figure 3B and Figure S2C). The sequences of these miRNAs are evolutionally conserved across mammals (Figure 3C and Table S1). We also detected a decrease in miR-16-5p, miR-17-5p, miR-20a-5p, and miR-106b-5p in CKD male and female mice receiving a diet containing 2.0% Pi for 4 weeks (Figure S2D through S2G).

Figure 3.

Figure 3. Uncovering the miRNA signature of small extracellular vesicles (sEVs) and functional miRNAs related to vascular calcification in CKD rodents and humans. A, The volcano plot graph of the miRNA transcriptomic profiling compares the peripheral circulating sEVs of the control and CKD rats (n=4 per group; Figure 1A). The expression levels of miRNA calculated by log2 transformation of the normalized data were compared using unpaired t tests. The horizontal line indicates that the fold change (FC) threshold of miRNA expression is ≥1.5, and the vertical line indicates that the threshold of P-value of the t test is 0.1. For the screening of meaningful functional miRNAs, the differentially expressed miRNAs were defined as those with P-values <0.1 and FC of ≥1.5. Red spots represent upregulated miRNAs, and blue spots represent downregulated miRNAs. B, The decreases in rno-miR-16-5p, rno-miR-17-5p, rno-miR-20a-5p, and rno-miR-106b-5p were validated with quantitative RT-PCR relative to rno-miR-139-5p (n=8 per group). The expression level of internal control miRNAs rno-miR-423-3p was not altered by CKD. C, The sequences of miR-16-5p, miR-17-5p, miR-20a-5p, and miR-106b-5p are evolutionally conserved across mammals. D, A Manhattan plot of the locations of the sEV-included miRNAs across the genome. The vertical axis indicates the FC in miRNA expression, and the horizontal axis indicates the position on each chromosomes. The cooperatively downregulated miR-17-5p and miR-20a-5p originate from the miR-17/92 cluster. E, miRNA expression levels relative to the miR-139-5p expression in 37 CKD patients. sEVs were purified using polymer-based precipitation (Figure 1G). Each miRNA expression level according to CKD stage (stage G5, eGFR <15.0 [n=8]; stage G4, eGFR 15.0–29.9 [n=11]; stage G3, eGFR 30.0–59.9 [n=13]; stage G1/2, eGFR ≥60.0 [n=5]) showing that CKD progression was significantly associated with the loss of miR-16-5p, miR-17-5p, miR-20a-5p, and miR-106b-5p expression in sEVs. Grubbs test was used to detect outliers, and Spearman correlation coefficient was used to identify correlations between nonparametric variables. P-values <0.01 were considered statistically significant. F, Multivariable linear regression models adjusting for age and sex revealed that a one quartile decrease in the expression of hsa-miR-16-5p, hsa-miR-17-5p, hsa-miR-20a-5p, and hsa-miR-106b-5 was associated with an eGFR decline, respectively. G, Alizarin red staining 48 hours after a culture in a CM containing 5% Ctrl or CKD serum. microRNA mimics were transfected 6 hours before culture in a CM. H, Each mimic of miR-16-5p, miR-17-5p, miR-20a-5p, and miR-106b-5p ameliorated calcified deposits (n=6, 6, 7, 4, 7, and 7 for each group, respectively). The Shapiro–Wilk test was used to test the normality of variables. For the comparison of variables with normal distribution and homogeneity of variance, unpaired t tests were used. For nonparametric variables or groups with n<6, we used Mann–Whitney U tests for 2-group comparisons and Wilcoxon tests for multiple comparisons. All data were presented as mean±SD. CKD indicates chronic kidney disease; CM, calcifying medium; eGFR, estimated glomerular filtration rate; and miRNA, microRNA.

We visualized the association between miRNA clusters and CKD with a Manhattan plot of the locations of the sEV-included miRNAs across the genome (Figure 3D). The vertical axis indicates the FC in miRNA expression, and the horizontal axis indicates the position on each chromosome. The cooperatively downregulated miR-17-5p and miR-20a-5p originate from the miR-17/92 cluster and share highly homologous sequences (Figure 3C). The miR-15/16 cluster-originated miR-15b-5p (FC, 0.43) and the miR-106b/25 cluster-originated miR-93-5p (FC, 0.49) were relatively low in CKD rats. However, the differences between control and CKD rats were not statistically significant in the miRNA transcriptome analysis.

We assessed if humans with CKD were deficient in hsa-miR-16-5p, hsa-miR-17-5p, hsa-miR-20a-5p, and hsa-miR-106b-5p expression in sEVs. For this purpose, we used 2 CKD cohorts consisting29 of 37 dialysis-independent patients (27% female; average age 71 years [interquartile range, 61–79 years]; Table S2). The participants’ age, body mass index, and proportions of hypertension, diabetes, CVD, calcium channel blockers use, and renin–angiotensin system inhibitors use were not different according to the CKD stage (stage G5, estimated glomerular filtration rate [eGFR] <15.0; stage G4, eGFR 15.0–29.9; stage G3, eGFR 30.0–59.9; stage G1/2, eGFR ≥60.0). Then, we examined miRNA expression levels relative to miR-139-5p expression. The median eGFR assessing renal function was 28.1 (19.5–51.5). Figure 3E shows each miRNA expression level according to the CKD stage, showing that CKD progression was significantly associated with the loss of miR-16-5p, miR-17-5p, miR-20a-5p, and miR-106b-5p expression in sEVs (P-value <0.005 for trend for all). Moreover, when analyzing the association between quartiles of miRNA expression and eGFR, the lower expression of each miRNA was correlated with lower eGFR (Figure S6A). Additionally, we found that a one quartile decrease in the expression of hsa-miR-16-5p, hsa-miR-17-5p, hsa-miR-20a-5p, and hsa-miR-106b-5 was associated with an eGFR decline of 9.48 (95% CI, 3.75–15.22; P=0.002), 9.93 (95% CI, 3.92–15.94; P=0.002), 10.80 (95% CI, 4.89–16.71; P=0.001), and 10.26 (95% CI, 3.97–16.55; P=0.002), respectively, in multivariable linear regression models adjusting for age and sex (Figure 3F). As a negative control, miR-423-3p was not associated with eGFR (β, 3.32; [95% CI, 3.32–9.97]; P=0.3).

To further evaluate the functions of these miRNAs in VC, we transfected each miRNA mimic into A10 cells treated with CM containing CKD serum. As shown in Figure 3G and 3H, all miRNA mimics suppressed mineralization in VSMCs. Mineralization induced by CKD serum and CM were not suppressed by the miR-423-3p and miR-455-5p mimics that were selected as potential negative controls from the miRNA transcriptome (FC, 0.85; P=0.1; and FC, 1.03; P=0.8 versus the control group, respectively; Figure 3A; Figure S4D and S4E). We also examined the effect of miRNA antagomirs against miR-16-5p, miR-17-5p, miR-20a-5p, and miR-106b-5p on mineralization. As shown in Figure S4G and S4H, all 4 miRNA inhibitors exacerbated calcification. In particular, the miR-16-5p inhibitor showed the largest effect on increased Ca deposition.

In Silico and Experimental Analyses Identified VEGFA as a Common Target of miRNAs Depleted in Circulating sEVs in CKD

To determine the critical targets of miRNAs depleted in the circulating sEVs of CKD, we performed in silico analyses of the target genes of these miRNAs and created molecular interaction networks. We first analyzed the KEGG molecular pathways associated with miRNA targets using the DIANA-miRPath v.3 platform.30Figure 4A shows the top 15 molecular networks. We found that the transforming growth factor-β signaling pathway promotes VC. Other significant pathways included the cell cycle and the Hippo signaling pathway, which regulate cell proliferation, differentiation, and apoptosis.

Figure 4.

Figure 4. In silico and experimental-based analyses identified VEGFA as a common target of miRNAs depleted in circulating sEVs in CKD. A, KEGG analysis revealed the top 15 molecular networks targeted by hsa-miR-16-5p, hsa-miR-17-5p, has-miR-20a-5p, and hsa-miR-106b-5p, using the DIANA-miRPath v.3 platform. B, Specific target molecules by 2 or more of hsa-miR-16-5p, hsa-miR-17-5p, has-miR-20a-5p, and hsa-miR-106b-5p, using MiRTargetLink. There were 3 molecules targeted by the microRNA quartet with strong sequence evidence supported by experimental validation in the software, including VEGFA. C, The predicted consequential pairing of the target region and each miRNA. The target motifs of miRNAs in the VEGFA 3′ untranslated region (3′-UTR) are evolutionally conserved across mammals. D, VEGFA mRNA levels were quantified with RT-PCR 24 hours after transfection of each mimic of miR-16-5p, miR-17-5p, miR-20a-5p, miR-106b-5p (n=4 per group), or negative control (NC; n=7) in A10 cells. E, All miRNA mimics, especially that of miR-16-5p, reduced the VEGFA protein expression level (n=4 per group). F, In the thoracic aorta of rats, the protein expression level of VEGFA was elevated in the CKD rat aorta (n=6 per group; Figure 1A). G, We used a model of CKD-induced vascular calcification in mice (Figure 2A). Chow containing 0.25% adenine and 2.0% phosphorus was given for 4 weeks, and chow containing 0.10% adenine and 2.0% phosphorus was subsequently given to mice for 12 weeks. The mice with CKD and a high-Pi diet treated with vehicle had a 16-fold increase in Vegfa transcription. GW4869, an inhibitor of sEVs, reversed the Vegfa transcription change (n=6, 6, and 5 in each group, respectively). H, GW4869 elevated miR-16-5p, miR-17-5p, miR-20a-5p, and miR-106b-5p expression levels in sEVs compared to vehicle treatment (n=5/group). I and J, Recombinant rat VEGFA dose-dependently facilitated mineralization (I) and calcium deposition (J) in VSMCs (n=6/group). K, VEGFA increased the phosphorylation of RUNX2, one of the key transcription factors that regulate the osteogenic differentiation of VSMCs, at S451 (n=4/group). L, Administration of 5 μg/mL VEGFA upregulated the mRNA expression levels of OSX (osterix), OCN (osteocalcin), and OPN (osteopontin), which are predominantly regulated by RUNX2 (n=4/group). The Shapiro–Wilk test was used to test the normality of variables. For the comparison of variables with normal distribution and homogeneity of variance, unpaired t tests were used for 2 groups and 1-way ANOVA followed by Turkey post hoc test for comparison of >2 groups. For nonparametric variables or groups with n<6 and, we used Mann–Whitney U tests for 2-group comparisons and Wilcoxon tests for multiple comparisons. All data were presented as mean±SD. CKD indicates chronic kidney disease; CM, calcifying medium; sEVs, small extracellular vesicles; miRNA, microRNA; and VSMCs, vascular smooth muscle cells.

Using MiRTargetLink,31,32 we specifically analyzed and visualized the target genes of hsa-miR-16-5p, hsa-miR-17-5p, has-miR-20a-5p, and hsa-miR-106b-5p, which show strong evidence based on sequences and supported by experimental validation.33,34 Given that all miRNAs are potentially protective for VC (Figure 3G and 3H), we mainly focused on genes commonly targeted by multiple miRNAs (Figure 4B), and we found VEGFA as one target common to the miRNA quartet. Figure 4C shows the predicted pairing of the target region and each miRNA. The target motifs of miRNA binding sites in the VEGFA 3′ untranslated region are evolutionally conserved across mammals.

Next, we confirmed that transfection of miR-16-5p, miR-17-5p, miR-20a-5p, and miR-106b-5p mimics decreased the transcription of VEGFA in A10 cells (Figure 4D), showing consistent findings with previous reports.33,34 All miRNA mimics, especially that of miR-16-5p, reduced the VEGFA protein expression level (Figure 4E). Also, we examined the mimic efficiency on mineralization and VEGFA mRNA expression when transfecting all 4 miRNA mimics. As shown in Figure S4A through S4C, transfection of the 4 miRNA mimics suppressed mineralization and VEGFA mRNA expression. However, these effects were not greater than the single transfection of each miRNA, suggesting that there is no synergetic effect between these miRNAs in this model. Among miRNA antagomirs against the 4 miRNAs, the miR-16-5p inhibitor increased VEGFA mRNA expression, consistent with the finding that this inhibitor showed the greatest impact on mineralization (Figure S4I). To clarify if the 4 miRNAs regulate and depend on each other, we quantified the miRNA expression levels 24 hours after transfection of each miRNA mimic in A10 cells. Transfection of miR-17-5p mimic upregulated miR-16-5p expression, miR-106b-5p mimic upregulated miR-17-5p-expression, and miR-16-5p -and miR-17-5p mimics upregulated miR-106b-5p (Figure S4J).

In the thoracic aorta of rats, the protein expression level of VEGFA was elevated in the CKD model (Figure 4F). When measured in the CKD-induced calcification murine model (Figure 2A), the mice with CKD and a high-Pi diet treated with vehicle had a 16-fold increase in Vegfa transcription. GW4869, an inhibitor of sEVs,23,24 reversed the Vegfa transcription change (Figure 4G). We also found that GW4869 elevated miR-16-5p, miR-17-5p, miR-20a-5p, and miR-106b-5p expression levels in sEVs compared to vehicle treatment (Figure 4H), which might lead to the suppression of the VEGFA mRNA level in the mouse aorta. The serum VEGFA level was higher in mice with CKD and a high-Pi diet treated with vehicle than those with non-CKD/nonhigh-Pi diet treated with vehicle (Figure S4K). GW4869 reversed this increase. In mice without CKD and a high-Pi diet, GW4869 did not influence Vegfa mRNA expression compared to vehicle treatment (Figure S4L). These findings indicate that miR-16-5p, miR-17-5p, miR-20a-5p, and miR-106b-5p propagated in healthy sEVs are essential for protecting against VC, and CKD pathologically suppresses this safeguard mechanism in mice.

VEGFA-VEGFR2 Signaling Drives VC and its Inhibition Mitigates VC

To elucidate if VEGFA signaling targeted by miRNAs is involved in the pathophysiology of and displays therapeutic potential in VC, we evaluated Ca deposition under treatment with recombinant rat VEGFA in in vitro cultures of VSMCs. Recombinant rat VEGFA dose-dependently facilitated mineralization in VSMCs (Figure 4I and 4J). We also found that VEGFA increased the phosphorylation of RUNX2, one of the key transcription factors that regulate the osteogenic differentiation of VSMCs, at S451. This phosphorylation site was previously shown to upregulate the transcriptional activity of RUNX2.35 Administration of VEGFA upregulated the mRNA expression levels of OSX, OCN, and OPN, which are predominantly regulated by RUNX2. Thus, these findings suggest that VEGFA facilitates the osteogenic differentiation of VSMCs by regulating the phosphorylation and transcriptional activity of RUNX2.

We also evaluated a therapeutic potential of the VEGFR1/2 inhibitors sorafenib, fruquintinib, or sunitinib in VSMCs. Calcification of A10 cells was dose-dependently inhibited by sorafenib, fruquintinib, and sunitinib (Figure 5A). Additionally, Ca levels were dose-dependently suppressed, particularly by fruquintinib, which is a more selective blocker of VEGFR1/2 in addition to PDGFR, c-KIT, and Flt-3 (Figure 5B). Cell viability was not affected by fruquintinib with the doses used (Figure S5A). As shown in Figure 5C, the mRNA expression of OSX and OCN was inhibited with fruquintinib at 48 hours after treatment with CM. Next, we validated if VEGFR1/2 inhibition suppresses the transcription of these osteogenic marker genes in the aorta of CKD mice under a high-Pi diet (Figure S5D). We found that a 4-week treatment with 2.5 mg/kg per day fruquintinib36 suppressed phosphorylation of VEGFR2 in the aorta (Figure 5D) and decreased the mRNA expression levels of OSX, OCN, and OPN in the thoracic aorta of CKD mice (Figure 5E).

Figure 5.

Figure 5. Inhibition of VEGFA-VEGFR2 signaling mitigated vascular calcification in aortic VSMCs. A, Calcification of VSMCs was assayed with Alizarin red staining 48 hours after a culture in a CM containing 5% Ctrl or CKD serum and 16–400 nM sorafenib, fruquintinib, or sunitinib. B, Calcium contents were particularly ameliorated with fruquintinib, a more selective blocker of VEGFR1/2 (n=5, 6, 7, 3, 4, 4, 3, 4, 4, 3, and 3 for each group, respectively). P values <0.05 were considered statistically significant versus the group treated with CM and CKD serum without VEGFR inhibitors. C, mRNA expression levels of OSX and OCN were inhibited with fruquintinib (FQT; n=4/group). D, A 4-week treatment with 2.5 mg/kg per day fruquintinib suppressed phosphorylation of VEGFR2 in the aorta (n=6 in the vehicle group and n=7 in the FQT group). E, The mRNA expression levels of OSX, OCN, and OPN were decreased with FQT in the thoracic aorta of mice (n=6, 6, and 7 in the 3 groups, respectively). F, Immunoblots for silencing efficiencies of VEGFR1, VEGFR 2, or both with siRNA (n=3/group). G and H, Alizarin red staining 48 hours after a culture in a CM containing 5% CKD serum. Silencing VEGFR1, VEGFR 2, or both with siRNA was performed 6 hours before culture in a CM. siVEGFR2 ameliorated mineralization (G) and calcified deposits (H) compared to the scrambled siRNA group. A combination of siVEGFR1 and siVEGFR2 did not further suppress calcification (n=9/group). I, The single siVEGFR2 and the combination of siVEGFR1 and siVEGFR2 suppressed the mRNA expression levels of OSX and OCN (n=6/group). The Shapiro–Wilk test was used to test the normality of variables. For comparisons of variables with normal distributions and homogeneity of variance, unpaired t tests were used for 2 groups and 1-way ANOVA, followed by Turkey post hoc test for >2 groups. For nonparametric variables, we used Wilcoxon tests for multiple comparisons. All data were presented as mean±SD. CKD indicates chronic kidney disease; CM, calcifying medium; OCN, osteocalcin; OPN, osteopontin; OSX, osterix; VEGFA, vascular endothelial growth factor A; and VSMCs, vascular smooth muscle cells.

To further determine if VEGFR1 or VEGFR2 is critical for signal transduction in VC, silencing of VEGFR1, VEGFR 2, or both with siRNA was performed in A10 cells. The silencing efficiency is shown in Figure 5F. As presented in Figure 5G and 5H, siVEGFR2 ameliorated calcified deposits compared to the scrambled siRNA group or sham transfection (Figure S5B and S5C). However, a combination of siVEGFR1 and siVEGFR2 did not further suppress calcification. The single siVEGFR2 and the combination of siVEGFR1 and siVEGFR2 suppressed the mRNA expression levels of OSX and OCN (Figure 5I). These findings suggest that VEGFA-VEGFR2 signaling drives VC, and CKD causes an insufficient delivery of the protective miRNA quartet targeting VEGFA, promoting VC.

Deficiency of miR-16-5p, miR-17-5p, miR-20a-5p, and miR-106b-5p Expression Levels in sEVs is Predictive of Aortic VC in CKD Patients

To determine if the expression levels of hsa-miR-16-5p, hsa-miR-17-5p, hsa-miR-20a-5p, and hsa-miR-106b-5p predict aortic or coronary arterial calcification, we measured the aortic calcification index (ACI) of the abdominal aorta from the renal artery bifurcation to the femoral artery bifurcation by dividing the abdominal aorta into 12 sections by 5-mm slices. The ACI was calculated as follows: ACI=(total score for calcification on all slices)/12×1/(number of slices)×100 (%).37Figure 6A shows the representative abdominal CT scans of 2 patients with a high or low ACI score among 33 CKD patients (Table S2). Figure 6B illustrates the association of quartiles of miRNA expression levels in sEVs relative to miR-139-5p, eGFR, and serum phosphate. There were significant associations between hsa-miR-16-5p, hsa-miR-17-5p, hsa-miR-20a-5p, and hsa-miR-106b-5p with ACI (P<0.05). Older age was also linked to the higher ACI, whereas serum phosphate, serum calcium corrected for albumin, and sex showed weaker associations with ACI (Figure 6B and S6B).

Figure 6.

Figure 6. Deficiency of miR-16-5p, miR-17-5p, miR-20a-5p, and miR-106b-5p expression levels in sEVs is predictive of aortic vascular calcification in CKD patients. A, The representative abdominal CT scans of 2 patients with a high or low aortic calcification index (ACI) score among 33 CKD patients. We measured ACI of the abdominal aorta from the renal artery bifurcation to the femoral artery bifurcation by dividing the abdominal aorta into 12 sections every 5 mm slice. B, The association of quartiles of miRNA expression levels in sEVs relative to miR-139-5p, eGFR, and serum phosphate. sEVs were purified using polymer-based precipitation (Figure 1G). There were significant associations between hsa-miR-16-5p (n=8, 8, 7, and 10 for Q1–Q4, respectively), hsa-miR-17-5p (n=9, 8, 6, and 10 for Q1–Q4, respectively), hsa-miR-20a-5p (n=9, 8, 7, and 9 for Q1–Q4, respectively), and hsa-miR-106b-5p (n=9, 9, 6, and 9 for Q1–Q4, respectively) with ACI. Grubbs test was used to detect outliers and Spearman correlation coefficient was used to identify correlations between nonparametric variables. P-values <0.01 were considered statistically significant. C, ROC curves for predicting aortic vascular calcification by miRNAs, eGFR, and serum phosphate level. The area under the curves (AUCs [95% CIs; P-values]) for ACI ≥15% were 0.7630 [0.5818–0.9441; 0.010], 0.7704 [0.5869–0.9538; 0.008], 0.7407 [0.5550–0.9265; 0.019], 0.7704 [0.5813–0.9594; 0.008], 0.5537 [0.3415–0.7659], and 0.7540 [0.5858–0.9221], respectively. D, A one-quartile decrease in the expression of hsa-miR-16-5p, hsa-miR-17-5p, hsa-miR-20a-5p, or hsa-miR-106b-5p increased the ACI by 8.01 [0.99–15.0; 0.027], 8.67 [1.43–15.9; 0.021], 8.32 [0.40–16.2; 0.040], or 9.67 [1.68–17.7; 0.027], respectively. The Shapiro–Wilk test was used to test the normality of variables. The Wilcoxon test was used for nonparametric analyses. All data were presented as mean±SD. CKD indicates chronic kidney disease; eGFR, estimated glomerular filtration rate; and sEVs, small extracellular vesicles.

As shown in Figure 6C, the area under the curve [95% CI; P-value] of each of the expression levels of hsa-miR-16-5p, hsa-miR-17-5p, hsa-miR-20a-5p, and hsa-miR-106b-5p for ACI ≥15% were 0.7630 [0.5818–0.9441; 0.010], 0.7704 [0.5869–0.9538; 0.008], 0.7407 [0.5550–0.9265; 0.019], and 0.7704 [0.5813–0.9594; 0.008], respectively. These miRNAs were more effective predictors of aortic VC than kidney function eGFR (0.5537; [0.3415–0.7659; 0.6]). The area under the curve of serum phosphate, a well-known CKD-mineral bone disorder-related variable, was 0.7540 [0.5858–0.9221; 0.015], similar to previous studies.38

Therefore, to exclude the confounding effect of serum phosphate, age, or sex, we analyzed the effect of miRNA expression levels in sEVs on the ACI with multivariable linear regression models. As shown in Figure 6D a one-quartile decrease in the expression of hsa-miR-16-5p, hsa-miR-17-5p, hsa-miR-20a-5p, or hsa-miR-106b-5p increased the ACI by 8.01 [0.99–15.0; 0.027], 8.67 [1.43–15.9; 0.021], 8.32 [0.40–16.2; 0.040], or 9.67 [1.68–17.7; 0.027], respectively. In all analyses, older age and lower eGFR were linked to higher ACI, while serum phosphate was not associated with ACI after adjusting for confounders. A one-quartile decrease in the expression of hsa-miR-106b-5p was equal to an age increase of 14.8 years and an eGFR reduction of 23.0 mL/min per 1.73 meter squared. As shown in Figure S6C, we also examined the association of miRNA expression in sEVs with coronary arterial calcification on thoracic CT scans among 35 patients who received this examination. Agaston’s evaluation method was used to determine the presence of coronary artery calcification with a threshold of ≥130 HU in the coronary arteries.39 The area under the curves of the expression levels of hsa-miR-16-5p, hsa-miR-17-5p, hsa-miR-20a-5p, and hsa-miR-106b-5p were 0.7288 [0.5568–0.9007; 0.021], 0.6863 [0.5036–0.8689; 0.060], 0.6863 [0.5036–0.8689; 0.060], and 0.6910 [0.5112–0.8707; 0.058], respectively (Figure S6C).

To assess the potential source of CKD-derived malicious sEVs, we quantified miRNA expression levels in the kidney lysates of CKD model rats. The expression level of miR-16-5p was decreased in CKD kidneys among the quartet of miRNAs, while those of the other 3 miRNAs were not altered by CKD (Figure S7A). Consequently, we speculated that sEVs released by the endothelial cells neighboring the smooth muscle cells are devoid of these miRNAs, given that endothelial cells are among the major sources of circulating sEVs. We examined the miRNA expression levels of human umbilical vein endothelial cells (HUVECs) treated with 5% CKD serum for 24 hours. The expression levels of miR-16-5p, miR-17-5p, miR-20a-5p, and miR-106b-5p were decreased by CKD (Figure S7B). Potential sources of sEVs devoid of miRNAs include endothelial cells; however, further investigation is required.

Figure 7 illustrates a schematic summary of this work. The CKD-derived sEVs in circulation display the malicious property of pathogenically promoting Ca deposition in VSMCs due to a change in miRNA profile propagated in sEVs. Particularly, sEVs in CKD are depleted of miR-16-5p, miR-17-5p, miR-20a-5p, and miR-106b-5p, which are enriched in healthy sEVs and cooperatively target VEGFA, leading to the suppression of the mRNA and protein expression of VEGFA in VSMCs. VEGFA activates VEGFR2 signaling as its primary ligand and facilitates the transcription of osteogenic marker genes, phenotypic switching of VSMCs, and VC.

Figure 7.

Figure 7. Schematic summary of interorgan communications between kidneys and remote VSMCs leading to vascular calcification. The CKD-derived sEVs in circulation display the malicious property of pathogenically promoting calcium deposition in VSMCs, due to a change in miRNA profile propagated in sEVs. Particularly, sEVs in CKD are depleted of miR-16-5p, miR-17-5p, miR-20a-5p, and miR-106b-5p, which are usually enriched in healthy sEVs and cooperatively target VEGFA, leading to the suppression of mRNA and protein expression of VEGFA (vascular endothelial growth factor A) in VSMCs. VEGFA activates VEGFR2 signaling as its primary ligand and facilitates transcription of osteogenic marker genes, phenotypic switching of VSMCs, and vascular calcification. These findings indicate that miR-16-5p, miR-17-5p, miR-20a-5p, and miR-106b-5p propagated in healthy sEVs are essential for protecting vascular calcification (VC, left), and CKD pathologically suppresses this safeguard mechanism in mice (right). CKD indicates chronic kidney disease; eGFR, estimated glomerular filtration rate; miRNA, microRNA; sEVs, small extracellular vesicles; and VSMCs, vascular smooth muscle cells.

Discussion

This study unveiled a transcriptomic signature of miRNAs in circulating sEVs of CKD model rodents and humans, showing depleted miR-16-5p, miR-17-5p, miR-20a-5p, and miR-106b-5p. The in silico target analyses and biological assays showed that VEGFA-VEGFR2 signaling targeted by this miRNA quartet is a key signaling pathway that drives VC in CKD. Using the CKD cohort patients, we also showed that each miRNA expression level is useful for predicting calcification of the abdominal aorta.38 These findings provide a potential therapeutic strategy for targeting these miRNAs and the VEGFA-VEGFR2 axis as biomarkers for the noninvasive prediction of VC.

This study clarified that a fraction of sEVs, but not sEV-depleted serum of CKD, facilitates VC in VSMCs and mice (Figures 1 and 2). A previous study found that the serum of patients undergoing hemodialysis drives Ca deposition in VSMCs compared to the serum of nondialysis patients.40 However, the biological mechanisms and key molecules propagated by serum were not identified. This study showed that sEVs play an essential role in this pathophysiology. The transcriptomic analyses of miRNAs in circulating sEVs revealed that a subset of miRNAs decreased in a CKD rodent model. Each expression level decreased as kidney function deteriorated in humans (Figure 3). We further showed that this miRNA quartet was protective against VC by cooperatively targeting VEGFA (Figure 4).

sEVs have long been recognized as cargos for discarding unnecessary molecules from cells. Since Valadi et al41 reported the intercellular transfer of miRNAs and their biological functions between cells, attempts have been made to find miRNAs associated with disease susceptibility that have diagnostic potential, particularly for cancer.13,41–43 However, few reports have identified miRNAs associated with CKD and CKD-related calcification.44 One of the reasons is the substantial challenge in detecting changes in expression levels of miRNAs secreted from normal cells in a dynamic steady state, in contrast to specifically secreted miRNAs from cancer cells. CKD exists on a continuum of severity that is graded from G1 to G5, while G5D is for patients who are dependent on dialysis.6,45 Therefore, after screening for candidate miRNAs propagated in sEVs from CKD model rodents, we analyzed the link between eGFR,46,47 miRNA expression levels, and ACI as continuous variables in humans with heterogeneous kidney function and comorbidities (Figures 3 and 6). We identified the key miRNAs and their target molecule in CKD-driven VC. Given the irreversibility of kidney dysfunction and VC, there is an urgent need for early noninvasive biomarkers. These findings provide novel insights into the diagnosis and therapy of VC.

We showed the therapeutic potential of targeting sEVs in VC (Figure 2). Circulating sEVs are abundant in the blood and urine.13,14 Notably, intraperitoneal administration in mice of GW4869, a systemic inhibitor of sEV biogenesis, ameliorated thoracic aortic calcification. GW4869 is a noncompetitive neutral sphingomyelinase inhibitor.23,24 GW4869 almost entirely inhibited the increased mRNA transcription of osteogenic marker genes. Previous studies have used GW4869 to elucidate the biological function of sEVs in intercellular communication, especially in local networks in the central nervous system.21,48,49 More recent studies have identified an essential role of sEVs in interorgan communication related to rodent survival, with little investigation on their functional cargos.50 This study found a critical role for sEVs, key miRNAs, and their target molecule that is essential in interorgan communication between the kidneys and remote VSMCs (Figure 7). The substantial improvement in vascular calcification and osteogenic marker gene expression with GW4869 treatment might be partly attributed to the suppression of procalcification inflammatory cytokines.50 The important limitation of this study was that, while the sEVs purified with the 2 isolation techniques exhibited similar EV properties, including protein composition27 and miRNA profiles, we could not conclusively demonstrate shared sEV characteristics between the isolation techniques. Further studies are needed to examine the specific effect of sEV removal in vivo. Although we aimed to use the same species among the cultured cells, serum, and sEV experiments, we examined the drug efficacy in mice rather than rats, which is a further limitation of our study.

VEGFA regulates bone formation and homeostasis by promoting osteoblast differentiation and angiogenesis,51 causing heterozygous embryonic lethality.52 A recent study identified a plasma VEGFA level as a strong predictor of kidney disease progression, demonstrating a pathophysiological link between VEGFA and kidney disease.53 However, the role of VEGFA in VC has not been elucidated. This study showed that VEGFA is a convergent target of miR-16-5p, miR-17-5p, miR-20a-5p, and miR-106b-5p, which are depleted in CKD sEVs. This finding is consistent with in silico prediction and is supported by previous experimental evidence.33,34 In the thoracic aorta of CKD-induced VC model mice, there was a 16-fold increase in Vegfa transcription, and GW4869 treatment reversed this increase. Inhibition of the VEGFA-VEGFR2 axis with sorafenib, fruquintinib, and sunitinib, initially developed for cancer therapy, mitigated calcification in VSMCs. Fruquintinib, the most efficient drug in the in vitro assay (Figure 5), received global approval for metastatic colorectal cancer and is potentially more effective in combination with antiprogrammed cell death receptor-1 antibodies.54,55 A 4-week oral administration of fruquintinib36 suppressed transcription of the osteogenic marker genes in the thoracic aorta of mice. VEGFR2 inhibition might be a therapeutic option for VC. Recent studies reported that VC was driven by a newly investigated drug for renal anemia, a HIF (hypoxia-inducible factor) PH (prolyl hydroxylase) enzyme inhibitor, in an in vitro assay of VSMCs.56 The mechanisms remain to be fully understood. Activated VEGFA transcription via the induction of HIF might be associated with the promotion of VC due to HIF-PH inhibitors.

In conclusion, this study revealed the transcriptomic landscape of miRNAs propagated in sEVs in rodent models of CKD. Four miRNAs that cooperatively target VEGFA are depleted in these sEVs, driving VC in CKD. We also investigated the therapeutic potential of VEGFR2 inhibition and the diagnostic efficiency of the miRNA quartet in humans.

Article Information

Acknowledgments

The authors thank all the study participants. They also thank all members of our laboratory for their helpful discussions regarding this work.

Supplemental Material

Supplemental Methods 57–58

Tables S1–S3

Figures S1–S7

Nonstandard Abbreviations and Acronyms

CKD

chronic kidney disease

CM

calcifying medium

CVD

cardiovascular disease

eGFR

estimated glomerular filtration rate

FC

fold change

FGF23

fibroblast growth factor 23

HIF

hypoxia-inducible factor

MGP

matrix gla-protein

miRNA

microRNA

OCN

osteocalcin

OPN

osteopontin

OSX

osterix

PH

prolyl hydroxylase

sEVs

small extracellular vesicles

VC

vascular calcification

VEGFA

vascular endothelial growth factor A

VSMCs

vascular smooth muscle cells

Disclosures None.

Footnotes

*T. Koide and S. Mandai contributed equally.

For Sources of Funding and Disclosures, see page 430.

Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/CIRCRESAHA.122.321939.

Correspondence to: Shintaro Mandai, MD, PhD, Department of Nephrology, Graduate School of Medical and Dental Sciences, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo, Tokyo 113-8519, Japan. Email

References

  • 1. GBD 2017 Causes of Death Collaborators. Global, regional, and national age-sex-specific mortality for 282 causes of death in 195 countries and territories, 1980–2017: a systematic analysis for the global burden of disease study 2017.Lancet. 2018; 392:1736–1788. doi: 10.1016/S0140-6736(18)32203-7CrossrefMedlineGoogle Scholar
  • 2. Roth GA, Dwyer-Lindgren L, Bertozzi-Villa A, Stubbs RW, Morozoff C, Naghavi M, Mokdad AH, Murray CJL. Trends and patterns of geographic variation in cardiovascular mortality among US counties, 1980–2014.JAMA. 2017; 317:1976–1992. doi: 10.1001/jama.2017.4150CrossrefMedlineGoogle Scholar
  • 3. World Health Organization. Cardiovascular diseases.2019. http://www.who.int/mediacentre/factsheets/fs317/en/.Google Scholar
  • 4. Sage AP, Tintut Y, Demer LL. Regulatory mechanisms in vascular calcification.Nat Rev Cardiol. 2010; 7:528–536. doi: 10.1038/nrcardio.2010.115CrossrefMedlineGoogle Scholar
  • 5. Durham AL, Speer MY, Scatena M, Giachelli CM, Shanahan CM. Role of smooth muscle cells in vascular calcification: implications in atherosclerosis and arterial stiffness.Cardiovasc Res. 2018; 114:590–600. doi: 10.1093/cvr/cvy010CrossrefMedlineGoogle Scholar
  • 6. GBD Chronic Kidney Disease Collaboration. Global, regional, and national burden of chronic kidney disease, 1990–2017: a systematic analysis for the global burden of disease study 2017.Lancet. 2020; 395:709–733. doi: 10.1016/S0140-6736(20)30045-3CrossrefMedlineGoogle Scholar
  • 7. Foley RN, Murray AM, Li S, Herzog CA, McBean AM, Eggers PW, Collins AJ. Chronic kidney disease and the risk for cardiovascular disease, renal replacement, and death in the United States Medicare population, 1998 to 1999.J Am Soc Nephrol. 2005; 16:489–495. doi: 10.1681/ASN.2004030203CrossrefMedlineGoogle Scholar
  • 8. Jankowski J, Floege J, Fliser D, Böhm M, Marx N. Cardiovascular disease in chronic kidney disease: pathophysiological insights and therapeutic options.Circulation. 2021; 143:1157–1172. doi: 10.1161/circulationaha.120.050686LinkGoogle Scholar
  • 9. Kono K, Fujii H, Nakai K, Goto S, Shite J, Hirata K, Fukagawa M, Nishi S. Composition and plaque patterns of coronary culprit lesions and clinical characteristics of patients with chronic kidney disease.Kidney Int. 2012; 82:344–351. doi: 10.1038/ki.2012.118CrossrefMedlineGoogle Scholar
  • 10. Ketteler M, Block GA, Evenepoel P, Fukagawa M, Herzog CA, McCann L, Moe SM, Shroff R, Tonelli MA, Toussaint ND, et al. Executive summary of the 2017 KDIGO Chronic Kidney Disease-Mineral and Bone Disorder (CKD-MBD) Guideline update: what’s changed and why it matters.Kidney Int. 2017; 92:26–36. doi: 10.1016/j.kint.2017.04.006CrossrefMedlineGoogle Scholar
  • 11. Lioufas NM, Pascoe EM, Hawley CM, Elder GJ, Badve SV, Block GA, Johnson DW, Toussaint ND. Systematic review and meta-analyses of the effects of phosphate-lowering agents in nondialysis CKD.J Am Soc Nephrol. 2022; 33:59–76. doi: 10.1681/ASN.2021040554CrossrefMedlineGoogle Scholar
  • 12. Chertow GM, Burke SK, Raggi P. Treat to Goal Working Group. Sevelamer attenuates the progression of coronary and aortic calcification in hemodialysis patients.Kidney Int. 2002; 62:245–252. doi: 10.1046/j.1523-1755.2002.00434.xCrossrefMedlineGoogle Scholar
  • 13. Kalluri R, LeBleu VS. The biology, function, and biomedical applications of exosomes.Science. 2020; 367:eaau6977. doi: 10.1126/science.aau6977CrossrefMedlineGoogle Scholar
  • 14. Sahoo S, Adamiak M, Mathiyalagan P, Kenneweg F, Kafert-Kasting S, Thum T. Therapeutic and diagnostic translation of extracellular vesicles in cardiovascular diseases: roadmap to the clinic.Circulation. 2021; 143:1426–1449. doi: 10.1161/circulationaha.120.049254LinkGoogle Scholar
  • 15. Tieu A, Lalu MM, Slobodian M, Gnyra C, Fergusson DA, Montroy J, Burger D, Stewart DJ, Allan DS. An analysis of mesenchymal stem cell-derived extracellular vesicles for preclinical use.ACS Nano. 2020; 14:9728–9743. doi: 10.1021/acsnano.0c01363CrossrefMedlineGoogle Scholar
  • 16. Kwon DH, Eom GH, Ko JH, Shin S, Joung H, Choe N, Nam YS, Min HK, Kook T, Yoon S, et al. MDM2 E3 ligase-mediated ubiquitination and degradation of HDAC1 in vascular calcification.Nat Commun. 2016; 7:10492. doi: 10.1038/ncomms10492CrossrefMedlineGoogle Scholar
  • 17. Yokozawa T, Zheng PD, Oura H, Koizumi F. Animal model of adenine-induced chronic renal failure in rats.Nephron. 1986; 44:230–234. doi: 10.1159/000183992CrossrefMedlineGoogle Scholar
  • 18. Osako MK, Nakagami H, Koibuchi N, Shimizu H, Nakagami F, Koriyama H, Shimamura M, Miyake T, Rakugi H, Morishita R. Estrogen inhibits vascular calcification via vascular RANKL system: common mechanism of osteoporosis and vascular calcification.Circ Res. 2010; 107:466–475. doi: 10.1161/circresaha.110.216846LinkGoogle Scholar
  • 19. Hodroge A, Trécherel E, Cornu M, Darwiche W, Mansour A, Ait-Mohand K, Verissimo T, Gomila C, Schembri C, Da Nascimento S, et al. Oligogalacturonic acid inhibits vascular calcification by two mechanisms: inhibition of vascular smooth muscle cell osteogenic conversion and interaction with collagen.Arterioscler Thromb Vasc Biol. 2017; 37:1391–1401. doi: 10.1161/atvbaha.117.309513LinkGoogle Scholar
  • 20. Schantl AE, Verhulst A, Neven E, Behets GJ, D’Haese PC, Maillard M, Mordasini D, Phan O, Burnier M, Spaggiari D, et al. Inhibition of vascular calcification by inositol phosphates derivatized with ethylene glycol oligomers.Nat Commun. 2020; 11:721. doi: 10.1038/s41467-019-14091-4CrossrefMedlineGoogle Scholar
  • 21. Bracamonte-Baran W, Florentin J, Zhou Y, Jankowska-Gan E, Haynes WJ, Zhong W, Brennan TV, Dutta P, Claas FH, van Rood JJ, et al. Modification of host dendritic cells by microchimerism-derived extracellular vesicles generates split tolerance.Proc Natl Acad Sci U S A. 2017; 114:1099–1104. doi: 10.1073/pnas.1618364114CrossrefMedlineGoogle Scholar
  • 22. Bhagirath D, Yang TL, Bucay N, Sekhon K, Majid S, Shahryari V, Dahiya R, Tanaka Y, Saini S. microRNA-1246 is an exosomal biomarker for aggressive prostate cancer.Cancer Res. 2018; 78:1833–1844. doi: 10.1158/0008-5472.CAN-17-2069CrossrefMedlineGoogle Scholar
  • 23. Airola MV, Shanbhogue P, Shamseddine AA, Guja KE, Senkal CE, Maini R, Bartke N, Wu BX, Obeid LM, Garcia-Diaz M, et al. Structure of human nSMase2 reveals an interdomain allosteric activation mechanism for ceramide generation.Proc Natl Acad Sci U S A. 2017; 114:E5549–E5558. doi: 10.1073/pnas.1705134114CrossrefMedlineGoogle Scholar
  • 24. Asai H, Ikezu S, Tsunoda S, Medalla M, Luebke J, Haydar T, Wolozin B, Butovsky O, Kügler S, Ikezu T. Depletion of microglia and inhibition of exosome synthesis halt tau propagation.Nat Neurosci. 2015; 18:1584–1593. doi: 10.1038/nn.4132CrossrefMedlineGoogle Scholar
  • 25. Zhang X, Wang S, Wang H, Cao J, Huang X, Chen Z, Xu P, Sun G, Xu J, Lv J, et al. Circular RNA circNRIP1 acts as a microRNA-149-5p sponge to promote gastric cancer progression via the AKT1/mTOR pathway.Mol Cancer. 2019; 18:20. doi: 10.1186/s12943-018-0935-5CrossrefMedlineGoogle Scholar
  • 26. Fricke A, Ullrich PV, Heinz J, Pfeifer D, Scholber J, Herget GW, Hauschild O, Bronsert P, Stark GB, Bannasch H, et al. Identification of a blood-borne miRNA signature of synovial sarcoma.Mol Cancer. 2015; 14:151. doi: 10.1186/s12943-015-0424-zCrossrefMedlineGoogle Scholar
  • 27. Théry C, Witwer KW, Aikawa E, Alcaraz MJ, Anderson JD, Andriantsitohaina R, Antoniou A, Arab T, Archer F, Atkin-Smith GK, et al. Minimal information for studies of extracellular vesicles 2018 (MISEV2018): a position statement of the International Society for Extracellular Vesicles and update of the MISEV2014 guidelines.J Extracell Vesicles. 2018; 7:1535750. doi: 10.1080/20013078.2018.1535750. eCollection 2018CrossrefMedlineGoogle Scholar
  • 28. Bobot M, Thomas L, Moyon A, Fernandez S, McKay N, Balasse L, Garrigue P, Brige P, Chopinet S, Poitevin S, et al. Uremic toxic blood-brain barrier disruption mediated by AhR activation leads to cognitive impairment during experimental renal dysfunction.J Am Soc Nephrol. 2020; 31:1509–1521. doi: 10.1681/asn.2019070728CrossrefMedlineGoogle Scholar
  • 29. Ishikawa S, Naito S, Iimori S, Takahashi D, Zeniya M, Sato H, Nomura N, Sohara E, Okado T, Uchida S, et al. Loop diuretics are associated with greater risk of sarcopenia in patients with non-dialysis-dependent chronic kidney disease.PLoS One. 2018; 13:e0192990. doi: 10.1371/journal.pone.0192990CrossrefMedlineGoogle Scholar
  • 30. Vlachos IS, Zagganas K, Paraskevopoulou MD, Georgakilas G, Karagkouni D, Vergoulis T, Dalamagas T, Hatzigeorgiou AG. DIANA-miRPath v3.0: deciphering microRNA function with experimental support.Nucleic Acids Res. 2015; 43:W460–W466. doi: 10.1093/nar/gkv403CrossrefMedlineGoogle Scholar
  • 31. Hamberg M, Backes C, Fehlmann T, Hart M, Meder B, Meese E, Keller A. miRTargetLink—miRNAs, genes and interaction networks.Int J Mol Sci. 2016; 17:564.CrossrefMedlineGoogle Scholar
  • 32. miRTargetLink.(Accessed on March 23, 2020). Available online: http://ccb-web.cs.uni-saarland.de/mirtargetlink.Google Scholar
  • 33. Qu Y, Liu H, Lv X, Liu Y, Wang X, Zhang M, Zhang X, Li Y, Lou Q, Li S, et al. MicroRNA-16-5p overexpression suppresses proliferation and invasion as well as triggers apoptosis by targeting VEGFA expression in breast carcinoma.Oncotarget. 2017; 8:72400–72410. doi: 10.18632/oncotarget.20398CrossrefMedlineGoogle Scholar
  • 34. Yue XH, Guo L, Wang ZY, Jia TH. Inhibition of miR-17-5p promotes mesenchymal stem cells to repair spinal cord injury.Eur Rev Med Pharmacol Sci. 2019; 23:3899–3907. doi: 10.26355/eurrev_201905_17819CrossrefMedlineGoogle Scholar
  • 35. Qiao M, Shapiro P, Fosbrink M, Rus H, Kumar R, Passaniti A. Cell cycle-dependent phosphorylation of the RUNX2 transcription factor by cdc2 regulates endothelial cell proliferation.J Biol Chem. 2006; 281:7118–7128. doi: 10.1074/jbc.m508162200CrossrefMedlineGoogle Scholar
  • 36. Sun Q, Zhou J, Zhang Z, Guo M, Liang J, Zhou F, Long J, Zhang W, Yin F, Cai H, et al. Discovery of fruquintinib, a potent and highly selective small -molecule inhibitor of VEGFR 1, 2, 3 tyrosine kinases for cancer therapy.Cancer Biol Ther. 2014; 15:1635–1645. doi: 10.4161/15384047.2014.964087CrossrefMedlineGoogle Scholar
  • 37. Tatami Y, Yasuda Y, Suzuki S, Ishii H, Sawai A, Shibata Y, Ota T, Shibata K, Niwa M, Morimoto R, et al. Impact of abdominal aortic calcification on long-term cardiovascular outcomes in patients with chronic kidney disease.Atherosclerosis. 2015; 243:349–355. doi: 10.1016/j.atherosclerosis.2015.10.016CrossrefMedlineGoogle Scholar
  • 38. Yokoyama K, Yoshida H, Maruyama Y, Ozaki A, Yamamoto H, Hosoya T. Should the targeted value of the phosphate be reviewed in the normal range from the viewpoint of vascular calcification?.Kidney Int. 2010; 77:928–929. doi: 10.1038/ki.2010.50CrossrefMedlineGoogle Scholar
  • 39. Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Viamonte M, Detrano R. Quantification of coronary artery calcium using ultrafast computed tomography.J Am Coll Cardiol. 1990; 15:827–832. doi: 10.1016/0735-1097(90)90282-tCrossrefMedlineGoogle Scholar
  • 40. Cazaña-Pérez V, Cidad P, Donate-Correa J, Martín-Núñez E, López-López JR, Pérez-García MT, Giraldez T, Navarro-González JF, Alvarez de la Rosa D. Phenotypic modulation of cultured primary human aortic vascular smooth muscle cells by uremic serum.Front Physiol. 2018; 9:89. doi: 10.3389/fphys.2018.00089. eCollection 2018CrossrefMedlineGoogle Scholar
  • 41. Valadi H, Ekström K, Bossios A, Sjöstrand M, Lee JJ, Lötvall JO. Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells.Nat Cell Biol. 2007; 9:654–659. doi: 10.1038/ncb1596CrossrefMedlineGoogle Scholar
  • 42. Sun Z, Shi K, Yang S, Liu J, Zhou Q, Wang G, Song J, Li Z, Zhang Z, Yuan W. Effect of exosomal miRNA on cancer biology and clinical applications.Mol Cancer. 2018; 17:147. doi: 10.1186/s12943-018-0897-7CrossrefMedlineGoogle Scholar
  • 43. Slack FJ, Chinnaiyan AM. The role of non-coding RNAs in oncology.Cell. 2019; 179:1033–1055. doi: 10.1016/j.cell.2019.10.017CrossrefMedlineGoogle Scholar
  • 44. Goettsch C, Hutcheson JD, Aikawa E. MicroRNA in cardiovascular calcification: focus on targets and extracellular vesicle delivery mechanisms.Circ Res. 2013; 112:1073–1084. doi: 10.1161/circresaha.113.300937LinkGoogle Scholar
  • 45. Carney EF. The impact of chronic kidney disease on global health.Nat Rev Nephrol. 2020; 16:251–251. doi: 10.1038/s41581-020-0268-7CrossrefMedlineGoogle Scholar
  • 46. Levey AS, Eckardt KU, Tsukamoto Y, Levin A, Coresh J, Rossert J, De Zeeuw D, Hostetter TH, Lameire N, Eknoyan G. Definition and classification of chronic kidney disease: a position statement from Kidney Disease: Improving Global Outcomes (KDIGO).Kidney Int. 2005; 67:2089–2100. doi: 10.1111/j.1523-1755.2005.00365.xCrossrefMedlineGoogle Scholar
  • 47. Matsuo S, Imai E, Horio M, Yasuda Y, Tomita K, Nitta K, Yamagata K, Tomino Y, Yokoyama H, Hishida A; Collaborators developing the Japanese equation for estimated GFR. Revised equations for estimated GFR from serum creatinine in Japan.Am J Kidney Dis. 2009; 53:982–992. doi: 10.1053/j.ajkd.2008.12.034CrossrefMedlineGoogle Scholar
  • 48. Dinkins MB, Dasgupta S, Wang G, Zhu G, Bieberich E. Exosome reduction in vivo is associated with lower amyloid plaque load in the 5XFAD mouse model of Alzheimer’s disease.Neurobiol Aging. 2014; 35:1792–1800. doi: 10.1016/j.neurobiolaging.2014.02.012CrossrefMedlineGoogle Scholar
  • 49. Guo M, Wang J, Zhao Y, Feng Y, Han S, Dong Q, Cui M, Tieu K. Microglial exosomes facilitate alpha-synuclein transmission in Parkinson’s disease.Brain. 2020; 143:1476–1497. doi: 10.1093/brain/awaa090CrossrefMedlineGoogle Scholar
  • 50. Essandoh K, Yang L, Wang X, Huang W, Qin D, Hao J, Wang Y, Zingarelli B, Peng T, Fan GC. Blockade of exosome generation with GW4869 dampens the sepsis-induced inflammation and cardiac dysfunction.Biochim Biophys Acta. 2015; 1852:2362–2371. doi: 10.1016/j.bbadis.2015.08.010CrossrefMedlineGoogle Scholar
  • 51. Hu K, Olsen BR. Osteoblast-derived VEGF regulates osteoblast differentiation and bone formation during bone repair.J Clin Invest. 2016; 126:509–526. doi: 10.1172/jci82585CrossrefMedlineGoogle Scholar
  • 52. Carmeliet P, Ferreira V, Breier G, Pollefeyt S, Kieckens L, Gertsenstein M, Fahrig M, Vandenhoeck A, Harpal K, Eberhardt C, et al. Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele.Nature. 1996; 380:435–439. doi: 10.1038/380435a0CrossrefMedlineGoogle Scholar
  • 53. Schmidt IM, Sarvode Mothi S, Wilson PC, Palsson R, Srivastava A, Onul IF, Kibbelaar ZA, Zhuo M, Amodu A, Stillman IE, et al. Circulating plasma biomarkers in biopsy-confirmed kidney disease.Clin J Am Soc Nephrol. 2022; 17:27–37. doi: 10.2215/CJN.09380721CrossrefMedlineGoogle Scholar
  • 54. Li J, Qin S, Xu RH, Shen L, Xu J, Bai Y, Yang L, Deng Y, Chen ZD, Zhong H, et al. Effect of fruquintinib vs placebo on overall survival in patients with previously treated metastatic colorectal cancer: the FRESCO randomized clinical trial.JAMA. 2018; 319:2486–2496. doi: 10.1001/jama.2018.7855CrossrefMedlineGoogle Scholar
  • 55. Wang Y, Wei B, Gao J, Cai X, Xu L, Zhong H, Wang B, Sun Y, Guo W, Xu Q, et al. Combination of fruquintinib and anti-PD-1 for the treatment of colorectal cancer.J Immunol. 2020; 205:2905–2915. doi: 10.4049/jimmunol.2000463CrossrefMedlineGoogle Scholar
  • 56. Nagy A, Pethő D, Gáll T, Zavaczki E, Nyitrai M, Posta J, Zarjou A, Agarwal A, Balla G, Balla J. Zinc inhibits HIF-prolyl hydroxylase inhibitor-aggravated VSMC calcification induced by high phosphate.Front Physiol. 2020; 10:1584. doi: 10.3389/fphys.2019.01584. eCollection 2019CrossrefMedlineGoogle Scholar

eLetters(0)

eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.

Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.