*3.5. GLA-Null CMs Secrete More Exosomes Than H9 CMs*

Given the fact that the proteome analysis revealed changes in proteins involved in vesicle secretion, we further investigated whether the exosome biogenesis was affected in GLA-null CMs. The exosomes were isolated from the culture medium, and their identity was confirmed by electron microscopy. Multivesicular bodies were observed and exhibited the typical characteristic cup-shaped morphology (Supplementary Figure S4A) and size (diameter between 50 and 150 nm). Exosome size was measured by the Nanosight tracking system (Supplementary Figure S4B) and their identity

was confirmed by detecting expression of exosomal markers TSG101 and CD63 (Supplementary Figure S4C). Quantification of exosome though CD63 PE-conjugated fluorescence by flow cytometry analysis revealed that at higher CD63 positive levels in exosomes derived from GLA-null CMs compared to H9 CMs (Figure 4D). Next, we investigated the number of isolated exosomes from H9 and GLA-null CMs. Interestingly, GLA-null CMs produced significantly larger number of exosomes than H9 CMs (10.8 <sup>×</sup> 108 vs. 5.1 <sup>×</sup> 108, *<sup>p</sup>* <sup>&</sup>lt; 0.001) (Figure 4E). These results indicate that GLA-null CMs increased the production and secretion of exosomes.

#### *3.6. Vesicle Turnover Impairment Induces Cardiotoxicity in GLA-Null CMs*

Several studies have shown that the molecular machinery and regulatory mechanisms are shared between exosome biogenesis and autophagy [25,26], suggesting that these two processes are intimately linked. Since the proteomic analysis revealed the downregulation of components of vesicular trafficking machinery in GLA-null CMs, we hypothesized that dysfunctional autophagy pathway may underlie the impairment of cellular homeostasis in FD-related cardiomyopathy. To investigate whether autophagy flux impairment plays a role in cardiac phenotype of FD, we used western blotting to analyze the expression of autophagy markers LC3-I and LC3-II. The levels of these proteins were monitored in a time course of four hours after induction of autophagy with HBSS medium, and it was revealed that LC3-II accumulated by two hours after autophagy induction in both H9 and GLA-null CMs indicating to initiation of autophagy flux (Figure 5A,B). However, after four hours of autophagy induction, LC3-II levels dropped significantly in H9, but remained high in GLA-null CMs, signifying the digestion of autophagosome by lysosomes in the former and the block of autophagy flux in the latter (Figure 5A,B). Previous studies have indicated that undigested material within the autophagosome, including dysfunctional mitochondria, may be a source of free radicals, which in turn can result in cellular dysfunction and apoptosis [27,28]. Mitochondrial integrity and turnover play an important role in CM bioenergetics and function. MitoSOX Red staining demonstrated significant activation of mitochondrial superoxide production in GLA-null CMs (Figure 5C). The quantification of the MitoSOX Red staining intensity indicated that mitochondrial superoxide production in GLA-null CMs was 3.4-fold higher than that in H9 CMs (Figure 5D). To assess CM apoptosis, annexin V assay was used. GLA-null CMs displayed 4-fold increase in staining with annexin V in comparison with H9 CMs (Figure 5E,F). Furthermore, we observed significant 2-fold increase of lactate dehydrogenase (LDH), a marker of necrosis, in the supernatant derived from GLA-null CMs (Figure 5G). Collectively, our results demonstrate that GLA-null CMs were characterized by autophagy impairment and active mitochondrial ROS production that caused apoptosis and necrosis.

**Figure 5.** Vesicle turnover impairment induces cardiotoxicity in GLA-null CMs. (**A**) Western blot showing expression of LC3 protein isoforms (LC3-I and LC3-II) in a time course of induction of autophagy with HBSS medium. GAPDH was used as a loading control. (**B**) The expression levels of LC3-II were measured by using ImageJ and the quantification results presented as mean ± standard deviation error bars from three independent experiments. (**C**) Staining of mitochondrial ROS in GLA-null CMs and H9 control CMs with MitoSOX Red. (**D**) MitoSOX Red fluorescence intensity quantified by flow cytometry presented as mean ± standard deviation from three independent experiments. (**E**) Flow cytometry analysis of annexin V-positive cells in populations of GLA-null and H9 CMs. (**F**) Quantitation of annexin V-positive cells. (**G**) The level of LDH secreted by H9 and GLA-null CMs. The data are presented as a fold change relative to H9 control. These data are presented as mean ± standard deviation error bars from three independent experiments.

#### **4. Discussion**

The lysosome storage dysfunction (LSD) is the major factor in etiology of inherited hypertrophic cardiomyopathy, including FD; however, the underlying mechanisms of FD-associated cardiomyopathy are not fully understood. In order to expand the therapeutic strategies for FD, the construction of in vitro disease models using primary human cells is essential and unavoidable. Whereas the patients' cells can be used to directly model the effects of drugs on humans, their availability and capacity for expansion are limited and finite compared to in vitro derived cell lines, especially for vital organs such as heart and brain. These drawbacks restrict the capacity of these models to faithfully simulate human disease. By comparison, ESCs can sidestep these limitations and thus provide a powerful and versatile tool for disease therapy, as well as basic research. Combined with the advancements in genome editing

technology though CRISPR/Cas9, it is now possible to model human diseases in a physiologically, pathologically, and genetically relevant manner.

Mass spectrometry-based proteomics has been recognized as a powerful tool with a potential to uncover detailed changes in protein expression [29]. To date, most of the proteomics studies performed on FD patients examined FD-affected renal tissue or plasma; however, few studies of protein expression have used FD-affected human heart tissue [30,31]. Although it has been revealed that Gb3 accumulation induces endothelial KCa3.1 degradation in *Gla*-knockout mice through clathrin and Rab5C, which are the critical components of endosome maturation machinery [32], the proteomic profiling performed in our study revealed that *GLA* knockout led to the downregulation of Rab GTPases, including RAB11 subfamily, which are involved in recycling from an endosomal compartment to the plasma membrane, and was shown to contribute to exosome secretion in neuronal cells [33], although the molecular mechanism of RAB11 function in exosome secretion has yet to be deciphered, especially regarding its downstream effectors. On the other hand, our proteomic profile also revealed downregulation of Rho GDP-dissociation inhibitor 2 (GDIR2) in GLA-null CMs. This observation is consistent with previously published clinic proteomic profiles, where PBMCs isolated from FD patients were analyzed, and among the downregulated proteins were calnexin, Rho GDP-dissociation inhibitor 1 (GDIR1), Rho GDP-dissociation inhibitor 2 (GDIR2), chloride intracellular channel protein [34]. Rho GDIs play an important role in regulating Rho GTPases, which are members of the Ras superfamily of GTP binding proteins that participate in the regulation of cytoskeleton and other cellular functions including proliferation, differentiation, and apoptosis [35,36]. Rho GDI is ubiquitously expressed and binds to all Rho family proteins, including RhoA [37]. The small G-protein RhoA regulates the actin cytoskeleton, and its involvement in cell proliferation has also been established. In cardiomyocytes grown in vitro, RhoA induces hypertrophic cell growth and gene expression [38,39]. In vivo, however, cardiac-specific overexpression of RhoA leads to development of heart failure [40,41]. These correlated evidences suggest that exosome secretion regulated by Rab GTPase/RhoGDI signaling pathway may utilize as a target for the potential therapeutic strategy for FD-associated cardiomyopathy.

FD is characterized by failures of cellular autophagy associated with accumulation of glycogen granules and intracytoplasmic vacuoles that contain autophagic material. Impairment of autophagic flux in FD, which was manifested as defects of autophagosome maturation in renal endothelial cells and mesangial cells [42,43]. Autophagy is an evolutionary conserved process of self-degradation of cellular components by autophagosomes, which are delivered to the lysosomal machinery. Several studies have shown that starvation-induced autophagy reduces exosome secretion due to the fusion of multivesicular bodies with autophagic vacuoles [44]. In contrast, cellular stresses, such as senescence and ER stress, increase exosome secretion [45,46]. It is not clear why cells respond to stress by releasing more exosomes, but this could be an alternative way of eliminating waste products. The secreted exosomes may be targeted to and degraded by phagocytes, but they may also have other destinations. Exosomes secreted as waste are likely to affect neighboring cells and possibly induce pathological conditions. Another possibility is that cells may communicate with neighboring cells about intracellular stress by increasing exosome release. Therefore, preventing waste accumulation and rescuing the autophagic ability in FD-affected CMs may be utilized as another therapy approach for FD cardiomyopathy. It has been observed that exosomes derived from CMs harbor a variety of mRNAs, miRNAs and proteins, which may be transferred to the adjacent endothelial cells and modulate their function [47]. Interestingly, exosomes derived from ESCs/iPSCs were shown to possess regenerative power on CMs by augmenting and modulating endogenous repair mechanisms [48,49]. Emerging evidence from the studies of normal development, as well as multiple disease studies, revealed that exosome secretion and autophagy act in a coordinated manner to maintain homeostasis through lysosomal degradation and/or release of cellular cargo [50]. Therefore, considering the role of exosomes in physiological and pathological conditions, strategies that interfere with the release of exosomes and impair exosome-mediated cell-to-cell communication could potentially be exploited therapeutically in FD cardiomyopathy.

The objective of the present study was to explore the influence of Gb3, a lysosomal glycolipid accumulating in FD-affected cells, on modulation of cellular vesicle cycling and the possible mechanism underlying cardiomyopathy. We demonstrated that disruption of *GLA* with CRISPR/Cas9 resulted in the complete ablation of GLA protein expression in hESCs. Results from this study may provide mechanistic insights into how Gb3 accumulation modulates vesicles formation, particularly the autophagy flux in CMs. Such information may be extremely important for potential application in prevention and intervention of adverse effects of FD-associated cardiomyopathy.

#### **5. Conclusions**

To summarize, in the present study we show that CRISPR/Cas9-mediated *GLA* knockout of hESC-derived CMs can serve as an in vitro FD model for studying hypertrophic cardiomyopathy. Here, we adopted CRISPR/Cas9-mediated genomic editing to successfully generate GLA-deficient hESC clones (Figure 1). These GLA-deficient hESC clones displayed the properties of pluripotency, and were differentiated into CMs, which exhibited the typical biochemical and pathological abnormalities of FD including ablated GLA expression, enlarged cellular size, increased expression of cardiac hypertrophy genes and Gb3 accumulation. Therefore, these GLA-null CMs clones recapitulated the typical characteristics of FD-associated cardiomyopathy (Figures 2 and 3). To determine the global protein expression changes involved in the hypertrophic response to GLA deficiency, we used proteomic analysis and demonstrated markedly reduced expression of proteins involved in cytoskeleton dynamics and secretion of extracellular vesicles (Figure 4). Since exosome biogenesis and autophagy are highly interrelated processes, we have shown that *GLA* knockout led to autophagic impairment, which resulted in increase of mitochondrial ROS production and cell death (Figure 5). Additional investigation of the role of autophagy in FD is highly promising for the development of novel chemical modulation-based therapeutic approaches that may be more economically viable than traditionally-used ERT.

**Supplementary Materials:** Supplementary materials are available online at http://www.mdpi.com/2073-4409/8/4/ 327/s1.

**Author Contributions:** Conceptualization: H.-Y.S., and C.-S.C.; Methodology: H.-Y.S., C.-S.C. and S.-J.C. writing—original draft preparation: H.-Y.S. and A.A.Y.; writing—review and editing: A.A.Y., M.-L.W. and Y.-P.Y.; Supervision: C.-Y.W., H.-B.L., W.-C.Y.; project administration and funding acquisition Y.-L.C. and S.-H.C. All authors read and approved the final manuscript.

**Funding:** This work was financially supported by the Ministry of Science and Technology in Taiwan (MOST 106-2911-I-010-502) New Partnership Program for the Connection to Top Labs in the World; MOST 107-2633-B-009-003−107-0210-01-19-01−and 106-2319-B-001-003; Research Foundation of Cardiovascular Medicine, the Department of Health Cancer Center Research of Excellence (MOHW105-TDUB-211-134003, MOHW105-TDU-B-211-133017, MOHW106-TDU-B-211-113001, and MOHW107-TDU-B-211-123001); VGH, TSGH, NDMC, AS Joint Research Program (VTA105-V1-5-1, and VTA107-V1-5-1), VGH, NTUH Joint Research Program (VN106-02, VN107-16); National Health Research Institutes in Taiwan (NHRIEX106-10621BI, and NHRI-EX107-10621BI); "Development and Construction Plan(107F-M01)" of the School of Medicine, National Yang-Ming University; "Aiming for the SPROUT Project-Center for Intelligent Drug Systems and Smart Bio-devices (IDS2B)" of National Chiao Tung University from The Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education (MOE) in Taiwan.

**Acknowledgments:** pSpCas9(BB)-2A-GFP (PX458) was a gift from Feng Zhang (Addgene plasmid # 48138; http://n2t.net/addgene:48138; RRID: Addgene\_48138). We thank Ms. Yi-Ching Tsai, Yu-Ling Kuo, Hsiao-Yun Tai, Yu-Hsuan Yang, Fu-Ting Tsai from Taipei Veterans General Hospital for isolation and maintain the hESCs, assisted the experiments, data analysis, and literature review.

**Conflicts of Interest:** The authors declare that they have no potential conflicts of interest in the present study.

#### *Cells* **2019**, *8*, 327
