Biodistribution of Mesenchymal Stromal Cells after Administration in Animal Models and Humans: A Systematic Review

Mesenchymal Stromal Cells (MSCs) are of great interest in cellular therapy. Different routes of administration of MSCs have been described both in pre-clinical and clinical reports. Knowledge about the fate of the administered cells is critical for developing MSC-based therapies. The aim of this review is to describe how MSCs are distributed after injection, using different administration routes in animal models and humans. A literature search was performed in order to consider how MSCs distribute after intravenous, intraarterial, intramuscular, intraarticular and intralesional injection into both animal models and humans. Studies addressing the biodistribution of MSCs in “in vivo” animal models and humans were included. After the search, 109 articles were included in the review. Intravenous administration of MSCs is widely used; it leads to an initial accumulation of cells in the lungs with later redistribution to the liver, spleen and kidneys. Intraarterial infusion bypasses the lungs, so MSCs distribute widely throughout the rest of the body. Intramuscular, intraarticular and intradermal administration lack systemic biodistribution. Injection into various specific organs is also described. Biodistribution of MSCs in animal models and humans appears to be similar and depends on the route of administration. More studies with standardized protocols of MSC administration could be useful in order to make results homogeneous and more comparable.

Since MSCs have variable phenotypes, with different expression of bio-markers depending on the source and means of isolation, as well as the tissue they come from, they cannot be considered as a homogeneous set of cells [11]. The International Society for Cellular Therapy set minimum criteria for characterizing human MSCs in order to promote a more uniform definition of MSCs. These criteria are: (a) Plastic adherence when maintained in standard culture conditions; (b) Expression of CD105, CD73 and CD90 and lack of expression of CD45, CD34, CD14, or CD11b, CD79a or CD19 and HLA-DR surface molecules; and (c) Differentiation into osteoblasts, adipocytes and chondroblasts in vitro [12].
MSCs are of great interest because of the possibility of using them as a part of therapeutic regimens in a wide variety of human diseases, e.g., rheumatic and autoimmune diseases, skin diseases and complex ulcers and wounds [13][14][15][16]. Some characteristics of MSCs are fundamental for this purpose: (a) MSCs can be obtained from adult donors and expanded in vitro without losing their immunomodulatory and differentiation potential; (b) MSCs have hypo-immunogenic properties, so allogenic sets can be used, avoiding the need for autologous cell cultures; (c) Their immunomodulatory and transdifferentiating capabilities into different cell lineages can be exploited as a novel approach to the treatment of different diseases [13][14][15][17][18][19].
Different routes of administration of MSC-based medical therapies have been described both in pre-clinical and clinical reports, and the possible differences between them, in terms of safety and efficacy, is an issue which is still under discussion [15,16,[20][21][22][23]. These differences may be explained by the variable biodistribution of MSCs after their administration. The most common reported routes of administration are topical, intravenous and intraarterial, intramuscular and intralesional (including different locations e.g., skin, spinal cord, tendons).
Given the presumable importance of the different mechanisms of MSC biodistribution and their impact on the therapeutic effects, the objective of this systematic review is to describe how MSCs are distributed after their inoculation through different administration routes in animal models and humans.

Search Strategy
A literature search from January 2015 to April 2021 was performed using the Medline database. The following search terms were used: MSC or MESENCHYMAL STEM CELL or MESENCHYMAL STROMAL CELL or MULTIPOTENT STEM CELL or MULTIPOTENT STROMAL CELL or STEM CELL AND BIODISTRIBUTION or DISTRIBUTION.

Inclusion and Exclusion Criteria
The search was limited to: (a) Human or animal data; (b) In vivo studies; (c) Studies addressing the biodistribution of MSCs after any source of administration; (d) Articles written in English or Spanish. All types of epidemiological studies (clinical trials, cohort studies, case-control studies and cross-sectional studies) regarding the biodistribution of MSCs were considered.

Study Selection
The titles and abstracts obtained in the first search were reviewed to assess relevant studies. The full texts of all articles meeting the inclusion criteria were reviewed and their bibliographic references were checked for additional sources. Articles considered relevant were included in the analysis. Uncertainties about the inclusion or exclusion of articles were subjected to discussion until a consensus was reached.

Research Questions and Variables Assessed
The research questions were as follows: • How do MSCs distribute after intravenous and intraarterial injection in animal models and humans? • How do MSCs distribute after intramuscular injection in animal models and humans? • How do MSCs distribute after intralesional injection in different organs and tissues in animal models and humans? • Which cell marking techniques have recently been used in studies on humans?
The variables assessed in order to answer these questions were the model which received the MSCs (human or animal), the route of administration, the disease treated, the cell-marking technique used, the biodistribution assessment method, the time when the assessment was performed, and the outcomes regarding the biodistribution of the MSCs.

Results
An initial search found 6808 references (see Figure 1). After reviewing the titles and abstracts, 159 articles underwent full-text review. From this list, 50 articles were eventually discarded due to various issues: 33 articles did not assess biodistribution; 7 were related to other types of cells, rather than MSCs; 6 were not accessible or written in a different language; 3 only addressed the issue of in vitro MSCs; and 1 article was duplicated. Finally, 109 studies met the eligible criteria and were included in the review.

Biodistribution Characteristics of Mscs Depending on the Route of Administration
An overview and summary of all the information collected in this study can be seen in Table 1.

MSC Biodistribution in Animal Models
First, the biodistribution of MSCs after their delivery or injection into animal models will be discussed. Intravenous and intraarterial infusion, intramuscular injection and a wide variety of intralesional administrations of MSCs will be addressed in this section. Cells do not reach other organs apart from lungs in great quantities.

Intraarterial (not selective) Yes
Cells bypass the pulmonary filter so there is a wide distribution in the rest of the organs (heart, brain, kidneys, liver, digestive system) Convenient route of administration. Useful to bypass the lungs and achieve broader distribution.
Not so widely used. Intraarterial infusion is not common in clinical practice Cells are distributed mainly in the territory irrigated by the cannulated artery. Distribution of cells to other organs is possible but in smaller amounts.
Targeted deposition of cells is achieved.

Distribution of MSCs after Intravenous Injection in Animal Models
Intravenous injection has emerged as the most widely used route in the various research studies. This route of administration is a simple and effective way to deliver MSCs systemically. Most of the studies discussed in this section agree on the general characteristics of how mesenchymal cells are distributed after being injected into the venous stream (Table 2, Figure 2). To begin with, some general ideas can be stated about this issue: (a) after IV injection, most cells are retained initially in the lungs, which is the first capillary filter; (b) there is later redistribution of the cells, mainly to the liver, spleen and kidney, with few MSCs redistributing to other organs; (c) in some studies, later redistribution is very limited; and (d) some pathological entities seem to alter this biodistribution pattern.
A good example of this general distribution pattern can be seen in one study assessing intravenous infusion of MSCs in a myocardial infarction model in dogs [24]. It showed high distribution during the immediate post-infusion time in the lungs, with a posterior decrease in the amount of MSCs and a later redistribution from day 1 to 7 in different tissues, mainly in the liver, spleen and kidney. A similar model of myocardial infarction in mice [25] showed early distribution in lungs but an insignificant amount of cells distributed to other organs (less than 1%). Intravenous infusion of MSCs in baboons [26], and a late evaluation of their distribution in a variety of tissues, have demonstrated a wide distribution of MSCs after a long period of time: gastrointestinal, kidney, skin, lung, thymus, and liver tissues contained MSCs. Similar results were shown in several other studies [27][28][29][30][31]. The redistribution might be explained by phagocitation of MSCs: monocytes might perform this action, and then change their immunophenotype, inducing Treg cells [32]. Biodistribution of MSCs after intravenous infusion. After intravenous infusion, there is initial biodistribution in the lungs. Later, most cells redistribute to the liver, kidney and spleen. Few cells can be found in other organs and tissues. In some cases, diseased tissues have been found to be capable of attracting MSCs.
The alteration of the general distribution pattern in specific diseases has also been reported in several studies. Zhang et al. [33] found a significant amount of MSCs in the kidneys of rabbits with acute kidney injury. Similar results have been shown in a model of Alzheimer's disease [34,35] with higher brain distribution of MSCs in diseased animals compared to healthy animals. This was also evidenced in another study performed on mice with cerebral tumors [36], rats and beagle dogs with spinal cord injury [37,38], and rats with intracerebral hemorrhage [39]. Moreover, acute distress respiratory syndrome or liver tumors may also affect the distribution of cells after intravenous injection [40,41]. In contrast, in a murine model of experimental autoimmune encephalomyelitis [42], MSCs were not distributed to the brain area.
Although the lungs seem to be the area MSCs mostly distribute to after intravenous injection, Schmuck et al. [43] concluded that this may be due to the lack of sensitivity of bioluminescence techniques, which are carried out in most biodistribution studies. In their study, which used a 3D cryo-imaging system, they demonstrated a higher concentration of MSCs in the liver when compared to the lungs after intravenous infusion in rats with acute lung injury. In this line, Schubert et al. [44] demonstrated a high distribution of MSCs to the lungs with bioluminescence techniques on day 1 after intravenous infusion in mice with acute kidney injury. Cells cleared on days 3 and 6. However, when RT-PCR was performed on several tissues on day 6, variable amounts of mRNA were detected in the blood, liver, kidneys and lungs. Therefore, RT-PCR could be a better option for detecting the late presence of MSCs in tissues and could be used to complement imaging techniques.
Other situations, such as the modification of MSCs or the selective infusion of MSCs into certain veins, might also affect biodistribution. Moreover, some studies have shown that modifying MSCs may lead to cells selectively targeting specific organs. The modification of specific "homing markers" or adhesion molecules can lead to targeted homing of MSCs. This has been proven by modifying MSCs to achieve specific distribution to the liver [30]. In addition, the selective intravenous delivery could lead to differences in biodistribution. For example, Li et al. [45] demonstrated that superior mesenteric vein infusion of MSCs leads to more selective and longer homing of MSCs in a model of acute liver injury when compared to intravenous and inferior vena cava delivery.
Finally, regardless of the source of administration, Fabian et al. [46] demonstrated that the age of both the recipient and the donor of MSCs seems to affect the biodistribution of the cells. The study demonstrated that old recipients and donors showed a very restricted biodistribution of MSCs in mice after 28 days (mainly in the brain cortex and spleen) whereas young receptors and donors showed a wide variety of distribution.

Distribution of MSCs after Intraarterial Injection in Animal Models
Intraarterial infusion of MSCs has been used as an alternative and has also been compared to IV injection in several situations (Table 3, Figure 3). Briefly, the main characteristics of this route of administration are: (a) IA injection bypasses the pulmonary filter, so low amounts of MSCs are retained in these organs; (b) MSCs distribute more widely into the rest of the body's organs after IA infusion compared to IV delivery; (c) like the IV route, biodistribution after IA injection might be modified by several diseases; d) selective intraarterial delivery of MSCs might be very useful for targeting diseased organs.
As an initial example of these characteristics, one study performed on pigs [22] compared intravenous and intraarterial infusion techniques. MSCs were detected using SPECT/TC imaging, which showed a lower pulmonary captation in the intraarterial group, and a relatively higher uptake in other organs such as the liver, spleen and kidney. This was also studied in an acute kidney injury model in mice [47]. In this study, a significantly higher amount of MSCs were detected in the kidneys after intraarterial infusion, especially in mice with AKI. In contrast, the vast majority of MSCs were distributed to the lungs after intravenous injection. Moreover, intracardiac injection has also been reported to be an effective delivery route. This route of administration can be considered to be equivalent to the IA route when the cells are injected into the left chambers of the heart. In fact, after intracardiac injection [48], MSCs seem to follow a similar path; widespread distribution is observed (lungs, brain, spleen, liver, kidneys).
The fact that the IA route leads to a significantly higher distribution of MSCs in peripheral organs might be an interesting characteristic when homing MSCs in the diseased area is desirable. For example, in other studies it has been proven that intraarterial injection improves distribution to the damaged cerebral areas when compared to intravenous injection [49][50][51].
Regarding the distribution of MSCs to the brain after intraarterial infusion, Cerri et al. [52] evaluated distribution to the brain of MSCs injected in the carotid artery of a Parkinson's disease murine model. One group was treated with mannitol as a transient permeabilizing factor of the blood-brain barrier. Later assessment showed that rats not treated with mannitol had an extremely low amount of MSCs homing to the brain, whereas the group treated with mannitol showed a significantly higher amount of MSCs. Moreover, most of the cells were distributed in the ipsilateral hemisphere to the carotid used to inject them. Therefore, the use of a permeabilizing agent could be essential to allow the passage of MSCs into the brain. On the other hand, selective delivery of cells might help MSCs reach the damaged areas [51,53]. As occurred with IV injection, some pathological entities can modify the biodistribution of MSCs after IA injection. In the specific case of mice with inflammatory bowel disease, MSCs do not significantly distribute to lungs or liver but distribute mainly to the affected areas of the intestine [54]. In contrast, in a model of kidney injury, MSCs did not distribute to damaged kidneys after intracardiac injection [55]. Moreover, the dose of MSCs seems to be important when administered IA. One study showed that an increased dose of IA-administered MSCs led to a wider distribution of cells but also to a high degree of intravascular cell aggregation and mortality [56]. Thus, the dose of MSCs should be assessed before intraarterial delivery to avoid intraarterial aggregation.
The homing of MSCs to diseased tissues can be improved by selective intraarterial infusion. With this technique, MSCs are directly injected into selected arteries. This results in a greater amount of MSCs in the targeted organs. Some examples are discussed here: When MSCs are delivered directly into the renal artery, MSCs seem to distribute only in the kidneys, without systemic significant distribution, and mainly in the renal cortex [57]. Therefore, renal intraarterial MSC infusion limits off-target engraftment, leading to efficient MSC delivery to the kidneys. Similar results were found after selective intraarterial infusion into the superior mesenteric artery regarding the intestine distribution of MSCs [58], and the selective intraarterial limb infusion [59,60], with MSCs distributed in the target area and a small quantity of MSCs in the rest of the organs.

Distribution of MSCs after Intramuscular Injection in Animal Models
As this is widely used with classic drugs, intramuscular injection of MSCs has also been studied as a possible way to administrate MSCs (Table 4, Figure 4). As a general idea, whereas intramuscular injection of conventional drugs leads to a significant systemic distribution, MSCs injected intramuscularly do not seem to distribute to the rest of the body.
One study performed on mice to assess the sensitivity and specificity of quantitative PCR [61] for detecting MSCs showed that, 3 months after intramuscular injection of MSCs, no MSCs were detectable in any internal organ. However, DNA from MSCs was still present in the muscles where it was injected. This could suggest that MSCs do not distribute to other organs after intramuscular injection. This was in line with the findings of similar studies performed following intramuscular injection [62][63][64], with MSCs remaining at the injection site, but without MSCs distributing to organs. However, it has been demonstrated that, despite the lack of distribution of MSCs, when injected intramuscularly in a contralateral muscle to an inflamed area, MSCs are capable of reducing inflammation. This is thought to be performed by the release of soluble factors rather than the movement of the cells [65]. A recent review of intramuscular MSCs showed that, to date, no articles have found significant systemic biodistribution after intramuscular injection of MSCs [65].

Distribution of MSCs after Intralesional Injection in Animal Models
Several different intralesional routes of administration for MSC delivery have been described. The most important routes of administration of MSCs into lesioned areas will now be addressed • Intraarticular (IAr) delivery of MSCs (Table 4, Figure 4): Intraarticular injection of MSCs has been widely studied in different animal models. As a general idea, IAr injection lacks systemic biodistribution, whereas it leads to a very targeted delivery of cells into the joints. This has been adequately demonstrated by studies on different mice models of healthy animals, arthritis and osteoarthritis, where it was shown that MSCs do not distribute to other organs following intraarticular injection [21,[66][67][68][69]. Markides et al. [70] assessed the biodistribution of MSCs in a sheep model of osteochondral injury. After intraarticular injection, MSCs were only detected in the synovium, with a lack of MSCs within the chondral defect. Khan et al. [71] showed similar results after intratendinous injection, with no MSCs spreading from the injection site.
In contrast with that already described, some studies show an incidental distribution of MSCs. In these cases, MSCs have been shown to be present in the blood, distant zones or tendon lesions near the injection site. One study performed on a horse model of tendon lesions [72] showed that, although the vast majority of cells remained at the site where they had been injected, a small amount of MSCs could be found in blood for the first 24 h after injection, as well as in the contralateral control tendon lesions which had not been injected. Similar results were observed by Shim et al. [73]; after intraarticular injection, MSCs were detectable in blood with a peak at 8 h. No systemic distribution was observed. Moreover, other studies show that MSCs seem to be able to migrate from the joint to nearby tendinous lesions [74,75]. No systemic distribution has been demonstrated after intramuscular or intraarticular injection. After intraarticular injection, MSCs have been found to be able to migrate to nearby damaged lesions and into the bloodstream. Moreover, the use of a magnet on MSCs with a magnetic label is useful for targeted deposition of cells within the joint.
As occurred with the IV and IA routes, elective accumulation of MSCs in selected areas of a joint (i.e., a chondral lesion within the joint) can be achieved. MSCs must be modified by magnetic labeling. The subsequent use of a magnet during the transplantation [76] leads to the movement of the cells within the joint so they can be deposited in the target zone.
Finally, as a variant of IAr delivery, one study was performed to assess biodistribution of MSCs which were pre-loaded into bone grafts [77]. This study also showed the lack of systemic biodistribution of MSCs and the long-lasting MSCs in the graft up to 6 weeks. Similar results were found when injecting MSCs into the femoral head of pigs [78].

Injection of MSCs into the Reproductive and Urinary System
Some studies have been found on the issue of biodistribution of MSCs after injection into the urinary and reproductive systems. In a rat model of birth-trauma injury [79], the presence of MSCs following local injection into the periurethral tissues was demonstrated up to 7 days post-injection. In this case, no tests were performed to assess the distribution to other organs after local injection. Ryu et al. [80] injected MSCs into the outer layer of the bladder in a interstitial cystitis model. It was demonstrated that cells are able to migrate from the outer layers of the bladder to the urothelium for the first 30 days after injection and to home as perivascular cells. Dou et al. [81] found that after intracavernous injection, MSCs distributed to the lower abdomen in a erectile dysfunction model in mice in the first hour. Moreover, MSCs can be found in kidney, prostate and hepatic tissues up to 7 days after injection. Finally, when injected into the ovaries, MSCs are able to distribute to the uterus, with no systemic distribution Table 5 [82].

Injection of MSCs into the Central Nervous System
There is a wide variety of reports concerning the injection of MSCs into the central nervous system Table 6: Intrathecal, intracerebral and intraventricular injections have been described: (a) Intrathecal injection of MSCs: After intrathecal injection, Barberini et al. [83] demonstrated that MSCs do not seem to distribute cranially (when injected in the lumbosacral area), whereas they can progress caudally (when injected in the altanto-occipital area).
In this study, no MSC engraftment was demonstrated. The systemic biodistribution of the MSCs was not specifically assessed, but imaging techniques did not show the presence of MSCs in areas other than the central nervous system. In contrast, Kim et al. [84] demonstrated that MSC migration from the spine to the brain is possible in a dose-dependent manner. Quesada et al. [85] also demonstrated brain migration after intrathecal injection; (b) Intracerebral injection of MSCs: Wang et al. [86] demonstrated that intracerebrally injected MSCs loaded with paclitaxel are capable of spreading from one cerebral hemisphere to another in a glioma model in mice in two days. These cells were found to spread from the healthy hemisphere to the glioma hemisphere and to invade the tumor. The ability of MSCs to migrate from one hemisphere to another has also been demonstrated in other studies [87]. In other reports [88][89][90][91][92][93], MSCs injected intracerebrally were detectable at the site of administration 1-3 weeks after injection, with a subsequent rapid decrease and no significant systemic distribution. Other studies [94] showed that MSCs can be detected with fluorescence and bioluminescence up to 7 weeks after transplantation; (c) Intraventricular injection of MSCs: Some studies showed that MSCs injected into cerebral ventricles are able to migrate to large blood vessels in a brain traumatic injury model [95], and also to brain parenchyma and the spinal cord [96]. In contrast, other reports [97] demonstrate that after intraventricular infusion, MSCs do not migrate to brain parenchyma and are hardly able to migrate to the spinal cord in a model of amyotrophic lateral sclerosis.
Finally, one review showed that intranasal delivery of MSCs led to significant intracerebral migration of MSCs [98].

Injection of MSCs into the Digestive System:
(a) Intrahepatic and intrasplenic injections have been studied in several reports as efficient delivery routes for administrating MSCs. After intrahepatic injection, Xie et al. [99] demonstrated that MSCs remain in the liver and are cleared in a short period of time, without systemic distribution. This short period of time might be in association with NK cell activation: Liu et al. [100] showed that mice with activated NK cells had a more rapid clearance of intrahepatic MSCs. Yaochite et al. [101] injected MSCs into the liver and spleen of diabetic mice. It was shown that intrasplenic MSCs were able to move to the liver whereas intrapancreatic cells remained at the site of the injection.
No systemic distribution was shown and cells remained for up to 8 days. Similar results were found in another study [102], with MSCs remaining for up to 4 weeks; (b) When injected intraperitoneally [103,104], MSCs seem to spread mostly to abdominal organs (liver, spleen and intestine) with little distribution to the lungs, heart, blood and lymph nodes. Other study shows that Wharton's Jelly MSCs are capable of distributing to the whole body after intraperitoneal injection at days 1, 7, 14 and 21 in piglets [105]. (c) When injected in the peri-fistula area [106], MSCs do not seem to distribute systemically.
3.1.9. Injection of MSCs into the Cardiovascular and Respiratory Systems Some articles have addressed the injection of MSCs into the pericardium or the myocardium. When injected intrapericardially [107] in a myocardial infarction model, MSCs seem to distribute to ventricles and atriums, with a preference for the left ventricle. Regarding intramyocardial injection, MSCs seem to distribute initially in the myocardium, with posterior redistribution to the lungs, liver and bone [108]. Moreover, it has been demonstrated that after intramyocardial injection, if a repeated ischemia model is performed, MSCs tend to home mainly to the heart with less distribution to peripheral organs [109]. Finally, some studies were related to the injection of MSCs into the respiratory system: When injected intratracheally or intrabronchially, MSCs do not distribute systemically [110,111].

Injection into the Skin, Subcutaneous Cellular Tissues and Lymph Nodes
Few studies have addressed the issue of biodistribution after intradermal injection of cells (Table 5). When injected into the skin of mice, Tappenbeck et al. [112] demonstrated that MSCs seem to remain in the skin and migrate to lymph nodes, without significant systemic distribution. Regarding the specific distribution in cutaneous wounds [113,114], MSCs seem to distribute with a diffuse pattern initially and later concentrate towards the wound edges. Finally, these cells seem to be engrafted with the newly developed skin tissue. No systemic distribution following intradermal injection had been reported. Only one study was performed to describe biodistribution after intranodal injection; in this study, most MSCs remain at the injection site or in the fat surrounding the injected nodes 48 h later [103], without systemic distribution of cells.

Biodistribution of MSCs in Humans
Only a few reports of the biodistribution of MSCs after the injection into human models have been recorded in this review (Table 7). These articles will be discussed in the following sections.

Distribution of MSCs after Intravenous Injection in Humans
Three studies regarding the intravenous injection of MSCs into humans were identified in order to assess biodistribution. In the first study, the intravenous infusion of MSCs in patients suffering from cirrhosis showed an early (pre-48 h) distribution mainly in the lungs, with a later decrease in lung captation and a high distribution in the spleen and liver [115]. In another study on breast cancer patients, MSCs were monitored in peripheral blood after intravenous injection [116], finding a rapid clearance of MSCs in blood, with no cells detected 1 h post injection. Finally, a third article showed that when injected intravenously into a patient with hemophilia A [117], MSCs distributed early to the lungs and liver, with a progressive decrease. Distribution to the usual bleeding places was shown at 24 h.
As can be seen, biodistribution after IV injection in humans seems to be similar to that described in animal models: (a) early captation in the lungs; (b) later distribution in organs such as the spleen and liver; and (c) distribution of MSCs into target areas have been described.

Distribution of MSCs after Intraarterial Injection in Humans
Only one study addressed the intraarterial infusion of MSCs in humans. This study was performed on 21 type 2 diabetes mellitus patients. MSCs were selectively injected intravenously or intraarterially (into the pancreaticoduodenal artery and the splenic artery). MSCs were labeled with 18-FDG and PET-TC images were used to assess biodistribution.
Selective intraarterial delivery led to MSCs homing to the pancreas head (when cells were injected into the pancreaticoduodenal artery) or body (when infused into the splenic artery); whereas no MSCs were found in the pancreas in the intravenous group. This report shows that biodistribution after IA infusion of MSCs seem to be similar to biodistribution in animal models, with systemic delivery, a lack of lung trapping and the possibility of selective infusion into certain areas.

Distribution of MSCs after Intralesional Injection in Humans
Only one study of intralesional injection of MSCs and their biodistribution was observed. Henriksson et al. [118] injected MSCs into intervertebral discs in 4 patients. Discs were explanted at 8 and 28 months post injection. Histologic examinations found the presence of MSCs in intervertebral discs after 8 months, with chondrocyte-like differentiation. No cells were found in the discs after 28 months, and no systemic distribution was assessed.

Distribution of MSCs after Intracoronary Injection in Humans
In one study, biodistribution of MSCs after intracoronary injection was assessed. Lezaic et al. [119] injected MSCs into the coronary arteries of 35 patients with idiopathic dilated cardiomyopathy. They showed that a very low number (0-1.25%) of MSCs are retained in the myocardium, with the majority distributed to the liver, spleen and bone marrow.

Which Cell-Marking Techniques Have Recently Been Used in Preclinical Studies?
A wide variety of cell-marking techniques have been used for preclinical studies involving cell therapy. Also, a wide variety of detection methods have been performed. Since it is not the objective of this review to address these techniques in depth, an overview of them is reviewed hereafter.
Most common cell-marking techniques can be divided into: (a) those related to the use of radionuclides; (b) those related to bioluminescence imaging systems; (c) those related to the use of magnetic resonance imaging (MRI); and (d) those related to the genetic marking of cells.
The use of radionuclides is a common technique which is useful to assess the distribution of previously marked cells in preclinical models. Some of the most common radionuclides include 99m Technetium-hexamethylpropyleneamine oxime ( 9m Tc-HMPAO) [21,83], or 111 Indium-Oxine ( 111 In-Oxine) [60,115]. After culture, these substances enter into the cells. Once the cells are administered, the emitted radioactivity can be detected by imaging techniques such as Single Photon Emission Computed Tomography (SPECT) or Positron Emission Tomography (PET), which allow us to track the fate of the cells. The main disadvantage of these methods are the limited duration of the radioactivity, which limits the assessment of the late distribution of cells, and the dangers related to the management of radioactive substances in the laboratory.
Bioluminescence imaging systems are based on the light which is emitted by cells which have been previously transfected with the firefly luciferase gene (luc gene) [99,100]. Once the cells transfected with this gene are administered to the animal, an injection of Dluciferin is performed. After D-luciferin has been administered, cells containing the specific gene fluoresce, emit light with a wavelength of 537 nm. This light can be detected by specific imaging systems to assess biodistribution. The main disadvantage of this method is that bioluminescence has limited spatial resolution and reduced tissue penetration due to the relatively weak energy of emitted photons. For these reasons, its use in large animal models is not advisable [120].
The use of superparamagnetic iron oxide nanoparticles (SPIONs) is also a useful technique to assess cell biodistribution. SPIONs are small synthetic iron particles which are coated with certain biocompatible polymers. When cells have been labeled with these particles, they are detectable by imaging techniques such as MRI techniques [121]. Given that magnetic resonance imaging is a technique that is not very accessible, the use of this method can be limited.
Finally, it is possible to label cells with specific genes that can be subsequently detected by PCR methods [26,44,105]. The main disadvantage of this cell-marking technique is that a tissue sample is required so that the distribution of cells cannot be assessed in vivo in most cases.

Which Cell-Marking Techniques Have Recently Been Used in Studies with Humans?
The studies included in this review used different cell-marking techniques. The most common techniques are those related to the use of a radioactive labeling: MSCs can be labeled with radionuclides in vitro, and then injected into humans. Radionuclides used in the reviewed articles include 111 Indium-Oxine ( 111 In-Oxine) [115,117], 18-Fluorodeoxyglucose ( 18 F-FDG) [122] and 99m Technetium-hexamethylpropyleneamine oxime ( 9m Tc-HMPAO) [119]. Cells are incubated in culture mediums containing these substances, which enter into the cells. Later assessment of the emitted radioactivity of these substances in the body with imaging techniques such as scintigraphy, Single Photon Emission Computed Tomography (SPECT) or Positron Emission Tomography (PET) allow us to detect the distribution of the cells in the body to be evaluated. The main disadvantage of radionuclide-based cell-marking techniques is the limited duration of the radioactivity; as the radionuclides disintegrate, the emitted signal becomes smaller and finally disappears, making it impossible to evaluate the late biodistribution of the cells administered.
Labeling cells with markers which can be detected in histologic samples is another technique used in humans. In the study reviewed [118], iron sucrose was used to label cells. This compound makes cells detectable in histologic samples. The main advantage of this kind of marker is its presumably long duration (longer than radionuclides). However, the use of histologic markers makes it necessary to perform ex vivo examination which is a limiting factor for its use in humans. The use of flow cytometry [116] to evaluate cell markers could be considered to be comparable to the use of histologic markers, and involves the extraction of biologic samples to evaluate cell distribution.

Discussion
Determining the fate of MSCs after administration is a major issue in the development of cell therapies. On one hand, as a part of their physiological functions, MSCs are able to produce several soluble substances and to modulate the immune response through different pathways; the production and induction of interleukin production and the release of microvesicles [123][124][125]. Cell interactions lead to the secretion of soluble factors and cell-to-cell contact which induces changes in the immunobiology of immune cells, such as changing the interleukin production, inducing anergy or triggering apoptosis. On the other hand, MSCs have been found to be able to differentiate into different cellular subtypes, which could play a role in regenerative medicine. Whether MSCs act by modulating the immune system or differentiating into tissular cells, understanding how and where cells are distributed after being administered by each route of administration is critical.
Intravenous injection of MSCs remains the most widely used form of administration. The widespread use of this route of administration for drugs which are used in clinical practice, and the ease of administering cells by this route, are probably the reasons why. As previously seen, IV administration might be beneficial when cell trapping in the lungs, liver or spleen is pursued, or when MSCs are capable of acting at a distance. However, intraarterial delivery might be of choice when wide systemic distribution into different tissues and organs is required. Moreover, if a targeted deposition of cells into a single organ is needed, intraarterial selective delivery could be the solution. In contrast to intravascular administration, intramuscular injection seems to lack significant systemic distribution of cells and might be preferred when the cells do not necessarily need to reach the target tissues.
Regarding intralesional administration of MSCs, there are several distribution patterns depending on the organ or tissue injected. Intraarticularly injected MSCs seem to remain in the joints, which could be of benefit when treating articular diseases. Administration into skin, lymph nodes, trachea, lungs and urinary tissues does not produce systemic distribution either, and might be useful when targeted delivery is required. In contrast, intrahepatic, intrasplenic, intracardiac and intrapericardial infusion led to a distribution of MSCs following the natural direction of the bloodstream. Moreover, the injection of cells into virtual anatomical cavities containing corporal fluids seems to produce a distribution of MSCs within the same anatomic areas: intraperitoneal, intra-cerebro-ventricular and intrathecal routes of administration make MSCs reach the organs and tissues in contact with the correspondent fluid. Finally, intracerebrally administered MSCs are able to move within the brain if induced to do so by appropriate stimuli.

Limitations and Future Studies
Although the fate of MSCs after each route of administration seems to be fairly well understood, the specific mechanisms which lead to these distribution patterns are still a matter of discussion. Moreover, as has been previously reviewed [30,46], these mechanisms might be modulated by specific factors such as the surface molecules expressed by MSCs or the age of the donors and recipients of cells. Although this is still unknown, other factors, such as the specific subtype of MSC or the donor and recipient model, could also be important. Moreover, there is great variability among different studies with respect to the exact forms of administration (e.g., the exact anatomical site injected or the concentration or volume of cells administered). The design of standardized protocols for mesenchymal cell administration could lead to less variability of results, making them more comparable.

Conclusions
Biodistribution of MSCs in animal models and humans appears to be comparable. In response to the research questions, some facts are worth noting: (a) Intravenous administration leads to an initial accumulation of cells in the lung with later redistribution to the liver, spleen and kidneys; (b) Intraarterial injection bypasses the pulmonary filter, so MSCs distribute more widely into the rest of the organs of the body; (c) In both of the two previous routes of administration, selective perfusion of selected blood vessels is useful for targeting specific organs; (d) MSCs are not distributed systemically in significant quantities after intramuscular, intraarticular, intradermal, intranodal, intratracheal, intrapulmonary and intraurinary tissue administration; (e) The injection into specific organs, such as the liver, spleen, pericardium or heart leads to a distribution of MSCs following the direction of the natural bloodstream; (f) The injection into anatomical cavities containing body fluids (cerebral ventricles, subarachnoid space and peritoneum) leads to a distribution of MSCs in tissues which are in contact with the fluid; (g) MSCs injected intracerebrally seem to be able to migrate within the central nervous system. Mongrel dogs [24] Miocardial infarction (7 animals) Intravenous (allogenic MSCs) 111 In oxine-labeled MSCs colabeled with ferumoxides-poly-l-lysine. Single-photon emission CT (SPECT) and x-ray CT (SPECT/CT) and MRI studies were used to evaluate the distribution.
Imaging was performed immediately after injection and at multiple time points between 1 and 7 days after infusion. Early imaging showed a high distribution to lungs, which later decreased drastically. After day 1, MSCs distributed from lungs to different organs (kidney, bone marrow, liver, spleen) and also to the infarcted area.
A high and early distribution to lungs is showed, with a progressive decrease of MSCs and a later redistribution to a wide variety of tissues.
Scintigraphy was performed 2 h after cell injection and ex vivo radioactivity was evaluated 24 h after cell transplantation. MSCs were mainly distributed to the lungs, kidneys, spleen and liver. Brain captation was low but it was relatively higher in the damaged hemisphere.  89 Zr-oxine and luciferase were used to label MSCs. PET-CT, bioluminescence and ex vivo radioactivity measures were used to assess biodistribution.
PET-CT at 1 h and 1, 2, and 7 days post-injection. At 7 days, radioactivity was measured from ex vivo organs. The majority of signal (60%) was found in the lung at 1 h before decreasing, while liver signal increased. From 1 to 7 days post-injection, the proportion of the 89  Human MSCs were used. Tisular PCR analyses (blood, bone marrow, brain, spinal cord, spleen, kidney, liver, heart, lung, gonad) were used to assess biodistribution.
Harvesting of tissues was performed at 24 h, 7 days, and 30 days after injection. MSCs were broadly detected both in the brain and several peripheral organs, including the liver, kidney, and bone marrow, of both species, starting within 7 days and continuing up to 30 days post-transplantation. MR imaging of the liver was carried out before and 1, 3, 7 and 12 days after injection. Liver, lung, kidney, muscle and heart tissues were harvested at 1, 7, 15 and 42 days after cell injection. Dual-labeled MSCs were retained in the fibrotic liver of rats. SPIO particles and EGFP-labeled BMSCs showed a different tissue distribution pattern in rats with liver fibrosis at 42 days after transplantation.
SPIO-based MR imaging may not be suitable for long-term tracking of transplanted BMSCs in vivo.
Kim et al. [36] (  Ex vivo examination was performed 7 days after injection. The green fluorescent protein-expressing AD-MSCs were clearly detected in the lung, spleen, and injured spinal cord; however, these cells were not detected in the liver and un-injured spinal cord. Li et al. [45] (2015) Mice Acute liver injury MSCs were labeled with luciferase. Bioluminiscece images were used to assess biodistribution.
Images were taken at 3 h, and at 1, 3, 7, 10, 14 and 21 days. After IVC infusion, MSCs were quickly trapped inside the lungs, and no detectable homing to the liver was observed. By IH injection, lung entrapment was bypassed, but MSCs-R distribution was only localized in the injection region of the liver. After SMV infusion, MSCs-R were dispersedly distributed and stayed as long as 7-day post-transplantation in the liver.
SMV is the optimal MSCs delivery route for liver disease.
Tissue samples were collected and analyzed at 60, 120 and 240 min and 2, 4 and 8 days after infusion. Distribution up to 240 min was detected mostly in liver, and also in lungs and spleen.
The number of cells detected at 2, 4, and 8 days was less than 0.06% of the total cells infused on day 0 and were mainly distributed also in lungs, liver and spleen but relatively higher captation was seen in the rest of the tissues studied.
Authors conclude that studies using bioluminescence to track cells underestimate cell retention in the liver because of its high tissue absorption coefficient MSCs were 111 In-tropolone labeled. Imaging with SPECT (in vivo) and gamma-counter (ex vivo) was performed to assess biodistribution.
Imaging and gamma-counter studies were performed at 24 h and 48 h post infusion. In Alzheimer's model, brain uptake of MSCs was significantly higher than in healthy animals. In both groups, MSCs distributed mainly to lungs, liver and spleen.
Distribution to brain seem to be higher in Alzheimer's models.
Leibacher et al. [28] ( Human MSCs were injected and PCR techniques were used to assess biodistribution. Mice were euthanized at 1, 3, 12, or 24 h and at 1, 4, or 13 weeks post injection. MSCs were detected soon in the lungs and disappeared before 1 week post injection. Then, MSCs were found mainly in the liver. No MSCs were found in other tissues (testis, ovary, spleen, pancreas, kidney, adrenal gland, thymus, and brain). MSCs were injected and a variety of techniques, including magnetic resonance imaging, immunohistochemistry, fluorescence in-situ hybridization, and quantitative polymerase chain were performed to assess biodistribution.
Assessment was focalized in the brain area. No evidence for immediate migration of infused MSC into the central nervous system of treated mice was found.
Kim et al. [30] (2016) Rats Healthy rats Intravenous (allogenic MSCs) MSCs were surface-modified with HA-wheat germ agglutinin (WGA) conjugate for targeted systemic delivery of MSCs to the liver and labeled with fluorescent dyes. Histologic examinations were performed.
Assessment was performed at 4 h post injection. Lungs and livers were collected. HA-WGA-MSCs had a greater distribution to the liver when compared to control MSCs, which were mainly trapped in the lungs.
HA-WGA conjugate has great potential to deliver MSCs to the liver efficiently within a short time and to reduce the entrapment of MSCs in the lung.

Intravenous (allogenic MSCs)
Fluorescein isothiocyanatedextran was used to label MSCs. Histological analyses and qPCR were used to assess biodistribution.
Assessment was performed immediately after cell injection, 2, 24, and 48 h later. Lung, heart, spleen, kidney, brain, and liver were collected. MSCs accumulated mainly in the lungs of control and diseased mice, with minor amounts distributed to other organs up to 2 h. Diseased animals showed less early distribution to lungs and higher distribution to the rest of the organs when compared to healthy animals.
Acute distress respiratory syndrome might lessen the pulmonary capillary occlusion by MSCs immediately following cell delivery while facilitating pulmonary retention of the cells. Genetic tests and histology were assessed after 28 days. Transplantation of MSCs obtained from old mice showed biodistribution only in the blood and spleen in both young and old mice. MSCs obtained from young mice showed a wide distribution in young receptors (lung, axillary lymph nodes, blood, kidney, bone marrow, spleen, liver, heart, and brain cortex). In contrast, these cells showed distribution only in the brain cortex in old mice.
Authors conclude that aging of both the recipient and the donor MSCs used attenuates transplantation efficiency.
Ohta et al. [37] (2017) Rats Spinal cord injury Intravenous (allogenic MSCs) MSCs were labeled with 3 H-thymidine. Histologic and radioactivity examination of the spinal cord segment containing the damaged region, blood and target organs were harvested. After 3, 24 and 48 h, organs were collected and radioactivity measured. The highest radioactivity was detected in the lungs 3 h after infusion, while radioactivity in the injured spinal cord was much lower. However, brain radioactivity was lower than damaged spinal cord.
MSCs distribute to the injured spinal crod.
Ex vivo assessment of lungs, heart, spleen, kidneys and liver was performed at 30 min, 1 day, 3 days and 7 days following injection. MSCs distributed to the lungs up to day 1; and to the liver up to day 3, with progressive subsequent decrease. No significant distribution was observed to heart, spleen and kidneys MRI was performed at days 0, 3, 7 and 14 after cells transplantation. Histological analyses were performed immediately after the MRI examination. MSCs were detected in the liver tumors, rather than the non-tumor liver tissue and other organs. At day 3, MSCs were mainly in the central part of the tumor, showing a posterior distribution in the periphery.

Intravenous and intraarterial (allogenic MSCs)
Feridex (Berlex Imaging) mixed with the transfection agent poly-l-lysine. Later evaluation with MRI and necropsies.
Imaging was performed before and after the infusion (2 to 24 h after). After intraarterial infusion, MSCs were detected in the brain of the rats. After intravenous infusion, no MSCs were detected in the brain. Imaging was performed on days 1, 3 and 6. RT-PCR was performed in kidney, lung, liver tissue and blood on day 6. Bioluminescence showed a high distribution of MSCs to lungs on day 1, which disappeared on days 3 and 6. RT-PCR on day 6 showed variables amounts of MSCs-mRNA in blood, liver and kidneys RT-PCR seems to be a more sensitive technique to demonstrate the late presence of MSCs in different tissues when compared to bioluminescence.  MRI was performed at days 0, 1, 3, 5, 7 and 9 after cell injection. Histological analysis was performed at days 8, 12 and 18. Intravenous injection did not lead to accumulation of MSCs in the tumor. However, intralesional and intraarterial injections showed a rapid accumulation of MSCs in the core of the tumor with a gradual decrease of the cells in the zone.

Intravenous injections does not lead to MSCs migration to central nervous system tumors, whereas intraarterial and intralesional injections do.
Taylor et al. [55] (2020) Mice Renal injury model

Intravenous and intracardiac (allogenic MSCs)
MSCs were labelled with luciferase and SPIO. MRI and bioluminescence were used to assess biodistribution. Images were taken up to 2 days after injection. Following intravenous administration, no MSCs were detected in the kidneys, irrespective of whether the mice had been subjected to renal injury. After intracardiac injection, MSCs transiently populated the kidneys, but no preferential homing or persistence was observed in injured renal tissue. Intracardiac MSCs distributed to the brain, heart, lungs, kidney, spleen and liver, with also a majority of cells distributing to the lungs.
Intracardiac injection led to a wide distribution of MSCs to peripheral organs. SPECT images were acquired 20 min, 3 h, and 6 h postinjection, after which rats were sacrificed for ex vivo examinations. The majority of the cells were located in the brain and especially in the ipsilateral hemisphere immediately after cell infusion. This was followed by fast disappearance. At the same time, the radioactivity signal increased in the spleen, kidney, and liver.
Human MSCs had faster clearance from the brain than rats MSCs.    111 In-oxine. Scintigraphy was performed to assess biodistribution.
Scintigraphic images were taken immediately after injection and at 1, 2, 24, 48 h and 1 week. Immediately after injection, MSCs were trapped in the capillary network of the limb and in the lungs. Subsequently, MSCs were also mainly in the injected limb, with a decrease in the lung captation and a relative increase in the liver captation. MSCs were able to reduce the contralateral inflammation and to lower the TNF-alfa serum levels without distributing systemically. MSCs were labeled with quantum dots with near-infrared properties. Near-infrared fluorescence imaging was used to assess biodistribution.
Imaging was performed at days 1, 3, 7, 11, 14, and 17. MSCs did not distribute systemically. MSCs tended to migrate from the joint to the place of the lesion. MSCs were labeled with luciferase. Bioluminescence imaging was performed to assess biodistribution.
Imaging was performed at 2 h, 24 h, 2, 4 and 5 days. After intraarticular injection, no distribution of MSCs was detected. After intravenous injection, most MSCs were trapped in the lungs and disappeared within 24 h. After intraperitoneal injection, MSCs were localized in the injection site without distribution up to 5 days.
In vivo imaging was performed up to 70 days weeks. MSCs were detected in the injected join up to 9 weeks. No systemic distribution was observed.
MSCs seem to stand long times in the injected joint with no systemic distribution. Human MSCs were injected and qPCR tests were used to assess biodistribution in the different organs.
At 15 min and 8 h after injection, samples were collected from eight organs (spleen, kidney, liver, lymph nodes, muscle, lung, heart, brain). Blood concentrations were also monitored. After intravenous injection MSCs were detected immediately in blood, with a progressive decrease. After intraarticular injection, MSCs were detected in blood with a peak at 8 h.
No systemic distribution was observed after intraarterial delivery. After intravenous injection, most MSCs were trapped in the lungs. MR images were acquired at injection and at 1, 4, 8, and 12 weeks. Ex vivo histological examination was performed at 12 weeks. MSCs were found in the joint up to 12 weeks, without systemic distribution. MSCs were labeled with DiR fluorescent dye and iron nanoparticles. MRI and fluorescent imaging were used to assess biodistribution. Histological exams were also performed.
Bioluminescence imaging was performed immediately and 1, 3, 7, 14, 21, and 28 days after cell transplantation. At day 28, organs were collected for ex vivo analyses. After intraarticular injection, MSCs remained in the joint. The use of the magnet led to magnetic MSCs accumulation in the target lesion.
The use of a magnet during magnetic-labeled MSCs transplantation can lead to selective accumulation of cells into the cartilage defects. Table 5. Biodistribution after intralesional administration (except for intra-central nervous system) of MSCs in animal models.

Cell-Marking Technique Detection Time and Outcome Comments
Dave et al. [54] (2017) Mice Chronic bowel inflammation Intra-cardiac (xenogenic MSCs-human MSCs) MSCs were labeled with luciferase and red fluorescent protein.
In vivo and ex vivo bioluminescence and histologic examinations were performed to assess biodistribution.
Images were taken up to 24 h after injection. Histology was performed at 24 h post injection. MSCs in healthy mice distributed mainly to lungs, spleen and liver. In contrast, MSCs in diseased mice were located mainly in the intestine, with low pulmonary captation.
After intracardiac injection, MSCs are able to distribute mainly to the inflamed intestine.

Intra-myocardial (allogenic MSCs)
MSCs were harvested from male rats and injected into female rats. qPCR was performed in different tissues to assess biodistribution (heart, lungs, spleen and liver) Examinations were performed 3 weeks after injection. MSCs had a greater homing in heart and a lower distribution to peripheral organs when repeated ischemia was applied. MRI was performed at days 3, 5 and 7. MSCs were detected to home mainly in the left ventricle. They were also detected in the right ventricle, and both atriums.
After intrapericardial injection, MSCs distribute mainly to left ventricle.

Pigs and mice
Osteonecrosis of the femoral head

Intraosseous (xenogenic MSCs-Human
MSCs) Human MSCs were injected and qPCR, cytometry and histologic analysis was performed to assess biodistribution in different tissues (Femoral head, adyacent tissues, liver, kidneys, spleen, and lungs).
Tissues were collected at either 30 min or 24 h after injection.
No MSCs were detected in other organs apart from the injection site.
Tendons were recovered post mortem at 1 day, and 1-2, 4, 12 and 24 weeks after MSC injection. MSCs distributed throughout the tendon synovial sheath but restricted to the synovial tissues, with no MSCs detected in the tendon or surgical lesion. After day 14, no MSCs were detected. Tracking techniques were performed up to 24 weeks after injection. Labeled cells could be traced at their injection site by MRI as well as histology for the whole follow-up period of 24 weeks. Furthermore, small numbers of labeled cells were identified in peripheral blood within the first 24 h after cell injection and could also be found until week 24 within the contralateral control tendon lesions that had been injected with serum Ryska et al. [ Different assessments were performed at 0, 1, 4 and 10 days after injection.
No positive Alu-stained nuclei were observed in urethras at 4, 10, and 14 days. PKH26-labelled cells were found in all urethras at 2 and 24 h. Bioluminescence study showed increased luciferase expression from day 0 to 1 following injection, with a progressive disappearance until day 7.
No MSCs were detected in periurethral tissue after intravenous injection. MSCs were detected for less than 7 days in periurethral tissues after local injection.  Human MSCs were injected and genetic tests (quantitative PCR) were done in tissue samples: blood, skin/subcutis and skeletal muscle at the injection site, lymph node, liver, spleen, lungs, brain, femur bone, and bone marrow, kidneys, thymus, thyroid/para-thyroid gland and ovaries or testes) to evaluate biodistribution.
After intradermal injection, mice were sacrificed at 1 week, 3 months and 4 months. After intravenous injection, mice were sacrificed. After intradermal injection, MSCs were detected in the skin up to 3 months and also in draining limph nodes after 1 week. No MSCs were detected in any other tissues. After intravenous injection, MSCs were detected mainly in the skin and muscle near to the injection site and also in the lungs on day 8. After 1 month, most MSCs were in the lungs. MSCs were also detected in low quantities in kidney and thymus after 1 month. MSCs were labeled with PKH26. Fluorescent microscopy was performed to assess biodistribution within the wound.
Skin samples were collected from respective wounds on 3, 7, 10 and 14 days. MSCs demonstrated a diffuse pattern of distribution initially and were later concentrated towards the wound edges and finally appeared to be engrafted with the newly developed skin tissue. MSCs (in this case, isolated from Wharton's Jelly) were labeled with SRY sequences and PKH26-labeled Ex vivo evaluation was performed with qPCR and confocal microscopy. Tissues were collected from the heart, lung, pancreas, liver, kidney, omentum, stomach, intestine, uterine horn, ovary, muscle, and bone marrow.
Biodistribution was assessed at 6 h, 24 h, and 7, 14 and 21 days after administration. All tissues were positive for MSCs for 1-day-, 1-week-, 2-week-, and 3-week-old recipients. MSCs were labeled with luciferase, red fluorescent protein and herpes simplex virus-1 thymidine kinase. PET, CT, bioluminescence imaging and histological analyses were performed to assess biodistribution.
Images and ex vivo analysis were collected for weeks 1 to 4. The intrahepatic group showed a confined signal at the injection site, while the intrasplenic group displayed a dispersed distribution at the upper abdominal liver area, and a more intense signal. MSCs were labeled with luciferase. Bioluminescence was used to assess the biodistribution.
Biodistribution of MSCs was measured in the main organs (liver, spleen, intestine, lungs, heart and blood) and lymph nodes (LNs, inguinal, popliteal, parathymic, parathyroid, mesenteric, caudal and axillary) 48 h after injection. Most MSCs distributed to abdominal organs (liver, spleen and intestine), with few remaining in lymph nodes, lungs, blood and heart. Biodistribution did not change significantly between healthy and diseased mice.
Intraperitoneal injection seems to lead to abdominal spreading of MSCs. Both administration routes are convenient for treating acute respiratory syndrome.  MSCs were double labeled with fluorescent nanoparticles and Hoechst-33258. Bioluminescence and histologic examinations were used to assess biodistribution.
In vivo and ex vivo analyses were performed at 1, 7, 21 days. By intravenous administration cells were sequestered by the lungs and rapidly cleared by the liver. MSCs transplanted in lateral ventricles remained on the choroid plexus for the whole duration of the study even if decreasing in number. Few cells were found in the spinal cord, and no migration to brain parenchyma was observed  Green fluorescent protein MSCs (GFP-MSCs). Imaging techniques and histology were used to assess biodistribution in blood vessels.
Techniques were performed at 10, 14, and 17 days. MSCs were found to home in large arteries (thoracic aorta, abdominal aorta, common iliac artery) 10, 14, and 17 days after transplantation.
MSCs seem to distribute after brain injury when injected intraventricularly.
Lee et al. [89] ( CM-Dil staining was used to label MSCs, which also contained Paclitaxel. Confocal laser-scanning microscopy was used to assess the distribution of MSCs. Later histological examinations assessed the distribution of MSCs within the brain.
Necropsies were performed 2 days after MSCs injection. MSCs were distributed in clusters in the injection area, and were also found within the glioma.
MSCs seem to spread within a short period of time from one hemisphere to another, after intracerebral injection. MSCs were labeled with SPION and PKH26. MRI imaging and histology were performed to assess biodistribution.
The distribution and migration of MSC were analyzed by MRI from day 1 to day 15, and histological methods on days 1, 2, 3, 7, and 15. After intracerebral injection, MSCs moved to corpus callosum and blood vessels. After intraarterial injection, most MSCs were detected in the ipsilateral hemisphere and most of them within the blood vessels.  MSCs were detected for several hours post-infusion in peripheral blood. MSCs are rapidly (less than 1 h) cleared from peripheral blood after intravenous infusion The presence of MSCs in peripheral blood was not detected after 1 h post-infusion.
MSCs were detected at 2 h, 4 h, 6 h, 24 h, 48 h, 7th and 10th days after infusion. Pre-48 h images showed a large majority of cells distributed in the lungs. Later images showed a drastic decrease in lung captation, with a higher amount of MSCs distributed in the spleen and few in liver.
There was a clear initial biodistribution in lungs, which decreased after 48 h. Spleen captation was higher than liver captation, maybe due to splenomegaly. MSCs were labeled with iron sucrose (Venofer ® ). Histologic examinations were performed to detect the cells.
Intravertebral discs were explanted at 8 months (3 patients) and 28 months (1 patient) post injection. MSCs were detected at 8 months, but not at 28 months. Detected MSCs had differentiated into chondrocyte-like cells.
MSCs seem to home in intravertebral discs after intralesional injection for long periods of time.
Imaging was performed 1 h and 18 h after transplantation. At 1 h after intracoronary administration, the majority of MSCs accumulated in the liver, spleen and bone marrow. Accumulation of MSCs in the myocardium ranged from 0 to 1.45% of injected activity in the field of view. The distribution of labeled stem cells in the myocardium corresponded to the area supplied by the vessel used for administration. At imaging 18 h post injection, the distribution of labeled stem cells appeared unchanged, but with decreased activity.
The retention of MSCs in the myocardium is low after intracoronary injection. Institutional Review Board Statement: Not applicable because no humans or animals were present in a systematic review.