Therapeutic role of adipose tissue–derived stem cells versus microvesicles in a rat model of cerebellar injury

Abstract Monosodium glutamate (MSG) is a controversial food additive reported to cause negative effects on public health. Adipose stem cells (ASCs) and their derived vesicles (MVs) represent a promising cure for human diseases. This work was planned to compare the therapeutic effects of adipose stem cells and microvesicles in MSG‐induced cerebellar damage. Forty adult healthy male Wister rats were equally divided into four groups: Group I (control group), group II (MSG‐treated), group III (MSG/ASCs‐treated), and group IV (MSG/MVs‐treated). Motor behaviour of rats was assessed. Characterization of ASCs and MVs was done by flow cytometry. The cerebellum was processed for light and electron microscopic studies, and immunohistochemical localization of PCNA and GFAP. Morphometry was done for the number of Purkinje cells in H&E‐stained sections, area per cent of GFAP immune reactivity and number of positive PCNA cells. Our results showed MSG‐induced deterioration in the motor part. Moreover, MSG increases oxidant and apoptotic with decreases of antioxidant biomarkers. Structural changes in the cerebellar cortex as degeneration of nerve cells and gliosis were detected. There were also a decrease in the number of Purkinje cells, an increase in the area per cent of GFAP immune reactivity and a decrease in the number of positive PCNA cells, as compared to the control. Rats treated with ASCs showed marked functional and structural improvement in comparison with MV‐treated rats. Thus, both ASCs and MVs had therapeutic potential for MSG‐induced cerebellar damage with better results in case of ASCs.

gastrointestinal tract and nervous system. The intestine is the primary site for catabolism. 3 The average daily intake of MSG is estimated to be 0.3-1.0 g in industrialized countries, which can be higher depending on the MSG content of food items and the individual taste preferences. 4 Under normal conditions, acute toxicity of glutamate is uncommon. The oral dose that is lethal to 50% of subjects (LD50) in rats and mice is 15,000-18,000 mg/kg body weight. 5 However, MSG has been tied to obesity and disorders in the respiratory, circulatory, reproductive and nervous systems. 6,7 It triggered symptoms, referred to collectively as 'Chinese restaurant syndrome' consisting of numbness at the back of the neck and arms, weakness and palpitations. 8 The young and the elderly are most at risk from MSG as the blood-brain barrier is not fully developed in the young and can be damaged by ageing or disease in the elderly. Ingredients that contain MSG are used in baby formula, allowing neurotoxins to be more accessible to the brain which is not yet well developed. 9 Exposure of rats to MSG at the neonatal stage was found to cause severe damage of the hypothalamic nuclei resulting in increased body weight, fat deposition and decreased motor activity. 10 Cell therapies are group of strategies that use live cells for healing aims. Their aims are to repair, replace and restore the biological functions of damaged tissue or organs. It can be applied in many clinical tests for the treatment of conditions as diabetes mellitus, liver disease and corneal, articular, neural and cutaneous lesions. 11 Regenerative medicine is the process of regenerating or replacing the damaged cells, tissues and/or organs restore normal function. Tissue damage caused by degenerative diseases or neoplasia remains a vast challenge to treat. 12,13 Stem cells can differentiate into many different mature cell types such as heart cells, skin cells or nerve cells under the right conditions, or given the correct signals. They carry on the great potential for regenerative medicine, particularly in replacing cells in tissues with barely intrinsic renewal capacity, including the heart and nervous tissue. 14 Mesenchymal stem cells (MSCs) are a promising alternative approach for healing human diseases. They primarily originate from the mesoderm and ectoderm during the early embryonic development and present mostly in adult tissues as bone marrow and adipose tissue, and foetal tissues and fluids as in the umbilical cord and amniotic fluid. 15 The role of MSCs in the treatment of diseases depends on their ability to mend damaged tissue and suppress inflammation.
In addition to their wide distribution, these cells are not immunogenic and can avoid immune cell recognition. 16,17 Adipose tissue-derived mesenchymal stem cells (ASCs) are unitary of the adult stem cells. They can be harvested from the stromal vascular fraction (SVF) of adipose tissues and have a great ability for multidirectional differentiation. 18 Due to the mesodermal origin of ASCs, they have the capacity for self-renewal and can also be differentiated into adipocytes, chondrocytes, myocytes, osteoblasts and neurocytes induced by a selective medium in vitro. 19 White adipose tissue is the main source of ASCs, and the MSCs derived from it have a stronger antiapoptotic ability than that from brown adipose tissue. Among the advantages of ASCs is the easy access by subcutaneous lipoaspiration which is a much less painful procedure than harvesting bone marrow stem cells. They are harvested from autologous fat, unlike embryonic stem cells. So, their use is less associated with ethical controversies. 20,21 Adipose tissue-derived MSCs can pack trophic mediators into extracellular vesicles that can transport not only pro-regenerative factors, but also mRNA and microRNA and even mitochondrial components over a long distance. 22 They play as mediators for intercellular communication with neighbouring cells by exchanging these bioactive molecules or disseminating genetic contents towards distal organs. They are classified into exosomes, microvesicles and apoptotic bodies according to their respective size, origin, biogenesis and composition. 23 Preclinical animal models showed that MSC-derived MVs (MSC-MVs) display pro-angiogenic and cellular protective effects and can be used for tissue regeneration. 24 Thus, MVs are promising therapeutic approach for various pathological conditions such as neural degeneration, liver fibrosis, kidney injury and myocardial infarction. 25 Thus, the purpose of our work was to compare the therapeutic effect of ASCs versus MVs on the cerebellar damage induced by MSG.

| Chemicals
Monosodium glutamate was supplied in the form of white powder. It was purchased from Cairo Pharma Co., Egypt.

| Experimental design
Rats were equally divided into four groups. Group I (Control group) included 10 rats that were subdivided into three subgroups: subgroup IA (4 rats) that received no treatment, subgroup IB (3 rats) that received 2 ml of 0.9% sodium chloride (NaCl), intraperitoneally, and subgroup IC (3 rats) that received 0.5 ml of AD-MSCs (1 × 10 7 cells/ml) dissolved in 0.5ml of phosphate-buffered saline (PBS), intravenous through the caudal vein. Group II (MSG-treated group) received 2 ml of 3.5 mg/g body weight/day of MSG dissolved in 2 ml of 0.9% NaCl, intraperitoneally, for 10 days. 26 Group III (MSG/ASC-treated group) Received MSG (as described in group II) for 10 days. On the 11th day, each rat had a single dosage of 0.5 ml of ASCs (1 × 10 7 cells/ml) suspended in 0.5 ml of PBS, intravenous through the caudal vein. Specimens were taken 4 weeks after stem cell administration. 27 Group IV (MSG/MV-treated group) received MSG (as described in group II) for 10 days. On the 11th day, each rat received a single dose of 200 μg MVs diluted in 1 ml of PBS, intravenous through the caudal vein. Specimens were removed 1 week after MV administration. 25 Three rats died (2 rats from group II and one rat from group IV). They were excluded from the experiment.

| Body weight
At the beginning of the experiment, rats of all groups were weighed for calculating the dose of MSG.

| General observation
During the experimental period, the general appearance of rats and the quantity of food intake were determined daily.

| Motor behaviour assessment
2.6.1 | Accelerating rotarod test 28 Rats were trained three times daily for one week before getting results. Each trial lasted for 5 min maximum, and rats were rested for 15 min between trials to avoid fatigue. The rats experienced linear acceleration from 4 to 40 rpm in 300 s. Latency to fall from rotarod was recorded in seconds ( Figure 1A).

| Balance beam test 29
This test examines the ability of the animal to remain upright and to walk on an elevated and fairly narrow beam ( Figure 1B). Score from 0 to 5 was used as follows: 0: Falls off the beam immediately, 1: Falls off the beam before completing the walk, 2:Walks the beam, but is very instable, almost falling off, may pause one or more times, and/ or takes more than 6 s, 3: Walks the beam, but is somewhat instable, may pause one or more times, takes more than 6 s to complete the walk, 4: Walks the beam, but is somewhat instable, completes the walk within 6 s, and 5: Walks the balance beam perfectly, does not need to check balance, does not pause, and completes the walk within 6 s.

| Isolation and culture of adipose-derived mesenchymal stem cells (ASCs)
Human adipose tissue was isolated from freshly subcutaneous adipose tissue samples obtained either from the abdomen or the inguinal fat. The adipose tissue was resected and placed into a labelled sterile tube containing 15 ml of phosphate-buffered saline (PBS; Gibco/Invitrogen). Then, it was re-suspended in 0.075% type collagenase II (sigma) in HBSS (approximately 2 ml/g) for enzymatic digestion, and incubated at 37°C for 60 min with shaking. At the end of digestion, 10% foetal bovine serum (FBS) (Gibco) was added to neutralize collagenase. The digested adipose tissue was passed through a 100μm filter to remove debris and centrifuged at 400 g for 10 min to obtain a cell pellet, and the erythrocytes were removed by treatment with erythrocyte lysis buffer. The cells were trans-  The microvesicles were harvested and labelled with PKH26 dye (red fluorescence cell linker). They were centrifuged and washed twice in serum free medium, then pelleted and suspended in dye solution. 36 2.14 | Detection of homing of microvesicles Cerebellar tissue was examined with a fluorescent microscope to detect and trace the cells stained with PKH26. 33

| Measurement of total oxidant status (TOS)
Total oxidative status (TOS) was measured in the cerebellar homogenates using a (Rel Assay Diagnostics kit). The assay was calibrated with hydrogen peroxide (H 2 O 2 ), and findings were expressed as μmol H 2 O 2 eq. /L.

| Antioxidants measurements in the cerebellar homogenate
The superoxide dismutase (SOD) activity in U/g cerebellar tissue was measured by the method of Ref. [38]. Reduced glutathione (GSH) level was determined according to the method of Ref. [39]. Also, glutathione reductase (GR) activity was measured as described by. 40 All the above kits produced from Biodiagnostic.

| Real-time analyses apoptotic genes in cerebellum
Total RNA was extracted from cerebellar tissue homogenate using the RNeasy extraction kit (Qiagen). The cDNA was obtained from extracted total RNA (5 μg) with 1 μl (20 pmol) antisense primer and 0.8 μl superscript AMV reverse transcriptase for 60 min at 37°C. The relative expression of mRNA was measured using SYBR Green method by Applied Biosystems. The sequences of the primers used in these assays were presented in Table 1 (Table 1).

| Light microscopic study
For light microscope study, specimens were fixed in a 10% formalin solution and processed to prepare 5μm-thick paraffin sections for H&E 43  Immunohistochemical reaction was run out on paraffin sections of the cerebellar cortex using a streptavidin system. Paraffin sections were de-paraffinized, rehydrated in descending grades of alcohol and incubated overnight with the primary monoclonal antibody.
Sections were rinsed three times with PBS, then incubated for 1 h with peroxidase-conjugated secondary antibody and washed three times with PBS. The reaction was developed with 0.05% diaminobenzidine (Dakopatts) as the substrate for peroxidase, and finally, the slides were counterstained with Mayer's haematoxylin. Negative control slides were prepared by replacing the primary antiserum with PBS. A tonsil was used as the positive control for PCNA. A brain slide was used as the positive control for GFAP.

| Electron microscopic study 45
Specimens were immediately fixed in 2.5% glutaraldehyde for 2 h and postfixed in 1% osmium tetroxide buffered with 0.1 M phosphate buffer at pH 7.4 for 1 h. Then, they were dehydrated in ascending grades of alcohol and embedded in resin to prepare semithin and ultrathin sections using a Leica ultracut (UCT) (Glienicker).

Gene Primer sequences
Annealing temperature product size

| Morphometric analysis
The following parameters were measured: Number of Purkinje cells in H&E-stained sections, number of PCNA immunoreactive cells and area percentage of GFAP immunoreactivity. Measurement was done using "Leica Qwin 500" software image analyzer computer system (Leica image system Ltd) at the Pathology Department, Faculty of Dental Medicine, Cairo University. Ten non-overlapping fields were randomly chosen for each section at ×400 magnification.

| Statistical analysis
Data management and statistical analysis were done using SPSS version 25 (IBM). Numerical data were summarized as means and standard deviations. Comparisons between the study groups were done using one-way ANOVA test. Post hoc analysis was done using Bonferroni's method. All statistical tests were two-sided. p values <0.05 were considered significant. 48

| General observation
During the experiment, the control rats were awake and alert while the MSG-treated rats were weak and lethargic and gradually improved after ASC and MV treatment.

| Accelerating rotarod test
Group II showed gradual deterioration in the latency to fall from the rotarod throughout the time of the experiment compared to the control group. Nevertheless, groups III and IV showed gradual deterioration through the 1st 2 weeks of assessment followed by some improvement along the remaining time of the experiment but did not come back to baseline values.

| Balance beam test
Rats in the control group walked easily and achieved a mean score of 5 while those in MSG-treated group were unstable and took longer to finish walking, with a mean score of 2. However, rats in group III (ASCs) showed some improvement as they were slightly unstable, but could finish walking in time with an average score of 4.
Moreover, group IV (MVs) showed some improvement but less than group III with an average score of 3.

| Identification and characterization of differentiated stem cells
Adipose tissue-derived mesenchymal stem cells (ASCs) showed adhesiveness and fusiform shape in culture (Figure 2A). They were positive for CD105 (ASC surface markers) and negative for CD106, CD14 and CD34 (hematopoietic markers) by flow cytometry ( Figure 2B).

| Detection of homing of ASCs
Cerebellar tissue showed PKH26-positive ASCs with a fluorescent microscope, indicating stem cell homing ( Figure 2C).

| Identification and characterization of MVs
By transmission electron microscope (TEM), MVs appeared as heterogeneous spheroid structures different in size ( Figure 3A). They were positive for CD63 (extracellular vesicles surface marker) by flow cytometry ( Figure 3B).

| Oxidative stress parameters (Cerebellar MDA, lipid peroxidation marker and TOS):
Our findings showed significant differences between the four groups Noteworthy, there were no significant differences of GR activity between groups. However, post hoc analysis revealed significant decreases of GR activity in group II compared to control group only.
There were significant differences of serum TAC between groups (p < 0.001). Deeper analysis by post hoc showed a significant decrease of serum TAC in groups II, III and IV compared to group I (p < 0.001, p = 0.02, p = 0.03 respectively; Table 2).

| Effect of ASCs and MVs on Apoptosis gene expression
The data of the current research demonstrated significant differences between the four groups regarding apoptotitic gene expression caspases 3, 8 and 9 (p < 0.001 for each). MNG increased the expression of these genes in the cerebellar tissue compared with the control group. This was significantly restored by ASC and MV treatment for all caspases genes expression levels. Furthermore, there was no significant dereferences concerning of caspase gene expression values between ASC-and MV-treated groups (Table 3). Thus, these finding shed the light on antiapoptotitic effects of ASCs and MVs.

| Histological results
Histological examination of the cerebellar sections of subgroups IA, IB and IC revealed no difference. Therefore, results of subgroup IA were represented as the control group.

| Light microscope results
Haematoxylin and eosin-stained cerebellar sections of the control group revealed rat cerebellar cortex with its three layers: molecular, astrocytes. The granular layer showed closely packed small granule cells with pale areas of cerebellar islands in between ( Figure 4A).
In MSG-treated group, Purkinje cells appeared dark and shrunken.
Many astrocytes were seen. Cerebellar islands were widened. The white matter showed degenerated neuropil ( Figure 4B,C). Group III showed Purkinje cells with architecture similar to that of the control group (arrowhead). Many astrocytes were seen ( Figure 4D). In group IV, Purkinje cells had architecture comparable to that of the control group. However, few cells appeared distorted. Many astrocytes were seen ( Figure 4E). Yet, few cells appeared dark and twisted. Many astrocytes were seen ( Figure 4H). In group IV, Purkinje cells showed an architecture comparable to that of the control group. Many astrocytes were seen.
The granular layer showed widened cerebellar islands ( Figure 4I).
Immunohistochemical localization of PCNA in the control group revealed intense immune reaction to PCNA in the nuclei of cells in the molecular layer, Purkinje cells and astrocytes in Purkinje cell layer, and few cells in the granular layer ( Figure 5A). In group II, the immune reaction was detected in few nuclei in the molecular layer, many astrocytes in Purkinje cell layer and few cells in the granular layer ( Figure 5B). However, in group III, the immune reaction was seen in the nuclei of few astrocytes in Purkinje cell layer and cells in the granular layer ( Figure 5C). Group IV showed positive reaction in the nuclei of a few cells in the molecular layer, many astrocytes in Purkinje cell layer and cells in the granular layer ( Figure 5D).
Immunohistochemical localization of GFAP in the control group revealed that Bergmann-glia or modified astrocytes were perpendicular to the surface and parallel to each other in the molecular layer, while in the granular layer, true astrocytes appeared irregularly ( Figure 5E). Nevertheless, group II showed an apparently increased GFAP-positive astrocytes in the molecular and granular layers ( Figure 5F). In group III, GFAP-positive astrocytes were seen

TA B L E 3 Relative apoptosis gene expression in the study groups
in the molecular and granular layers. They were nearly similar to that of the control group ( Figure 5G). Group IV showed GFAP immune reactivity in Bergmann-glia and true astrocytes ( Figure 5H).  Figure 6L).

| Electron microscope results
Group IV showed myelinated axons with intact myelin sheath were seen included mitochondria and few unmyelinated fibres with vacuolated mitochondria. A granular cell with heterochromatin clumps in its nucleus was seen ( Figure 6M). A Purkinje cell appeared with an irregular, ill-defined nucleus, dilated cisternae of rough endoplasmic reticulum and electron-dense bodies ( Figure 6N). Multiple granule cells show nuclei (N) with heterochromatin clumps, and thin rim of cytoplasm. Few cells appeared dark and shrunken ( Figure 6O).

| Morphometric and statistical results
The non-significant difference was detected between subgroups IA, IB and IC. Therefore, the average of group I was used as a control to be compared with the other groups of the study. in group I was significantly higher (6.1) than in group IV (4.5). There were no significant differences between group III, group I and group IV (Table 4 and Figure 7A).
The number of positive PCNA cells showed an overall significant difference between the four groups (p-value <0.001). Post hoc analysis revealed that the number of positive PCNA cells in group III (220.5) was significantly higher compared to all other groups. Likewise, the number of positive PCNA cells in group VI was significantly higher (165.7) than groups I (130.5) and II (104.1). There was a borderline significant difference between groups I and II, with a higher number of positive PCNA cells in group I (Table 5 and Figure 7B).
Area percentage of GFAP immune reactivity showed an overall significant difference between the four groups (p < 0.001). Post hoc analysis revealed that the area percentage of GFAP in group II (32.8) was significantly higher compared to all other groups. Also, in group IV, it was significantly higher (24.7) than groups I (4.8) and III (9.5). No substantial difference was reported between groups I and III (Table 6 and Figure 7C).

| DISCUSS ION
MSG is considered the most widely used food additive. While having huge benefits to the food industry, the ubiquitous use of this food additive could have negative consequences for public health. 6 In the present study, assessment of cerebellar function revealed deterio- pigmentosa. 65 The current study showed increased oxidative parameters, cerebellar MDA and TOS in the MSG-treated group, which was reversed in ASC-treated group and MV-treated group. In contrast, the cere-   77 In the same context, MSG was found to decrease PCNA immunoreactivity in spermatogonia and spermatocytes revealing increased apoptosis and decreased proliferation processes of the testicular germ cells. 78,79 On the contrary, Hegazy et al. 80 showed high PCNA immunoreaction in the MSG-treated cells that were explained by tendency of these cells to proliferate more to compensate degeneration and damage.
On the contrary, the immune expression of PCNA was increased in groups III and IV as compared to MSG-treated group. These results were confirmed by morphometric and statistical results.
Similarly, Cetinkaya et al. 81 and Jiao et al. 82 reported increased expression of PCNA and Ki67 in MSCs group. MSCs could markedly suppress the expression of P21, and promote the level of PCNA, thus improve function of ageing spleen and thymus and suppress oxidative stress. 83 SCs and microvesicles could help in the proliferation and regeneration of degenerated cells. 84,85 Thus, they might be recommended as an alternative therapy for treating peripheral neuropathy. 86 Our immunohistochemical results revealed intense reaction to GFAP in MSG-treated group which was statistically confirmed by a significant increase in area percentage of GFAP immune reaction, compared to the control group. Sriram et al. 87 and Baydas et al. 88 reported that any mechanical, chemical or degenerative insult to the brain stimulates astrocyte proliferation and hypertrophy with increased GFAP synthesis resulting in vigorous astrogliosis. Expression of GFAP inhibits inflammatory response after injury and limiting the damage. 89 In contrast to our results, previous researches reported toxic effect of MSG on the astrocytes themselves. Also, the toxic effects of reduced glutathione and reactive oxygen species accumulation on astrocytes have been reported, [90][91][92][93][94] which confirmed our finding regarding the oxidative stress reported in the present work.
We found that overexpression of caspase 3, caspase 8 and caspase 9 genes in cerebellar tissues in the MSG group compared to control group, which restored by ASC or MV treatment. These findings are in accordance with. 95 They proposed that cellular stress induced by MSG leads to apoptosis. Additionally, they reported that multiple and distinctive death programmes as caspase 3-dependent microvascular disruption mediated neuronal and glial cell death occur in the brain stem and disruption of blood-brain barrier integrity as well as neural connectivity as a result of MSG exposure. 95 On the contrary, the area percentage of GFAP reaction in groups III and IV were significantly lower than MSG group. In group VI, it was significantly higher than groups I and III. No significant difference was found between groups I and III. In the same context, Ou et al. 96 and Korayem et al. 29 detected GFAP-positive astrocytes, 28 days after stem cell transplantation in rats exposed to middle cerebral artery occlusion. ASCs were suggested to inhibit the activation of glia and microglial and reduce the expression levels of pro-inflammatory cytokines TNFα, IL-1β and IL-6 in the cerebellum. 97,98,99 In conclusion, Monosodium glutamate (MSG) is the most commonly used food additive. While having huge benefits to the food industry, the ubiquitous use of MSG may have negative consequences for public health. In the present study, MSG induced degenerative changes in the cerebellar cortex and functional disturbance.
Administration of ASCS and MVs were proved to have neuroprotective effects, possibly due to neural differentiation and paracrine, antiapoptotic, anti-inflammatory and antioxidant effects. However, our results revealed better results in case of ASCs. Thus, ASCs and MVs can be promising in the treatment of neural diseases to rescue the degenerating cells. However, future clinical studies are required to assure whether this treatment strategy is clinically relevant to patients.

CO N FLI C T O F I NTE R E S T
There is no conflict of interest to declare.  TA B L E 6 Area percentage of GFAP in the study groups Desoky: Conceptualization (equal); Writing -review and editing (equal). Amal S. El-Shal: Conceptualization (equal); Investigation (equal).

CO D E AVA I L A B I LIT Y
Leica Qwin 500 software image analyzer computer system (Leica image system Ltd; Cambridge, England).

CO N S E NT TO PA RTI CI PATE
Not applicable.

CO N S E NT FO R PU B LI C ATI O N
Not applicable.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data generated or analysed during this study are included in this published article.