The role of ultrasound in enhancing mesenchymal stromal cell‐based therapies

Abstract Mesenchymal stromal cells (MSCs) have been a popular platform for cell‐based therapy in regenerative medicine due to their propensity to home to damaged tissue and act as a repository of regenerative molecules that can promote tissue repair and exert immunomodulatory effects. Accordingly, a great deal of research has gone into optimizing MSC homing and increasing their secretion of therapeutic molecules. A variety of methods have been used to these ends, but one emerging technique gaining significant interest is the use of ultrasound. Sound waves exert mechanical pressure on cells, activating mechano‐transduction pathways and altering gene expression. Ultrasound has been applied both to cultured MSCs to modulate self‐renewal and differentiation, and to tissues‐of‐interest to make them a more attractive target for MSC homing. Here, we review the various applications of ultrasound to MSC‐based therapies, including low‐intensity pulsed ultrasound, pulsed focused ultrasound, and extracorporeal shockwave therapy, as well as the use of adjunctive therapies such as microbubbles. At a molecular level, it seems that ultrasound transiently generates a local gradient of cytokines, growth factors, and adhesion molecules that facilitate MSC homing. However, the molecular mechanisms underlying these methods are far from fully elucidated and may differ depending on the ultrasound parameters. We thus put forth minimal criteria for ultrasound parameter reporting, in order to ensure reproducibility of studies in the field. A deeper understanding of these mechanisms will enhance our ability to optimize this promising therapy to assist MSC‐based approaches in regenerative medicine.


| MESENCHYMAL STROMAL CELL BIOLOGY
Within the field of regenerative medicine, mesenchymal stromal cells (MSCs) have been a popular area of research owing to their antiinflammatory effects, secretion of growth factors, and ability to home to damaged tissue. 1,2 MSCs are multipotent cells that, as their name suggests, 2 can give rise to various mesenchymal lineages, including bone, cartilage, and adipose tissue. Though they were first isolated from bone marrow, 3 MSCs have since been purified from a variety of other tissues, including adipose, 4 muscle, dermis, 5 dental pulp, 6 perivasculature, 7 and Wharton jelly from the umbilical cord. 8,9 MSCs are believed to play a natural regenerative role in the human body: in response to tissue damage, MSCs are released into circulation, where they home to the site of injury in response to inflammatory signals. 2 MSC homing is a multistep process which can be split into five steps: (a) tethering and rolling, (b) activation, (c) arrest, (d) transmigration/ diapedesis, and (e) nonsystemic migration. 10 During tethering, CD44 expressed on the MSC surface catch onto selectins on the endothelium, after which they begin rolling along the vessel wall. 11 Activation is facilitated by G-protein coupled chemokine receptors, most prominently CXCR4, which binds stromal cell-derived factor 1 (SDF-1) released by inflamed tissue. 12 These interactions activate integrins (VLA-4) on the MSC surface, which then bind to receptors on the endothelium (VCAM-1) to trigger cell arrest. After arrest, MSCs undergo transmigration or diapedesis to pass through the endothelium. This step is facilitated by the secretion of enzymes like matrix metalloproteinases (MMPs) that break down the endothelial basement membrane. 13 Finally, having exited the systemic circulation, MSCs undergo further nonsystemic migration to reach the injured tissue, guided by chemokines and growth factors. 14 Within the tissue, they secrete a variety of factors with powerful immune-modulating, angiogenic, and antiapoptotic effects. [15][16][17] MSCs are highly immunosuppressive, being able to convert pro-inflammatory environments into anti-inflammatory environments by suppressing T cell, B cell, natural killer (NK) cell, and dendritic cell populations, as well as by expanding regulatory T-cell pools. 18 Their angiogenic ability is also well documented, owing to their ability to secrete potent angiogenic factors like vascular endothelial growth factor (VEGF), insulin-like growth factor 1 alpha, and hepatocyte growth factor (HGF), 19 which activate the PI3K-Akt pathway in endothelial cells to inhibit apoptosis, increase survival, and stimulate new blood vessel formation. 20 Given these regenerative abilities, there has been great interest in exploiting the therapeutic potential of MSCs. MSCs can be cultured in vitro and then transfused into patients, after which they home to damaged tissue to aid in recovery and serve as an effector for tissue regeneration. 1 Several properties make them attractive platforms for cell-based therapy. They are easy to harvest from bone marrow or adipose tissue, expand in culture, and can then be transplanted into patients via an intravenous injection. MSCs appear to be somewhat immune-privileged, [21][22][23] and many clinical trials have demonstrated their safety in humans. Indeed, there are over 100 registered clinical trials using MSCs for applications such as immune modulation in multiple sclerosis and type 1 diabetes, tissue protection following myocardial infarction or liver cirrhosis, and tissue regeneration for bone and cartilage repair. 24 The results of such trials, though promising, leave much room for improvement. Perhaps the biggest hurdle encountered by MSCs is their ability to be targeted to their intended destination. When MSCs are infused intravenously, only a few percent ultimately reach the target tissue due to inefficient homing. 25 Another hurdle is stimulating the MSCs to secrete regenerative factors in sufficient quantities once they reach the damaged tissue, in order that they have an appreciable clinical effect.
Many strategies have been used to improve the homing and regenerative capabilities of MSCs, including genetic modification, cell surface engineering, and in vitro priming. [26][27][28] One novel method for improving MSC-based therapies comes in the form of ultrasound, which has been shown to be effective both for improving MSC homing and their regenerative capabilities. This review discusses ultrasound-based methods that have been demonstrated to enhance MSC-based therapies and the potential molecular mechanisms by which they do so.

| THERAPEUTIC ULTRASOUND
Although ultrasound is most commonly used for diagnostic imaging, it has been adopted for a variety of therapeutic applications since the 1950s. 29 Therapeutic ultrasound often utilizes acoustic pressures and intensities well above those of diagnostic ultrasound (DUS) in order to elicit some form of biological effect or response. Typically, the ultrasound beam is focused to a point within the body, thereby selectively targeting a specific tissue of interest and avoiding bioeffects in the tissues lying between the ultrasound transducer and the target tissue.

| Forms of therapeutic ultrasound
Under the umbrella of ultrasound therapy, a variety of methods have been investigated, with different modes of delivery, intensity, and

Significance statement
Mesenchymal stromal cells (MSCs) are a popular platform for regenerative medicine due to their ability to home to damaged organs and secrete molecules that spur cell growth and suppress inflammation. However, there remains a need to optimize their therapeutic effect for clinical translation.
One such strategy is the use of ultrasound. Ultrasound can be applied to MSCs to enhance their ability to secrete regenerative molecules or applied to a target organ to make it a more attractive destination for infused MSCs. The present article reviews the current knowledge of ultrasound's biological effects and preclinical applications for MSC-based therapies.
biological mechanisms. A few specific examples of therapeutic ultrasound include high-intensity focused ultrasound (HIFU) for tissue and tumor ablation, 30 histotripsy (the mechanical fractionation of tissue) to break up and liquefy diseased tissue, 31 low-intensity pulsed ultrasound (LIPUS) for aiding bone fracture healing, 32 and extracorporeal shockwave therapy (ESWT) for breaking kidney and bladder stones. 33 Even DUS has also been used in therapeutic contexts, though always in conjunction with adjuvants. Adjuvants are agents used to amplify the effect of ultrasound: in ultrasound-mediated microbubble destruction (UMMD), microscopic bubbles are injected into the bloodstream, and upon exposure to focused ultrasound, the bubbles cavitate to cause a variety of physical and biological effects. Researchers have broadly categorized ultrasound, into low vs high-intensity and continuous vs pulsed methodologies ( Figure 1). The labels, however, are arbitrary and often inconsistent. For organizational purposes, we will keep the labels as reported in the literature. However, these different forms of ultrasound can be better described using a spectrum of intensities and other parameters.

| Ultrasound parameters
To understand the various forms of therapeutic ultrasound discussed in this review, we will briefly review their basic physical parameters.
Ultrasound frequency is the number of times per second a particle experiences a complete compression and rarefaction cycle, and is In this review, we focus specifically on those forms of therapeutic ultrasound that have been tested in conjunction with MSC-based therapies. These strategies can be broadly categorized into two approaches. First are the ones that apply ultrasound to the target tissue, upregulating the expression of homing factors so as to make it a more attractive target for MSCs (Table 1). Second are the ones that apply ultrasound to cultured MSCs in vitro, so as to modulate their self-renewal, differentiation, and production of regenerative factors (Table 2). These approaches have been applied to a variety of organ systems and disease models (Table 3).

| PULSED FOCUSED ULTRASOUND
Pulsed focused ultrasound (pFUS), sometimes referred to as pulsed high intensity focused ultrasound, is a therapeutic ultrasound method that uses short-duration, high-intensity pulses to nondestructively target tissues of interest. Though there is wide variation in the parameters that constitute pFUS, many of the studies discussed in this section report I SATA = 133 W/cm 2 , PRF 5 Hz at 5% duty cycle, frequency 1 MHz. pFUS has been shown to be relatively safe, causing minimal histological alterations. 37,38 Although one study found enlarged gaps between muscle fiber bundles following pFUS sonication, these differences went away within 72 hours. 106  homing to the sonicated area. Importantly, the increased homing following pFUS seems to result not from increased leakiness of the vasculature but rather from the induced molecular changes. 38

| Molecular mechanism of pFUS-mediated MSC homing
Research on the molecular mechanisms underlying pFUS and its therapeutic potential, though scant, has been gaining steady interest  etanercept (a TNF-α inhibitor), as well as in COX2-knockout mice. 39 Similar results have been achieved in the kidney. Ziadloo et al  One study investigated the effect of pFUS on the native pancreas. 42 Here, pFUS to the pancreas had no effect on tissue histology and did not elevate serum amylase or lipase (markers of pancreatitis).
Based on these studies, a rudimentary understanding of pFUSinduced MSC homing is beginning to emerge ( Figure 3A). However, much remains unknown regarding the complete molecular mechanism and the involved signaling pathways. Different intensities and parameters of pFUS may also elicit different tissue responses, a question that is only beginning to be rigorously explored.  stores. 68 Parts of this proposed mechanism have also been weakly supported by previous studies, such as the involvement of CX43 69 and release of extracellular ATP. 80 The release of intracellular Ca 2+ has interesting parallels with the mechanisms of pFUS as discussed above ( Figure 3A).

| In vitro effects of LIUS
Most studies on LIUS have applied it to cultured MSCs in vitro ( Figure 3B). Several studies have demonstrated that both cLIUS 74

| EXTRACORPOREAL SHOCK WAVE THERAPY
ESWT uses high-amplitude acoustic waves to deliver mechanical forces to the tissue. In ESWT, a shockwave is induced by transmitting high-pressure ultrasound wave (generally a 1 microsecond spike at roughly 50 MPa). 29 ESWT is traditionally used in kidney stone lithotripsy 33 and in physical therapy. 114 Data suggest that the underlying mechanism of ESWT is based on its ability to reduce inflammatory reactions, enhance angiogenesis, suppress oxidative stress and apoptosis, and upregulate SDF-1. 51,52 This has led some groups to investigate whether it can enhance MSC-based therapies. As was the case with LIUS, ESWT has been applied both to cultured MSCs in vitro ( Figure 3B), as well as to target tissues in vivo. Though different types of ESWT exist, it is infrequently reported in studies, and thus we will treat them here collectively.

| In vitro effects of ESWT
In vitro, there is evidence that ESWT increases MSC proliferation by activation of the MAPK pathway. 69 Shockwaves also seem to induce MSCs to secrete more growth factors and cytokines, such as VEGF and CXCL5. 53,60,70,77 Indeed, the conditioned media of ESWT-treated MSCs better enhances neurite growth and endothelial tube formation in vitro. It also increases the expression of homing factors like SDF-1 and improves in vitro migration. 53,71 Similar to LIUS, ESWT seems to enhance the differentiation of MSCs toward osteoprogenitor cells in vitro, as evidenced by greater expression of osteogenic markers like RUNX2, BMP2, ALP, OCN, and OSX. 77,86,87,99 This effect may be mediated by the activation of focal adhesion kinases. 88  In a rat model of a segmental femoral defect, ESWT applied to the bone defect was able to increase MSC homing. 60 The MSCs were found to differentiate into both osteoblastic and chondrocytic fates.

| In vivo effects of ESWT
Furthermore, ESWT increased the local expression of both TGFβ and VEGF, which likely played chemotactic and mitogenic roles.
In a rat model of stroke, UMMD has been shown to enhance MSC homing to the brain twofold, compared with either MSCs alone or ultrasound without microbubbles. 101 MSCs + UMMD better reduced infarct volume, cerebral edema, and the neurological severity score, though no molecular mechanisms were investigated.
In the prostate, UMMD has been shown to enhance MSC homing in a rat model of chronic bacterial prostatitis, reducing inflammatory cell infiltration and fibrous tissue hyperplasia. 58 The combined MSC + UMMD treatment reduced TNF-α and IL-1β levels in the prostate, reflecting reduced inflammation; the individual treatments, however, did not result in such a reduction.
In the kidney, UMMD has been used to enhance MSC homing in a mouse model of diabetes. 36 The sonication resulted in increased  61 Wu et al developed microbubbles loaded with SDF-1. 105 Infusion of the SDF-1-loaded microbubbles followed by focused ultrasound to the kidney enhanced MSC homing 1.8-fold compared with normal microbubbles and 6.6-fold over ultrasound alone.
UMMD appears to elicit a greater therapeutic response than ultrasound alone. However, it does come with an inherent problem: microbubble cavitation disrupts tissue integrity and cell membranes and can thus cause hemorrhage. [122][123][124] Though some studies report finding no evidence of such micro-hemorrhages, 125,126 UMMD in its current state is still faced with some safety concerns. There are few studies specifically seeking to improve the safety of UMMD, which may depend on various parameters of the sound waves used for cavitation.

| DISCUSSION
Here we have presented a comprehensive review of the methods by which ultrasound has been leveraged to enhanced MSC-based therapies. A variety of strategies exist: whether to sonicate the cultured MSCs or the target tissue, whether to utilize adjuvants like microbubbles, and whether to use low-or high-intensity sound waves. The large number of studies in this field has implicated several potential pathways by which sound waves exert their biological effects through mechanotransduction; however, further studies still need to be performed to completely understand the exact mechanisms involved.
Indeed, future studies need to systematically investigate the immediate and long-term responses in sonicated tissue as a function of ultrasound intensity.
7.1 | The need for standardization in ultrasound parameter reporting One of the most pressing needs in the field is standardization in ultrasound parameter reporting. The simple categorization scheme shown in Figure 1 is insufficient; for instance, the intensities used under the umbrella of "pFUS" span orders of magnitude. More problematic is that many studies do not report intensity parameters, and of the ones that do, most do not state what kind of intensity was measured. This trend is troubling for the reproducibility of studies in this field and makes meta-analyses impossible. We therefore recommend that all studies using therapeutic ultrasound to optimize cell therapy report temporal average intensities (I SATA and I SPTA ), because the induced bioeffects in these studies are dependent on the temporal application of ultrasound. I SATA describes the average acoustic power applied to tissue over time. I SPTA , although perhaps less useful, describes the maximum power applied to the tissue over the course of the treatment. MI (or frequency and peak negative pressure) should also be reported, as it indicates the extent of mechanical bioeffects. Furthermore, all studies utilizing microbubbles should additionally report the temporal peak intensities (I SATP and I SPTP ). Because bubbles are responsive to the instantaneous pressure, the temporal peak intensities will be informative of the therapeutic effect of microbubbles.
With improved reporting, researchers in the field will be better equipped not only to reproduce studies but also to expand upon, and