Tipifarnib Reduces Extracellular Vesicles and Protects from Heart Failure

Abstract

BACKGROUND:

Heart failure (HF) is one of the leading causes of mortality worldwide. Extracellular vesicles (EVs) including small EVs, or exosomes and their molecular cargo are known to modulate cell to cell communication during multiple cardiac diseases. However, the role of systemic EV biogenesis inhibition in the models of HF is not well documented and remains unclear.

METHODS:

We investigated the role of circulating exosomes during cardiac dysfunction and remodeling in a mouse transverse aortic constriction (TAC) model of HF. Importantly, we investigate the efficacy of Tipifarnib (Tip), a recently identified exosome biogenesis inhibitor that targets the critical proteins (Rab27a, nSMase2 and Alix) involved in exosome biogenesis for this mouse model of HF. In this study, 10-week-old male mice underwent TAC surgery, were randomly assigned to groups with and without Tip treatment (10 mg/kg three times/week) and monitored for 8 weeks and a comprehensive assessment was conducted through performed echocardiographic, histological, and biochemical studies.

RESULTS:

TAC significantly elevated circulating plasma exosomes and markedly increased cardiac left ventricular (LV) dysfunction, cardiac hypertrophy, and fibrosis. Furthermore, injection of plasma exosomes from TAC mice induced LV dysfunction and cardiomyocyte hypertrophy in uninjured mice without TAC. On the contrary, treatment of Tip in TAC mice reduced circulating exosomes to baseline and remarkably improved LV functions, hypertrophy, and fibrosis. Tip treatment also drastically altered the miRNA profile of circulating post-TAC exosomes, including miR331-5p which was highly downregulated both in TAC circulating exosomes and in TAC cardiac tissue. Mechanistically, miR331-5p is crucial for inhibiting fibroblast-to-myofibroblast transition by targeting HOXC8, a critical regulator of fibrosis. Tipifarnib treatment in TAC mice upregulated the expression of miR331-5p that acts as potent repressor for one of the fibrotic mechanisms mediated by HOXC8.

CONCLUSIONS:

Our study underscores the pathological role of exosomes in HF and fibrosis in response to pressure overload. Tipifarnib-mediated inhibition of exosome biogenesis and cargo sorting may serve as a viable strategy to prevent progressive cardiac remodeling to HF.

Graphical Abstract

INTRODUCTION:

According to the World Heart Report 2023, cardiovascular diseases (CVDs) are the major life-threatening diseases accounting for 33% of global mortality. Clinical evidence signifies the prevalence of Heart failure (HF) caused by various etiologies including Hypertrophic cardiomyopathy (HCM), ischemic heart disease, stroke and hypertension adding to 85% of all CVDs worldwide ^1-3. The prospect of discovering molecular mechanisms regulating cardiac remodeling has been a constant challenge.

Extracellular vesicles (EVs), especially small EVs (sEVs) also called exosomes, have recently emerged as key paracrine regulators of cardiovascular diseases ^4-10. Exosomes are 30 - 200 nm lipid bilayer membrane vesicles carrying biologically active compounds such as proteins, nucleic acids, miRNA, metabolites etc. secreted from its originating cell and eventually internalized by recipient cells both within tissue and at long distance organs ¹¹. Primarily, exosomes can change the fate of recipient cells through autocrine or paracrine signaling in response to specific donor cell exosome cargo ¹².

Although previous studies have demonstrated the importance of exosome biogenesis and secretion in injured cardiac tissue and initiation of different molecular mechanisms downstream of exosomes, there have been only a modest number of studies attempting a systemic inhibition of exosome biogenesis in models of cardiac injury and heart failure ^13-17. Although a lot of exosome inhibitors have been discovered to date, ironically studies have been largely limited to the use of GW4869 compound that targets neutral Sphingomyelinase 2 (nSMase2), as an exosomes inhibitor in CVD models. Recent studies have shown the promising role of Tipifarnib for exosome inhibition in cancer that was initially developed as a farnesyltransferase inhibitor (FTIs). Tipifarnib was investigated for the treatment of various solid tumor cancers, specifically, for HRAS mutant head and neck cancer as it interferes with the farnesylation of H-Ras, thereby retaining it in the cytosol and preventing its activity. However, some studies detected no change in Ras inhibition with Tipifarnib ^18-22. Tipifarnib targets three key enzymes; Alix, nSMase2 and Rab27a which are involved in exosome biogenesis, secretion, and cargo sorting. However, its potential role in models of cardiac injury and heart failure has not yet been tested.

Thus, for the first time to the best of our knowledge, we investigated the potential of Tipifarnib as an exosome biogenesis inhibitor in reducing systemic exosome secretion in a TAC pressure overload mouse model. We show that Tipifarnib not only significantly reduces TAC-induced increase in circulating exosomes but importantly this intervention significantly protects from TAC-induced cardiac hypertrophy, LV dysfunction and fibrosis. We also provide the first evidence for a positive correlation between exosomal miR331-5p and HOXC8 axis in cardiac fibroblast activation and fibrosis. Taken together, our study enhances understanding of the circulating EVs/exosomes in cardiac remodeling and preclinical evidence for tipifarnib as a potential treatment strategy for heart failure.

METHODS

Data availability.

All original data and materials used for this study are available from the corresponding authors upon reasonable request. The RNA sequencing data were deposited in the Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/projects/geo/) and are publicly available under GEO accession no. GSE266185.

A detailed description of all materials and methods can be found in the Online Supplemental Material. The data that support the findings of this study are available upon reasonable request.

Experimental animals and study design:

All animal experiments were conducted per institutional guidelines and were approved by the Institutional Animal Care and Use Committee of Temple University School of Medicine. Adult male C57BL/6J mice (8 weeks old) were purchased from Jackson Labs (Bar Harbor, ME) and randomly assigned to the experimental groups.

Transverse Aortic Constriction Murine Model of Heart Failure:

Transverse Aortic Constriction (TAC) was performed on 9-10 weeks old C57BL/6J male mice and mice were followed for 8 weeks as described previously ^23-25. For pre- and postoperative analgesia, mice were injected intraperitoneally with buprenorphine (0.08 mg/kg). During the surgical procedure, anesthetized mice (2% isoflurane, vol/vol) underwent partial thoracotomy and ligation of the transverse aorta using 7.0 nylon suture and 27G needle. In sham surgery, only the chest was opened, but no ligation of the aorta was performed. Cardiac dimensions and function were analyzed by two-dimensional (2-D) echocardiography before TAC/sham surgery and during the experiment, before euthanizing the animals.

Animal treatment:

For this study, mice were randomly divided into three groups: Sham, TAC, and Tipifarnib-treated TAC group. Tipifarnib (MedChemExpress; Cat. #HY-10502, USA) or GW4869 (MedChem Express, Cat. No.: HY-19363) were administered intraperitoneally (IP) in a 90% (20% Hydroxypropyl-β-Cyclodextrin in saline) and 10% DMSO vehicle as a volume of 0.10 ml of solution ²⁶. Control groups received the same dosage/volume in the 20% β-cyclodextrin vehicle. Sham mice were intraperitoneally (IP) injected with phosphate-buffered saline (PBS). Mice in the tipifarnib group were subjected to IP injection with tipifarnib (10 mg/kg body weight/three times a week).

In vivo exosome administration

In a separate experiment, C57BL/6J male mice were injected with 1-1.5× 10¹⁰ exosomes isolated from serum of TAC and sham mice through retro-orbital vein injections ²⁷. The blood for exosome isolation was collected by carotid artery puncture of the sham mice and TAC mice after 8 weeks of study. Then it was incubated on ice for 1 hour followed by centrifugation at 3,000g for 10 minutes and the supernatant was filtered through a 0.22 μm PES membrane filter unit (Millex^®-GP, Millipore, USA). The filtered serum was collected and stored at −80 °C until use. For isolating exosomes, 50 μl of filtered serum was diluted with an equal volume of PBS and exosomes were isolated using ExoQuick^® Exosome Isolation and RNA purification kit (System Biosciences, Palo Alto, CA) as per manufacturer’s instruction ²⁷.

Echocardiography

Echocardiography was performed as previously described, both before surgical intervention for baseline cardiac function, post-surgical for pressure gradient analysis and cardiac function analysis ⁴. Ultrasound gel was applied to the chest of isoflurane-anesthetized mice, and echocardiographic monitoring was performed by using Vevo 2100 (FUJIFILMS VisualSonics, Canada) equipped with 18-38 MHz transducer (MS400) according to a standard protocol, with a target heart rate of 450–500 bpm. The images were obtained in the parasternal long and short-axis views.

Following TAC, the peak pressure gradient was determined by application of pulsed-wave Doppler to the aortic arch, distal to the TAC constriction, as previously described ²³. Color Doppler imaging revealed either laminar blood flow in sham-operated mice or turbulent blood flow in TAC mice. A 13–24 MHz transducer (MS250) was used to detect blood velocities across the TAC constriction site. The pulsed-wave peaks were then measured and analyzed for peak pressure gradient. All M-mode and pressure peak images were analyzed using Vevo 2100 software (v3.1.1, FUJIFILM VisualSonics, Canada), with a minimum of three consecutive breaths analyzed.

Exosome Characterization

Blood was collected at stipulated time points by carotid artery puncture, to isolate exosomes from serum of different groups of mice. Serum exosomes were isolated by ExoQuick^® Exosome Isolation and RNA purification kit (System Biosciences, Palo Alto, CA) as mentioned earlier and were characterized using nanoparticle tracking analysis (NTA) for size and concentration. The appropriate volume of Exoquick Exosome Precipitation Solution (System Biosciences) was added to the filtered serum and mixed well by inverting. After 1 hour incubation on ice, the mixture was centrifuged at 1,500 g for 30 minutes and all supernatant was removed by aspiration. The sEVs/exosomes were resuspended in 1 ml of PBS. The quality of exosomes was examined using Exo-Check^™ Antibody Array (System Biosciences, Palo Alto, CA) following the manufacturer’s protocol to confirm the presence of exosome-specific marker proteins including CD63, CD81, ALIX, FLOT1, ICAM1, EpCam, ANXA5, and TSG101 in the isolated mice serum exosomes. The GM130 was used to confirm no cellular contamination in exosome samples.

RESULTS:

Tipifarnib effectively inhibits exosome biogenesis and reverses TAC-induced increase in circulating exosomes.

To establish the effectiveness of Tipifarnib as a potent systemic exosome inhibitor in TAC-induced HF, we first did a preliminary study to determine the effective dose of Tipifarnib in reducing exosome concentration in systemic circulation of mice without any significant changes in body weight during the period of administration (data not shown). This study established that a dosage of 10 mg/kg Tipifarnib effectively suppressed the exosome secretion in the systemic circulation of mice when intraperitoneally injected three times a week for 8 weeks (Figure S1A and S1B). Subsequently, we clarified whether Tipifarnib had any adverse effect on the cardiac function in control (non-TAC) mice. The 2-D echocardiography showed no significant difference between control mice with or without Tipifarnib treatment for the Left ventricular (LV) cardiac functions especially, Ejection fraction (EF%) and Fractional shortening (FS%) (Figure S1C and Supplementary Table 3). Further, Tipifarnib treatment did not affect either the heart weight/body weight (HW/BW) ratio or cardiomyocyte cross-section area (Figure S1D, S1E and S1F).

We next sought to further elucidate whether Tipifarnib at 10mg/kg dosage inhibits exosome secretion in the setting of cardiac pressure overload. The mice were subjected to TAC surgery and the pressure gradient analysis was performed on Day 7. Mice that developed a pressure gradient >60% were divided into two groups for no treatment and treatment of Tipifarnib (Figure 1A). Induction of TAC led to a significant surge of sEVs/exosomes number in the systemic circulation. Strikingly, the treatment of Tipifarnib in TAC mice led to a remarkable reduction in the circulating exosome number that was comparable to the level of sham mice (Figure 1B). The decrease of exosomes in circulation was concomitant with a significant reduction in the expression of Rab27a, Alix (ESCRT-dependent exosome biogenesis pathway proteins) and nSMase2 (ESCRT-independent pathway) proteins involved in exosome biogenesis in the cardiac tissue of Tipifarnib treated TAC mice (TAC + Tipifarnib mice) as opposed to the TAC mice alone (Figure 1C and 1D). Likewise, there was a significant reduction of TSG101, an exosome marker protein although there was no statistical difference in CD63 expression in cardiac tissue (Figure S2A through S2D).

Figure 1: Tipifarnib is effective in reducing exosome biogenesis and secretion in TAC mice.

(A) Schematic presentation on the timeline of the Tipifarnib study in 9–10-week-old C57BL/6J male TAC mice (n=10/group). (B) Nanoparticle tracking analysis (NTA) for exosome concentration in serum of sham, TAC and Tipifarnib treated TAC (TAC+Tipifarnib) mice (n=10/group). (C) Representative image of the western blotting analyses of protein extracted from mouse hearts to study the effect of Tipifarnib treatment on the expression of proteins involved in exosome biogenesis. GAPDH served as the loading control. (D) Mean ± SEM of western blotting experiment (n=6/group). Data are expressed as Mean ± SEM. Mean of the different groups was compared using one-way ANOVA followed by Tukey post-tests with 95% confidence interval.

Further, the particle size of exosomes isolated from mice serum of all three groups showed no significant differences and overall were within the normal range of typical size of exosomes (Figure S3A and S3C). The characteristics of serum-derived exosomes were further verified by positive detection of several exosome protein markers (Figure S3B). Taken together, these data indicate the effectiveness of Tipifarnib in reducing circulating exosomes and their biogenesis especially in heart during TAC-induced cardiac injury.

Inhibition of circulating exosomes by Tipifarnib leads to improved post-TAC LV function and reduction in cardiac hypertrophy and fibrosis.

Next, we evaluated whether inhibition of exosome biogenesis and secretion by Tipifarnib treatment can promote a protective effect against TAC-induced cardiac dysfunction. 2-D echocardiography (echo) was used to assess changes in LV cardiac functions (Figure 2A and 2B; Supplementary Table 1 and Table 2) starting with the baseline study of cardiac functions that showed no significant difference in the various groups. The TAC mice developed a significantly lower ejection fraction (LVEF), fractional shortening (LVFS) and LV stroke volume (SV) with an accelerated development of LV wall thickness by 8 weeks post-TAC compared to sham. Interestingly, there was a remarkable improvement in the LV cardiac functions in the Tipifarnib-treated TAC mice with enhanced LVEF, LVFS and SV by 4 weeks and continued to show better preserved cardiac functions until 8 weeks with reduced LV wall thickness (Figure 2A, Supplementary Table 2). Strikingly, the Tipifarnib-treated TAC mice showed a marked reduction in cardiac hypertrophy and significantly reduced HW/BW and heart weight/ tibia length (HW/TL) ratios in comparison to TAC mice as shown in Figures 2C and 2D. Importantly, to justify the effectiveness of Tipifarnib over GW4869 compound as an exosome inhibitor, we injected 2 mg/kg dosage of GW4896 to TAC mice and monitored them for 4 weeks. As anticipated, GW4869 was effective in reducing the serum exosome concentration (Figure S4A). However, our data indicates that blockade of exosome generation in the serum of TAC mice with GW4869 had no significant improvement in their cardiac functions (Figure S4B, S4C and Supplementary Table 3). Next, we wanted to understand the physiological basis of improvements in cardiac function with Tipifarnib treatment by assessing the pathologic hallmarks of pressure overload-induced HF — hypertrophy and fibrosis. Treatment of Tipifarnib exhibited remarkable protection against TAC-induced cardiomyocyte hypertrophy and showed a significant decrease in the cardiomyocyte cross-sectional area and the expressions of hypertrophy-associated fetal cardiomyocyte genes ANP, BNP and MYH7 (Figure 3A, 3B and 3C). Furthermore, we found that Tipifarnib treatment in TAC mice markedly decreased the extracellular matrix deposition (Figure 3D and 3E) and significantly inhibited the mRNA expression of fibrotic genes in cardiac tissue compared to TAC mice (Figure 3F). We additionally evaluated the protein expression of both farnesyltransferase β (FTase β) subunit and Ras in the cardiac tissues of Tipifarnib-treated TAC mice and found that there was no significant difference in the protein expression of these proteins (Figure S5)

Figure 2: Treatment with Tipifarnib improves cardiac function.

(A) M-Mode 2D-echocardiographic measurements at baseline/0 weeks, 4 weeks and 8 weeks to determine Left Ventricular (LV) functions i.e., ejection fraction (EF), fractional shortening (FS), LV mass, LV internal dimension in diastole (LVID;d), LV internal dimension in systole (LVIDd) and stroke volume (SV) (n=10/group). Data are expressed as Mean ± SD. Mean of three groups was compared using one-way ANOVA, followed by Tukey post-tests with 95% confidence interval. ^a1Sham vs TAC; ^b1TAC vs TAC+Tipifarnib and ^c1Sham vs TAC+Tipifarnib for 4 weeks and ^a2Sham vs TAC; ^b2TAC vs TAC+Tipifarnib and ^c2Sham vs TAC+Tipifarnib for 8 weeks. (B) Representative echocardiographic images (M-mode) from sham, TAC and Tipifarnib treated TAC mouse heart at 8 weeks. (C) Post TAC, heart-to-body weight ratio (HW/BW) and Heart weight (HW) to Tibia length (TL) (gm/cm) ratio at 8 weeks for sham, TAC and TAC+Tipifarnib. Data are expressed as Mean ± SEM. Mean of the different groups was compared using one-way ANOVA followed by Tukey post-tests. with 95% confidence interval. (D) Images of sham, TAC and TAC+Tipifarnib mouse hearts post 8 weeks. (Scale, in cm.)

Figure 3: Tipifarnib treatment reduces hypertrophy and fibrosis in cardiac tissue of pressure overload HF mice.

(A) Representative image of the immunofluorescence of myocardial sections stained with Alexa fluor 488-labeled wheat germ agglutinin (WGA) antibody, defining myocyte boundaries. Scale bar = 50 μm). (B) Mean ± SEM of cardiomyocytes≈50 myocytes per mouse in each experimental group; n=10/group. Data are expressed as Mean ± SEM. Mean of sham, TAC and Tipifarnib-treated TAC (TAC+Tipifarnib) mice groups was compared using one-way ANOVA followed by Tukey post-tests with 95% confidence interval. (C) Quantitative Reverse Transcription Polymerase Chain Reaction (qRT PCR) of mice heart tissues for hypertrophy–related genes (ANP, BNP and MYH7) in each experimental group after 8 weeks post-TAC; n=7/group. (D) Representative image for Masson’s trichrome staining in the heart tissues of sham, TAC and Tipifarnib treated TAC (TAC+Tipifarnib) Scale bar = 50 μm. (E) Representative coimmunostaining images for Fibronectin and α-SMA in mice cardiac tissue. Immunofluorescence staining was performed using anti-Fibronectin antibody (green), anti-α-SMA antibody (red) and DAPI (blue). Scale bar = 25 μm. (F) qRT PCR of mice heart tissues for fibrosis-related gene markers (ACTA2, COL1A1, POSTN and TGFB1) in each experimental group; n=7/group. Data are expressed as Mean ± SEM. Mean of the different groups was compared using one-way ANOVA followed by Tukey post-tests with 95% confidence interval. ANP: Natriuretic peptide A, BNP: Natriuretic peptide B, MYH7: myosin heavy chain 7, α-SMA/ACTA2: alpha-smooth muscle actin, POSTN: Periostin, COL1A1: Collagen, Type I, TGFB1: Transforming Growth Factor beta1.

Since systemic and local inflammation has been consistently reported in various pre-clinical and clinical studies as a strong regulator for myocardial remodeling and HF ^25,28-30, we evaluated the protein expression of circulating proinflammatory cytokines in the serum of mice. We found that compared to TAC mice, treatment with Tipifarnib showed a significant reduction in the levels of pro-inflammatory cytokines in the systemic circulation (Figure S6A). Further, the qRT PCR analysis of mice heart tissue showed decreased expression of IL 1B and IL 6 with Tipifarnib treatment in comparison to non-treated TAC mice (Figure S6B).

Together these findings suggest, that Tipifarnib is highly effective in reducing exosome biogenesis and secretion that helps in preserving cardiac function by preventing hypertrophy, fibrosis, and inflammation during HF.

Circulating plasma exosomes from TAC and Tipifarnib-treated TAC mice differentially affect cardiac fibroblasts and cardiomyocytes.

We next evaluated if improved cardiac remodeling and function by Tipifarnib are due to a reduction in exosome secretion or whether Tipifarnib also alters the exosome cargo since two of the Tipifarnib target proteins, Rab27a and Alix are also involved in exosome cargo sorting. To address this question, we used equal concentrations of circulating exosomes from TAC and TAC+Tipifarnib mice to treat adult cardiac fibroblasts (CF) and neonatal rat ventricular myocytes (NRVMs), in vitro. CFs and NRVMs were incubated with equal particle numbers of serum exosomes from TAC or TAC+Tipifarnib mice for 48 hours. The addition of TAC exosomes to CFs led to significant upregulation of the mRNA expression of myofibroblast genes (Figure 4A). Similarly, immunofluorescence studies showed increased expression of myofibroblast marker proteins, α-SMA (in red) in response to TAC exosomes (Figure 4B) implying that TAC exosomes induce myofibroblast phenotype in CFs. Interestingly, circulating exosomes from TAC mice treated with Tipifarnib significantly reduced the expression of these markers (Figure 4A, 4B). Likewise, TAC exosome treatment led to significant upregulation of the ANP and MHY7 genes in NRVMs, which was not observed in cells treated with TAC+ Tipifarnib circulating exosomes (Figure 4C). These data suggest that Tipifarnib not only reduces exosome biogenesis but also likely alters the cargo of the circulating exosomes.

FIGURE 4: Effect of circulating exosomes in cardiomyocytes (CMs) and cardiac fibroblasts (CFs).

Quantitative reverse transcription polymerase chain reaction (qRT PCR) analysis for the expression of the (A) Fibrosis gene markers (ACTA2, COL1A1 and POSTN) (B) Representative immunostaining images for the expression of α-SMA in mouse primary cardiac fibroblast cells. Immunofluorescence staining was performed using anti- α-SMA antibody (red). DAPI was used to stain cell nuclei (blue). Scale bar = 50 μm. (C) Hypertrophy gene markers (ANP, BNP and MYH7) after treatment of serum exosomes derived from sham, TAC and Tipifarnib-treated TAC (TAC+Tipifarnib) mice; n=3/group. Data are expressed as Mean ± SEM. Mean of the different groups was compared using one -way ANOVA followed by Tukey post-tests with 95% confidence interval. α-SMA/ACTA2: alpha-smooth muscle actin, POSTN: Periostin, COL1A1: Collagen, Type I, ANP: Natriuretic peptide A, BNP: Natriuretic peptide B, MYH7: myosin heavy chain 7.

Intravenous injection of circulating exosomes from TAC mice induces cardiac hypertrophy and dysfunction in normal uninjured mice.

The above in vitro studies indicated that Tipifarnib not only inhibits exosome synthesis but also modulates exosome cargo contents. To evaluate if exosomes from TAC mice potentially influence its dysfunctional cargo, fostering fibrosis and hypertrophy in cardiac tissue, we isolated exosomes from the serum of TAC mice and sham mice (as control) and injected them intravenously into the uninjured healthy/control mice without TAC surgery as shown in the experimental scheme in Figure 5A. We continued injecting the TAC or sham exosomes three times a week for 8 weeks. After 8 weeks, we performed echocardiography and euthanized the mice for further analysis. The gross morphology analysis showed an increase in the HW of mice injected with TAC exosomes and echo analysis revealed a significant reduction in LVEF and LVFS with the TAC exosomes administration in the absence of TAC whereas sham exosomes did not affect these parameters in control mice (Figure 5B, 5C and Supplementary Table 4). In addition, there was a significant increase in CM size in mice administered with TAC exosomes as evaluated by WGA staining of heart tissue sections (Figure 5D, E). Although we found some degree of perivascular fibrosis in mice injected with TAC exosomes, we observed no significant effect of TAC exosomes on cardiac fibrosis in these control mice. (Figure S7A).

FIGURE 5: Effect of circulating TAC exosomes on cardiac function and structure in control mice.

(A) Schematic presentation on the timeline of the sham or TAC exos (exosomes) study in C57BL/6J male mice. (B) Heart to Body weight ratio (HW/BW) and Heart weight (HW) to Tibia length (TL) (gm/cm) ratio at 8 weeks for sham, TAC and TAC+Tipifarnib. Data are expressed as Mean ± SEM. Mean of the different groups was compared using one-way ANOVA followed by Tukey post-tests with 95% confidence interval. (C) M-Mode echocardiographic measurements of Left Ventricular (LV) functions i.e., Ejection fraction (EF) and Fractional Shortening (FS) after 8 weeks of exosomes injection in mice; n=5/group. Data are expressed as Mean ± SEM. Mean of the Control, sham exos and TAC exos injected Control mice groups were compared using one-way ANOVA followed by Tukey post-tests with 95% confidence interval. (D) Representative image of the immunofluorescence of thin myocardial sections stained with Alexa fluor 488-labeled wheat germ agglutinin (WGA) antibody, defining myocyte boundaries. Scale bar = 100 μm. (E) Mean ± SEM of cardiomyocytes≈25 myocytes per mouse in each experimental group; n=5/group. Data are expressed as Mean ± SEM. Mean of the Control, sham exos and TAC exos mice groups was compared using one-way ANOVA followed by Tukey post-tests with 95% confidence interval. (F) Representative image of the immunofluorescence of thin myocardial sections of mice receiving an intravenous injection of PKH26-labeled TAC exosomes stained with α-Actinin (green) and DAPI (blue) receiving PKH26-labeled TAC exosomes (red). Scale bar = 50 μm.

To examine the biodistribution of the injected TAC exosomes, specifically in the heart tissue, we labeled the serum TAC exosomes with PKH26 dye and retro-orbitally injected into the control mice and 24 hours post-injection, the mouse hearts were harvested and analyzed by immunofluorescence. We observed the presence of TAC exosomes in cardiac tissue, as shown by red fluorescence signals. Moreover, these exosomes were present in the CMs (Figure 5F).

Tipifarnib treatment upregulates miR-331-5p expression that inhibits cardiac fibroblast activation to myofibroblast.

To further investigate the mechanism underlying the deteriorating effects of serum TAC exosomes in pressure-overload induced HF mice, we analyzed the miRNA transcriptomic profile of circulating TAC and Tipifarnib-treated TAC exosomes using a small RNA sequencing platform.

The RNA sequencing data were deposited in the Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/projects/geo/) and are publicly available under GEO accession no. GSE266185. Figure 6A displays the expression changes of miRNAs as the Differentially Expressed miRs (DEmiRs) by the volcano plot determined by Small RNA sequence analysis of the serum exosomes of TAC versus (vs) Tipifarnib treated TAC mice. The correlation plot, heat maps along with gene enrichment and gene ontology analysis for small RNA seq. of TAC vs Sham and TAC vs Tipifarnib treated TAC serum exosomes are shown in Figures S8 and S9. The green spots in the volcano plot represent downregulated miRNAs, and the red spot represents the upregulated miRNAs (Figure 6A). Expression of the topmost differentially expressed miRNAs was further confirmed by qRT PCR (data not shown). miR 331-5p was found to be most significantly upregulated both in the serum exosomes (Figure 6B) and the cardiac tissue (Figure 6C) of Tipifarnib-treated TAC mice compared to untreated TAC mice. Since Tipifarnib treatment resulted in a remarkable reduction in TAC-induced cardiac fibrosis and TGFβ1 expression (Figure 3), we mainly focused on the function of miR 331-5p in regulating cardiac fibrosis through the TGFβ signaling pathway ^31,32. To investigate whether cardiac miRNA 331-5p acts as one of the drivers of myofibroblast activation, we transfected miR 331-5p mimic and its antagomir to overexpress and inhibit miR 331-5p respectively, in primary adult murine cardiac fibroblast (CF) cells (Figure 6D). The overexpression of miR 331-5p in CF cells significantly downregulated TGFβ1-induced expression of myofibroblast genes ACTA2 and POSTN. Additionally, treatment of TGFβ1 to these cells prevented the upregulation of ACTA2 and POSTN levels supporting the validity of our approach that TAC+Tip exosomes enriched with miR 331-5p limit myofibroblast activation by suppressing TGFβ signaling (Figure 6E). Altogether, our data suggest that miR 331-5p gets highly expressed in the exosomes during Tipifarnib treatment in TAC mice and it exerts protection against fibrosis during HF.

FIGURE 6: Transcriptomic analysis of serum-derived TAC and Tipifarnib-treated TAC mice exosomes.

(A) Volcano plot analysis of the differential expression miRNAs between Tipifarnib treated TAC (TAC+Tip) mice and TAC mice with a comparison of fold change (log2) and significance (-log10 P-value). Significance is determined as adj-p<0.01. Quantitative reverse transcription polymerase chain reaction (qRT PCR) analysis of miR 331-5p in (B) Serum exosomes and (C) Heart tissue of sham, TAC and Tipifarnib-treated TAC (TAC+Tip) mice after 8 weeks post TAC surgery by (n=6/group). (D) Adult primary mouse cardiac fibroblasts (CFs) were transfected with miR 331-5p antagomir (inhibitor), or miR 331-5p mimic to quantify the differential miR 331-5p transfer. The cycle of threshold values was normalized to RNU6B. Data are expressed as Mean ± SEM. Mean of the different groups was compared using one-way ANOVA followed by Tukey post-tests (for >2 groups) with 95% confidence interval. (E) The qRT PCR analysis of the expression of fibrotic marker genes: ACTA2 and POSTN in CFs upon TGFβ1 treatment (10 ng/ml), transfection of miR 331-5p with or without TGFβ1 (10 ng/ml) treatment. Data are expressed as Mean ± SEM. Mean of the untreated, TGF β1, miR mimic control, miR 331-5p mimic ± TGFβ1 was compared using one-way ANOVA followed by Tukey post-tests with 95% confidence interval. ACTA2: alpha-Smooth Muscle Actin, POSTN: Periostin, TGFβ1: Transforming Growth Factor β1.

miR 331-5p targets HOXC8 and specifically represses myofibroblast transformation.

To evaluate the downstream effector molecules of TGFβ signaling pathway targeted by miR 331-5p, we specifically did an in-silico study for its putative interactors using miR databases such as miRBase and TargetScanMouse. Among various gene targets, HOXC8 emerged as a promising candidate, given that distinct species showed conserved regions within the HOXC8 gene that interact with the miR 331-5p (Figure 7A). To date, the regulatory mechanism by which endogenous cardiac HOXC8 signaling works in HF is poorly understood. Upon assessment of the heart tissue at mRNA and protein levels of sham, TAC and Tipifarnib treated TAC (TAC+Tip) mice, we observed elevated levels of HOXC8 in TAC mice compared to sham and TAC-Tip mice (Figure 7B, 7C and 7D). Further, the expression of HOXC8 in miR 331-5p mimic and antagomir-transfected CF cells was verified. The overexpression of miR 331-5p in CF cells remarkably reduced HOXC8 and a similar trend was maintained in cells even after TGF β treatment. This indicated miR 331-5p inhibits HOXC8 activity during TGFβ-mediated stress in cardiac fibroblasts. Notably, inhibition of miR 331-5p in CFs led to upregulation of HOXC8 expression which was suppressed by subsequent treatment with TAC-Tip exosomes (Figure 7E). Collectively, these findings confirm the role of HOXC8 being regulated by miR 331-5p in cardiac fibroblasts.

FIGURE 7: miR 331-5p regulates HOXC8 to prevent fibrosis.

(A) Bioinformatics algorithms predicting miR 331-5p binding sites in 3′-UTR or coding regions (CDS) across different species for HOXC8 gene. (B) The quantitative reverse transcription polymerase chain reaction (qRT PCR) analysis of the expression of HOXC8 in heart tissue of sham, TAC and Tipifarnib-treated TAC (TAC+Tip) mice at 8 weeks, n=7/group. (C-D) Representative image of the protein expression levels of HOXC8 and GAPDH in HOXC8 in heart tissue of sham, TAC and TAC+Tipifarnib mice at 8 weeks and bar graph is the quantitative analysis of HOXC8 protein levels after normalization with GAPDH, n=5/group. Data are expressed as Mean ± SEM. Mean of the sham, TAC and Tipifarnib-treated TAC (TAC+Tip) mice was compared using one-way ANOVA followed by Tukey post-tests (for >2 groups) with 95% confidence interval. (E) The qRT PCR analysis of the expression of HOXC8 in cardiac fibroblast cells after transfection of miR 331-5p with or without TGFβ1 (10 ng/ml) treatment and after transfection of miR 331-5p antagomir with and without treatment of Tipifarnib-treated TAC exosomes (TAC+Tip). Data are expressed as Mean ± SEM. Mean of the untreated, miR mimic control, miR 331-5p mimic ± TGFβ1, miR antagomir control and miR 331-5p antagomir ± Tipifarnib treated TAC exosomes were compared using one-way ANOVA followed by Tukey post-tests (for >2 groups) with 95% confidence interval. HOXC8: Homeobox C8, TGF β1: Transforming Growth Factor β1.

To further assess whether miR 331-5p targeted HOXC8 is directly regulated in TGFβ signaling pathway, we overexpressed (Figure 8A) and knocked down HOXC8 (Figure 8C) in CFs and examined the expression of the fibrotic genes ACTA2 and POSTN in comparison to TGFβ treatment (Figure 8B and 8D). As expected, TGFβ treatment led to the upregulation of these genes. Interestingly, overexpression of HOXC8 also significantly upregulated ACTA2 and POSTN genes indicating that HOXC8 is one of the critical mediators for TGFβ signaling during cardiac fibrosis activation (Figure 8B). Importantly, TGFβ treatment in HOXC8 knock-down CF cells (Figure 8C) prevented the upregulation of ACTA2 and POSTN genes (Figure 8D) even in the presence of TGFβ1. Collectively, these findings indicate a strong correlation between expression of HOXC8 and miR 331-5p in cardiac fibrosis for regulating myofibroblast activation mediated by TGFβ signaling pathway.

FIGURE 8: Silencing HOXC8 in cardiac fibroblast (CF) cells reduced fibroblast activation towards myofibroblast formation.

The quantitative reverse transcription polymerase chain reaction (qRT PCR) analysis of (A) overexpression of HOXC8 in CFs. (B) Fibrotic marker genes ACTA2 and POSTN expression after TGF β1 (10 ng/ml) treatment and overexpression of HOXC8, n=3/group. (C) Knockdown of HOXC8 in CFs. (D) Fibrotic marker genes ACTA2 and POSTN expression after TGF β1 (10 ng/ml) treatment and TGFβ1 in the HOXC8-knockdown cells, n=3/group. Data are expressed as Mean ± SEM. Mean of different groups was compared using one-way ANOVA followed by Tukey post-tests (for >2 groups) with 95% confidence interval. ACTA2: alpha-Smooth muscle actin, HOXC8: Homeobox C8, POSTN: Periostin, TGF β1: Transforming Growth Factor β1.

Transfections of miR 331-5p mimic and antagomirs were also performed in NRVMs to evaluate their role in CM hypertrophy. miR 331-5p mimic and antagomir treatment showed moderate downregulation and upregulation of HOXC8 in CMs respectively (Figure S10). In addition, the overexpression of HOXC8 showed enhanced expression of hypertrophy-related genes while knockdown of HOXC8 decreased the same in CMs (Figure S11). Together these findings support the data demonstrating the reduction and cargo alteration in exosomes by Tipifarnib significantly prevents TAC-induced cardiac remodeling and progression to HF.

DISCUSSION

Cardiovascular function depends on the synchronized interplay and interaction between cardiomyocytes and noncardiomyocytes wherein exosomes act as a critical mediator for cell-to-cell communications ^{5,8,14-16,33-35}. The exosomes derived from multi-vesicular bodies (MVB) bud directly from the cell membrane by two distinct mechanisms, i.e., endosomal sorting complexes required for transport machinery (ESCRT)-dependent and ESCRT-independent pathways involving ESCRT components (e.g., Alix and syntenin) vesicle transport molecules (e.g., Rab27a and Rab27b), and ceramide metabolism proteins (e.g., neutral sphingomyelinase and sphingosine1-phosphate receptor) ^11,36. Several studies support the fact that during cardiac stress/injury, there is a marked increase in exosome biogenesis in the heart tissue that aggravates fibrosis ^34,37. Moreover, it has been observed that the enrichment of cardiac-specific exosomes as well as exosomes from other distant organs in circulation during HF promotes adverse cardiac remodeling and proinflammatory responses ^16,38,39. Thus, it is important to emphasize sEVs/exosomes as crucial regulators of pathophysiological processes leading to HF. We observed a significant increase in systemically circulating exosomes as well as enhanced expression of Rab 27a, Alix and nSMase2 in heart tissue of TAC mice suggesting that exosome biogenesis and release is indeed upregulated during HF (Figure 1C). Most importantly, this necessitates the relevance of exosome inhibition for treating CVDs, particularly, identifying and developing strategies to reduce sEVs/exosome secretion and biogenesis during adverse cardiac remodeling.

Studies have shown an improvement in cardiac function due to inhibition of sEVs/ exosomes in mice through genetically and chemically targeting EV biogenesis pathway during cardiac injury. Previously, Rab27a KO mice showed better survival and recovery during MI due to reduced exosome biogenesis and release ⁴⁰. Similarly, treatment with GW4869 compound inhibits EVs secretion in ischemic reperfusion (IR) injured mice and prevents cardiac damage ²⁹. Although GW4869 has been extensively studied in cardiac injury animal models for exosome inhibition, this drug only targets a single protein involved in exosome release and does not inhibit other proteins that are involved in a complex process of exosome biogenesis and cargo sorting. Therefore, many new compounds or drugs are being constantly experimented on for a better translational approach as exosome inhibitors. The present study using Tipifarnib that targets multiple proteins responsible for the exosome biogenesis, cargo sorting and release is quite novel in the field the CVDs. Tipifarnib was shown as a potent exosome biogenesis inhibitor by inhibiting Rab27a, Alix and nSMase2 which are responsible for cargo sorting and exosome secretion in cancer cell lines ^41,42. Besides, the preclinical findings by Liu et al demonstrated a novel mechanism in which tipifarnib-mediated inhibition of sEV secretion served as a possible treatment strategy to enhance the antitumor efficacy of anti-GD2 immunotherapy for neuroblastoma patients ²⁶. For the first time, we demonstrated that Tipifarnib can not only successfully inhibit TAC-induced surge in circulating exosomes, but also that this intervention significantly protects mice from TAC-induced cardiac fibrosis, hypertrophy, and heart failure. The decrease of circulating exosomes and its biogenesis in the heart subsequently reduced the delivery of TAC-induced specific exosome cargo enrichment and cell-to-cell crosstalk in cardiac tissue and decreased the hypertrophy, fibrosis and inflammation during cardiac injury leading to a remarkable improvement in LV function.

Interestingly, the circulating exosomes during cardiac injury or stress have a distinct cargo content along with enhanced secretion that contributes to worsening of cardiac remodeling. Various clinical studies in HF patients showed the patients with higher circulating EVs had distinct EVs from healthy individuals with enriched proteins or miRNA content that can be diagnosed as biomarkers specific to cardiac origin ^43,44. Studies also showed individuals with diabetes had significantly higher levels of EVs in their circulation with the modified levels of insulin signaling and inflammatory protein cargo that functionally alters endothelial cells ⁴⁵. Our study also attempted to address this key question of whether the circulating exosomes from TAC mice are enriched in specific cargo that can be detrimental. To our knowledge, we for the first time demonstrated that circulating exosomes are a critical determinant of cardiac damage in a heart failure model by intravenously injecting the serum exosomes of TAC mice into the healthy uninjured mice. Intriguingly, infusion of circulating exosomes from TAC mice caused depression in cardiac functions and CM hypertrophy in otherwise healthy mice. In vitro studies showed the TAC exosomes enhanced myofibroblast activation and increased hypertrophy in CM. The current study indicated that circulating exosomes with their modulated cargo content contribute to promoting adverse cardiac remodeling.

Coincidently, inhibiting exosome secretion by Tipifarnib also resulted in the altered miRNA packaging in circulating exosomes since Tipifarnib targets Alix and Rab27a/b proteins that are also involved in miRNA packaging ^46,47. Prior studies support this notion. For example, Ostenfeld et al. showed Rab27A/B knockdown in mice with bladder cancer attenuated exosome secretion and reorganized the miRNAs content that further promoted metastasis ⁴⁷. Our small RNA sequencing analysis specifically for miRNAs of serum exosomes showed miR 331-5p as the most significantly downregulated in TAC mice in comparison to Tipifarnib-treated TAC and sham mice. The miR-331 family constitutes a group of three miRNAs (miR-331, miR-331-3p and miR-331-5p) ^48,49. Normal human heart tissue highly expresses miR 331-5p and during stress/injury it gets reduced/inhibited as evident from TAC mice in our study ⁴⁹. Our studies demonstrate that miR-331-5p plays a key role in inhibiting TGF β1-mediated signaling pathway. During pulmonary arterial hypertension, miR-331-5p inhibits Phosphofructokinase (PFKP) that activates BMP2-PPARγ axis which further leads to inhibiting both canonical and non-canonical TGFβ1 signaling pathway³¹. Similarly, miR-331 attenuated isoproterenol-induced cardiac fibrosis by regulating TGFβ signaling pathway in cardiac myofibroblasts ³². These findings encouraged us to focus on the mechanism by which miR 331-5p regulates fibrosis mainly through TGFβ signaling pathway in HF. We observed the overexpression of miR 331-5p led to a marked reduction of TGFβ1-induced myofibroblast gene activation in CF cells. Our data firmly establishes the significance of miR 331-5p as an anti-fibrotic miRNA that gets highly expressed with the exosome biogenesis inhibition by Tipifarnib during cardiac damage. In addition, for the first time, we confirmed miR 331-5p suppresses transcription factor HOXC8, a member of the homeobox gene family in HF mice model. HOXC8 is one of the master regulators of embryonic development along with different physiological processes, like cell proliferation and differentiation, etc. ^50-52. Previous studies have documented the upregulation of HOXC8 in solid tumors that mediates TGFβ/SMAD2/SMAD3 signaling-induced fibrosis and migration ^53-56. However, the role of HOXC8 in adult cardiac pathologies has not been well-studied. However, during embryonic stage, HOXC8 in cardiac progenitor cells (CPCs) co-expresses with cardiac transcription factor Tbx3/4 and Wnt5a associated with cardiac development ⁵⁷. Specifically, we observed that HOXC8 is highly expressed during cardiac injury after TAC in mice and the treatment of Tipifarnib significantly reduced it. Besides, TGFβ treatment of cardiac fibroblasts significantly upregulated HOXC8 expression. Interestingly, in adult cardiac fibroblasts, overexpression of HOXC8 mimicked the effect of TGFβ on induction of myofibroblast gene program while knockdown of HOXC8 reduced TGFβ-induced myofibroblast gene activation. We also studied the role of miR 331-5p interaction with HOXC8 in the context of CM hypertrophic response induced by Isoproterenol. There was a reduction in hypertrophy marker genes, but it was not significant enough to investigate further (Figure S10-S11) and our focus was mainly on fibrosis and fibroblast.

Collectively, our data indicate that Tipifarnib effectively decreases the elevated exosome biogenesis and secretion in the pressure overload HF model and reduces enhanced fibrosis induced by HOXC8 due to miR 331-5p enrichment in the modified exosomes. Altogether, our findings provide both a novel method to inhibit circulating exosomes and insights into the importance of targeting the exosome-biogenesis pathway for the prevention of HF using Tipifarnib. Further investigations are required to explore the potential of Tipifarnib as an efficient exosome inhibitor in other CVDs and the potential of this drug as a therapeutic modality in clinical settings.

Novelty and Significance.

What is Known?

• Cardiovascular diseases (CVDs), particularly heart failure (HF) accelerate small extracellular vesicles (sEVs) secretion in the circulation leading to the propagation of signaling molecules such as proteins, small non-coding RNA (miRNAs, lncRNA), and cytokine passage between donor, recipient cells. Their cargo can potentially serve as biomarkers for disease.

• Current therapeutic strategies mainly focus on the inhibition of fibrosis, immunomodulation, and cardiomyocyte proliferation. However, mechanisms targeting exosome biogenesis, cargo sorting, release and/or uptake may provide a novel pathway for targeting maladaptive cell-cell communications during HF.

• The compound GW4869 has been extensively used as an exosome inhibitor but has not yet been used as a therapeutic drug candidate for the treatment of CVD. New drug candidates have been tested for effectiveness in targeting sEVs, especially in cancer. Tipifarnib has been developed as a farnesyltransferase (FT) inhibitor for the treatment of solid tumors by inhibiting the function of the FT substrate H-Ras.

What New Information Does This Article Contribute?

• We provide evidence that exosome secretion and exosome contents are important for mediating the pathogenesis of HF and reducing exosome biogenesis and secretion in circulation.

• Our study provides promising evidence that Tipifarnib efficiently inhibits exosome biogenesis and exosomes release in systemic circulation in a pressure-overload HF model, in vivo, and this could play a role in cardiac remodeling during HF.

• Treatment of Tipifarnib inhibits exosome biogenesis proteins Rab27a, nSMase2 and Alix, which are also involved in exosome cargo sorting. This mechanism enforces major reshuffling of the miRNAs which prevents fibrosis progression in HF.

• There are multiple studies supporting the clinical safety and efficacy of Tipifarnib during multiple clinical trials on solid tumors. Thus, repurposing this drug as a therapeutic agent has possible potential for treating HF.

Emerging evidence from literature suggest that extracellular vesicle (EV) including small EVs known as exosomes play important roles in cell to cell communication in cardiac homeostasis and disease. The number of circulating EVs has been documented in number of cardiovascular diseases including myocardial infarction and heart failure. However, whether EVs play a pathological role in the progression of heart failure or whether inhibition of systemic EV secretion may protect from heart failure is not well established. We provide evidence that Tipifarnib, a farnesyl transferase inhibitor, targets multiple genes involved in EV biogenesis and cargo sorting and significantly reduces circulating EVs and alters their microRNA contents, including that of microRNA 331-5p. Reduction of EVs and alteration of microRNA 331-5p in EVs by Tipifarnib protects heart failure in a mouse model of pressure overload induced heart failure. Our findings support a novel mechanisms indicating miR331-5p in inhibiting fibroblast-to-myofibroblast transition by targeting HOXC8, a critical regulator of fibrosis.

NONSTANDARD ABBREVIATIONS AND ACRONYMS:

ACTA2

Actin Alpha 2

ALIX

ALG-2-interacting protein X

ANP

Atrial Natriuretic Peptide

ANXA5

Annexin A5

BMP2

Bone Morphogenetic Protein 2

BNP

Brain Natriuretic Peptide

CD63

Cluster of Differentiation 63

CD81

Cluster of Differentiation 81

CF

Cardiac Fibroblasts

CM

Cardiomyocytes

COL1A1

Collagen Type I Alpha 1 Chain

CPC

Cardiac Progenitor Cell

EF

Ejection Fraction

ELISA

Enzyme-Linked Immunosorbent Assay

ESCRT

Endosomal Sorting Complex Required for Transport

EV

Extracellular Vesicle

EpCam

Epithelial Cell Adhesion Molecule

FS

Fractional Shortening

FT

Farnesyltransferase

FTI

Farnesyltransferase Inhibitor

FTase β

Farnesyltransferase Beta

FLOT1

Flotillin-1

GM130

Golgi matrix protein 130

HF

Heart Failure

HCM

Hypertrophic Cardiomyopathy

HOXC8

Homeobox C8

H-Ras

Harvey Rat Sarcoma Viral Oncogene Homolog

ICAM1

Intercellular Adhesion Molecule 1

IL1B

Interleukin 1 Beta

IL6

Interleukin 6

IR

Ischemia Reperfusion

LV

Left Ventricle

LVEF

Left Ventricular Ejection Fraction

LVFS

Left Ventricular Fractional Shortening

MI

Myocardial Infarction

MVB

Multi-Vesicular Bodies

MYH7

Myosin Heavy Chain 7

NRVM

Neonatal Rat Ventricular Myocyte

PFKP

Phosphofructokinase

PPARγ

Peroxisome Proliferator-Activated Receptor Gamma

POSTN

Periostin

RNU6-2

RNA, U6 small nuclear 2

SV

Stroke Volume

TAC

Transverse Aortic Constriction

TGFβ

Transforming Growth Factor Beta

TSG101

Tumor Susceptibility Gene 101

Tip

Tipifarnib

WGA

Wheat Germ Agglutinin