Exosomal MicroRNA Clusters Are Important for the Therapeutic Effect of Cardiac Progenitor Cells
Although current evidence indicates that cardiac progenitor cells (CPCs) may improve cardiac repair after ischemic injury, the exact mechanism for this cardioprotective effect to date remains undefined. In this issue of Circulation Research, Gray et al1 report on the use of systems biology to link the change in exosomal microRNAs (miRNA) to the beneficial effects of CPCs during ischemic injury in mice. Expression analysis showed that exosomes from hypoxic CPCs contain a higher level of a subset of miRNAs than exosomes coming from normoxic CPCs. To connect the relevance of the change in miRNA levels to the biological effect of exosomes, the authors performed a principle component analysis of the miRNA expression profiles in exosomes, which identified 4 unique miRNA clusters, which after further mathematical modeling seemed to correlate to a biological effect related to the therapeutic benefit. Instead of studying the biological function of a single miRNA, computational modeling can be applied to integrate multiple variants to gain more insight into the overall relationship and interaction between the different datasets. Gray et al1 show that especially for miRNA biology, where a miRNA can target up to hundreds of genes in a parallel, systems biology could turn out to be a suitable approach for predicting a biological outcome or therapeutic benefit.
Article, see p 255
Extracellular Vesicles, a Focus on Exosomes
Cells communicate with each other via direct cell–cell contact and secretion of soluble factors. Soluble factors that are gaining an increasing amount of attention are extracellular membrane vesicles. The ability of vesicles to transport different molecules, such as proteins, peptides, mRNA, and microRNAs, from one cell to the other and their general body-wide distribution make them attractive candidates for horizontal information transfer between cells.2 Although the release of vesicles to influence other cells seems to be straightforward, multiple mechanisms, size distributions, and RNA profiles are observed among different vesicle populations released by cells.3
Exosomes, which are the best characterized vesicle population to date (40–100 nm in size), represent a specific subset of vesicles because of their intracellular origin. Exosomes are formed by inward budding of the late endosomal membrane of multivesicular bodies. This results in the formation of intraluminal vesicles in the multivesicular bodies, which can be released into the cellular environment through the fusion with the cell membrane.4 Because exosomes originate from the endosomal compartment, their content is dynamic and likely reflects the cellular origin and physiological state of the cell. Exosomes can be released from cells via a constitutive or inducible manner, like in response to increased intracellular Ca++, DNA damage, extracellular ATP, hypoxia, and lipopolysaccharide stimulation.5,6 The observation that the composition of exosomes is flexible and highly regulated, and that they are produced by almost all cell types, indicates that exosomes are ideal biological devices to mirror and regulate a wide variety of local and systemic cellular processes, playing a role during homeostasis and the course of diseases.
Exosomal Transfer of MicroRNAs in the Cardiovascular System
The myocardium is a complex mixture of different cell types, and myocardial performance is reliant on a intricate interplay of well-organized interactions, including controlled intra- and intercellular exchanges. Recent work suggests that also in the heart extracellular vesicles, and especially exosomes, are a tool to support long range cell–cell communications between a variety of different cardiac cell types.5,6
MiRNAs are short, single-stranded, noncoding RNAs that modulate intracellular gene expression through posttranscriptional regulation of targeted mRNAs. Although exosomes can contain DNA, RNAs, and protein, several reports to date ascribe at least part of the functional effects of extracellular vesicles to their active delivery of miRNAs to other cell types and tissues. In line with this, Hergenreider et al7 already in 2012 showed that the extracellular vesicles that are secreted by endothelial cells in response to shear stress deliver miR-143/145 to smooth muscle cells to exert an atheroprotective effect.
Although the biological relevance of the exosomes and their cargo is likely determined by the state of their donor, as well as the recipient cell, exosomes derived from a diseased heart seem to transmit a negative signal, as for example shown by the observation of Bang et al8 that cardiac fibroblasts promote cardiomyocyte hypertrophy by the exosomal transfer of miR-21*. Also peripartum cardiomyopathy seems to be under the influence of exosomal miRNA transfer because endothelial cells secrete more miR-146a–enriched exosomes that are transferred to cardiomyocytes where they subsequently reduce metabolic activity and contractile function.9
Circulating miRNAs have also shown to be relevant for coronary heart disease. Patients with coronary heart disease showed a less efficient transfer of miRNAs to cultured HUVECs (human vascular endothelial cells), which may contribute to the abnormal intercellular communication that underlies coronary heart disease initiation and progression.10 This underscores that appropriate delivery of miRNAs is required to maintain healthy conditions.
Exosomes derived from cardiac stem or progenitor cells in general seem to elicit a more beneficial response.11 Recently, it was shown that the regenerative effect of cardiosphere-derived cells in response to cardiac ischemic injury is largely dependent on exosomal transfer of miR-146a, and that cardiosphere-derived exosomes alone are sufficient to confer the same effects as the cardiosphere-derived cells themselves, whereas miR-146a alone did not.12 Also exosomes coming from mesenchymal stem cells showed to contain an increased amount of several miRNAs in response to preconditioning. Injecting these preconditioned exosomes resulted in less apoptosis and remodeling after myocardial infarction,13 an effect that was largely contributed to the exosomal delivery of miR-22 to neighboring cells.14
Exosomes From CPCs Contain Cardioprotective MicroRNAs
Several studies have shown beneficial effects of intramyocardial delivery of CPCs during myocardial infarction. Although it was originally thought that cell differentiation and the secretion of growth factors and cytokines were the basis for the beneficial effects, accumulating evidence now points toward extracellular vesicles as being responsible for the therapeutic effects observed.
Recently, Barile et al15 showed that extracellular vesicles secreted by CPCs show an enrichment for several miRNAs, including miR-210 and miR-132, which have been implicated to have a function in cardiac biology. In vitro experiments showed that both miRNAs protected myocyte-like cells against apoptosis and miR-132 additionally promoted endothelial tube formation, implying that the transfer of these miRNAs from CPC-secreted vesicles could explain part of the protective mechanism. Although the approach of singling out specific miRNAs is valid and insightful, because we are dealing with a complex interrelated system, it might be more suitable to apply more integrative mathematical modeling systems to include multiple variables and biological changes in parallel to get a more complete view on the situation.
To explore the cardioprotective effects of CPCs in more detail, Gray et al1 set out to determine whether CPCs secrete proregenerative exosomes under hypoxic conditions. In vitro exosomes from hypoxic CPCs show an enhanced effect on tube-forming capacity of endothelial cells, which is likely because of the presence of miRNAs because inhibition of the RISC negated this effect. Gene expression analysis in the endothelial cells, however, showed marginal changes in response to the exosome treatment. On the contrary, pretreated fibroblasts showed a lesser increase in a subset of fibrosis-related genes when exposed to the exosomes of the hypoxic CPCs.
Seeking to identify the difference between exosomes from normoxic and hypoxic CPCs, the authors next determined exosomal miRNA content by microarray and identified 11 upregulated miRNAs of which only a subset were confirmed by qPCR.
Principle component analysis of the miRNA microarray data to explore covarying relationships between the differential regulation of miRNAs and the normoxic or hypoxic state of the CPCs indicated 4 distinct clusters of covarying miRNAs.
Subsequent modeling using the partial least square regression analysis was applied to determine a relationship between the treatment conditions of the CPCs, the miRNA levels, and the putative biological response. To do so, all miRNAs from the array were matched to the response of tube formation and the expression of fibrosis-related genes. Based on this analysis, the authors defined additional miRNAs to be proactive for tube formation, anti-CTGF (connective tissue growth factor) expression (as readout for fibrosis), or both.
Trying to explain the protective effects of hypoxic exosomes by the delivery of miRNAs to recipient cells in the heart by mathematical modeling is an interesting approach, but some aspects deserve some additional consideration. While the read outs in vitro seem to indicate that these exosomal miRNAs are at least able to influence endothelial tube formation, it remains undefined what triggers the effect of the exosomes on fibroblasts.
Focusing on miRNAs as the biologically active component in exosomes is likely only a part of a complex puzzle because other active compounds will play a role in the effect on the recipient tissue as well. Also for the exosomes coming from the hypoxic CPCs, it will be interesting to determine its full content.
While Gray and others have linked exosomal miRNA transfer (and presumably subsequent delivery to recipient cells) to a biological effect in distant cells or tissues, details on the quantity of miRNA that is actually being transported and delivered, and to which level this effects gene expression in the recipient cells, also warrants further investigation. The marginal effects that Gray et al found in endothelial gene expression after exposure to exosomes might reflect the small amount of transferred biologically active materials to the cells, which is likely the nature of the system but will increase the difficulty to study the direct downstream effects of exosomes.
Although the complexity of a biological structure, such as the heart, is intriguing, it also increases the difficulty of determining the relevance of many small changes that occur in response to a cue, like disease. The outcome is obviously the combined effect of many underlying impulses, however small they are. Computational modeling has been instrumental in increasing our understanding and providing predictive models for how these signals contribute to the greater whole.
While mathematical modeling might help to make sense of large data sets with relative small changes, one should be cautious of the validity of the datasets used. Gray et al here use microarrays for miRNAs as starting point for their modeling, but at the same time they show that one-third of their microarray based data could not be confirmed (only 7 out of 11 significantly regulated miRNAs could be confirmed by qPCR). It therefore seems appropriate to validate the outcome of these models by follow-up experiments, in this case measure a direct biological effect of the newly identified miRNAs on angiogenesis or fibrosis.
In the article by Gray et al,1 the authors showed proof of principle by injecting exosomes from hypoxic CPCs into mice, which leads to an improvement in function and decline in fibrosis. While it is interesting to see the beneficial effects of exosomal delivery, it is feasible that other exosomal compounds besides miRNAs (also) contribute to the therapeutic benefits in mice. Currently there are no data to support the conclusion that miRNAs are the ones giving the benefit. Also, since the identified miRNAs in the study of Gray et al seem to have an effect on angiogenesis and fibrosis related genes, it would have been nice to confirm these effects in vivo. Future experiments will have to show that it is actually the miRNAs that are conferring the cardioprotection that is observed.
Looking to the Future
Cell–cell communication via exosomal transfer of biologically active material is incredibly exciting. Although the molecular mechanisms are yet undefined, once the exsosomes are released by their donor cell they are selectively taken up by specific recipient cells. Also the fact that exosomes are capable of transferring lipid RNAs and proteins to specific cells in the heart create exciting possibilities for therapeutic applications.
However, as also shown in the paper of Gray et al,1 the exosomal content is dependent on the state of the donor cell, and disease conditions seem to be of influence. Since the cargo of an exosome determines the effect is has on recipient cells, it is of great importance to standardize the conditions under which the exosomes are collected. And even though it appears clear that the cargo of these stem cell–derived exosomes can have a beneficial effect on the diseased heart, the underlying mechanism needs further investigation.
Undoubtedly, the coming few years will provide many more insights into the biological significance of cell-to-cell transport via exosomes. The use of systems biology creates an opportunity to increase our understanding and paves the way for designing optimized vesicles to enhance their effect on cardiac regeneration.
Sources of Funding
The work from J.P.G. Sluijter was supported by a grant from the ZonMw-TAS program (116002016), and the Netherlands CardioVascular Research Initiative (CVON): the Dutch Heart Foundation, Dutch Federation of University Medical Centers, the Netherlands Organization for Health Research and Development, and the Royal Netherlands Academy of Sciences. The work of E. van Rooij was supported by an ERC Consolidator grant and a grant from the Leducq Foundation.
E. van Rooij is a scientific cofounder and member of the Scientific Advisory Board of miRagen Therapeutics, Inc. J.P.G. Sluijter reports no conflicts.
The opinions expressed in this article are not necessarily those of the editors or of the American Heart Association.
- © 2015 American Heart Association, Inc.
- Gray WD,
- French KM,
- Ghosh-Choudhary S,
- Maxwell JT,
- Brown ME,
- Platt MO,
- Searles CD,
- Davis ME
- Bobrie A,
- Théry C
- Raposo G,
- Stoorvogel W
- Ailawadi S,
- Wang X,
- Gu H,
- Fan GC
- Sahoo S,
- Losordo DW
- Hergenreider E,
- Heydt S,
- Tréguer K,
- Boettger T,
- Horrevoets AJ,
- Zeiher AM,
- Scheffer MP,
- Frangakis AS,
- Yin X,
- Mayr M,
- Braun T,
- Urbich C,
- Boon RA,
- Dimmeler S
- Halkein J,
- Tabruyn SP,
- Ricke-Hoch M,
- Haghikia A,
- Nguyen NQ,
- Scherr M,
- Castermans K,
- Malvaux L,
- Lambert V,
- Thiry M,
- Sliwa K,
- Noel A,
- Martial JA,
- Hilfiker-Kleiner D,
- Struman I
- Barile L,
- Lionetti V,
- Cervio E,
- Matteucci M,
- Gherghiceanu M,
- Popescu LM,
- Torre T,
- Siclari F,
- Moccetti T,
- Vassalli G