Conditioned Medium of Human Adipose Mesenchymal Stem Cells Increases Wound Closure and Protects Human Astrocytes Following Scratch Assay In Vitro

(1)

Conditioned Medium of Human Adipose Mesenchymal Stem Cells Increases Wound Closure and Protects Human Astrocytes Following Scratch Assay In Vitro

Eliana Baez-Jurado¹&Oscar Hidalgo-Lanussa¹&Gina Guio-Vega¹&

Ghulam Md Ashraf²&Valentina Echeverria^3,4&Gjumrakch Aliev^5,6,7&

George E. Barreto^1,8

Received: 2 August 2017 / Accepted: 11 September 2017

#Springer Science+Business Media, LLC 2017

Abstract Astrocytes perform essential functions in the preservation of neural tissue. For this reason, these cells can respond with changes in gene expression, hypertrophy, and proliferation upon a traumatic brain injury event (TBI). Different therapeutic strategies may be focused on preserving astrocyte functions and favor a non-generalized and non-sustained protective response over time post-injury. A recent strategy has been the use of the conditioned medium of human adipose mesenchymal stem cells (CM-hMSCA) as a therapeutic strategy for the treatment of various neuropathologies. However, although there is a lot of information about its effect on neuronal protection, studies on astrocytes are scarce and its specific action in glial cells is not well explored. In the present

study, the effects of CM-hMSCA on human astrocytes subjected toscratchassay were assessed. Our findings indicated that CM-hMSCA improved cell viability, reduced nuclear fragmentation, and preserved mitochondrial membrane potential. These effects were accompanied by morphological changes and an increased polarity index thus reflecting the ability of astrocytes to migrate to the wound stimulated by CM-hMSCA. In conclusion, CM-hMSCA may be considered as a promising therapeutic strategy for the protection of astrocyte function in brain pathologies.

Keywords Human astrocytes . Brain injury . Conditioned medium . Mesenchymal stem cells . Migration .Scratch

Introduction

Astrocytes are the most abundant non-neuronal cells in the brain [1,2]. Their abundance provides them with essential functions such as the preservation of neural tissue [3], synaptic remodeling [4], maintenance of protective barriers [5], and cerebral homeostasis [6–8]. Moreover, astrocytes contribute to the maintenance of the antioxidant system and production of neurotrophins [5]. However, astrocytes not only perform neuroprotective functions but may also be detrimental to the maintenance of neuronal function [9]. This occurs when astrocytes undergo from a resting state to a reactive one acquir- ing a different morphology and function, with important consequences to neuronal homeostasis [10–12].

The brain tissue is frequently affected by pathologies in which glucose supply is diminished. Some of these diseases include hypoglycemia [13] or diabetes mellitus [14], and traumatic brain injury (TBI) [15–17]. Both types of pathologies have adverse effects on cognitive, sensory, and/or physical

* George E. Barreto

[email protected]; [email protected]

1 Departamento de Nutrición y Bioquímica, Facultad de Ciencias, Pontificia Universidad Javeriana, Bogotá, D.C., Colombia

2 King Fahd Medical Research Center, King Abdulaziz University, Jeddah, Saudi Arabia

3 Research & Development Service, Bay Pines VA Healthcare System, Bay Pines, FL 33744, USA

4 Fac. Cs de la Salud, Universidad San Sebastián, Lientur 1457, 4080871 Concepción, Chile

5 Institute of Physiologically Active Compounds, Russian Academy of Sciences, Chernogolovka, Moscow Region 142432, Russia

6 GALLY International Biomedical Research Consulting LLC, San Antonio, TX 78229, USA

7 School of Health Science and Healthcare Administration, University of Atlanta, Johns Creek, GA 30097, USA

8 Instituto de Ciencias Biomédicas, Universidad Autónoma de Chile, Santiago, Chile

DOI 10.1007/s12035-017-0771-4

(2)

performance with severe and permanent consequences [18–20]. In this regard, TBI is considered as the leading cause of mortality and disability in children older than 1 year [21, 22] and in older adults in developed countries [17, 23–25].

TBI is more prevalent than some major diseases like Parkinson’s disease, multiple sclerosis, and breast cancer [18]. Even after such a high level of incidence of TBI, there is still no effective treatments available to prevent neuronal damage. This occurs, in part, because of the high diversity of cells involved in the lesion and the lack of knowledge about their functional changes during injury [26]. For example, the overall effects of TBI-reactive astrocytes are still uncertain [3], and it is unclear as to what extent these cells are neuroprotective or deleterious for the brain [27,28].

Understanding the response of astrocytes and other types of cells to TBI can shed light on the mechanisms leading to functional deterioration of the nervous tissue, as well as to identify possible treatments or prevention targeted to avoid progression of this pathophysiology [18]. Recently, conditioned medium of human mesenchymal stem cells from adipose tissue (CM-hMSCA) has been used to treat pathologies of the central nervous system (CNS). Recent studies confirmed that the CM-hMSCA may release factors with anti-apoptotic, anti-inflammatory functions including hor- mones, extracellular matrix proteins, and neurotrophic factors [29,30]. Recent studies provided evidence on the protective effects of CM-hMSCA by enhancing the recovery of neuronal networks [31]. For example, CM-hMSCA reduced brain damage in rat models of stroke by increasing the proliferation of endothelial cells, reducing neuronal apoptosis and astrogliosis [32–34]. In this study, we explored the effect of CM-hMSCA in protecting human astrocytes fromscratchinjury.

Materials and Methods Primary Culture of hMSCA

Human mesenchymal stem cells from adipose tissue (hMSCA) were isolated according to a previously described method [35]. The procedures were performed according to a protocol approved by the ethics committee of the Pontificia Universidad Javeriana. Briefly, hMSCA were isolated from human adipose tissue that was obtained by liposuction in pa- tients between 24 and 28 years of age. hMSCA were characterized by assessing the expression of CD34(−), CD73(+), CD90(+), and CD105(+) following the criteria established by the International Society for Cell Therapy. The cells were cultivated in Dulbecco’s modified Eagle’s medium (DMEM) (LONZA, Walkersville, USA) supplemented with 10% fetal bovine serum (FBS) (LONZA, Walkersville, USA) and 1%

penicillin, at 37 °C in an atmosphere of 5% CO2.

Preparation of CM-hMSCA

hMSCA were cultured until 80% confluency with DMEM supplemented with 10% FBS. Thereafter, the cells were then cultured in serum-free DMEM and the CM-hMSCA was collected after 48 h. The supernatants were centrifuged at 3000 rpm for 3 min and stored at−80 °C. CM-hMSCA was collected from hMSCA cultures between passages III and V.

Human Astrocyte Cell Culture

We used GIBCO® Human Astrocytes (N7805-100, Life Technologies,Warsaw, Poland), a cell line previously used and validated [36–38]. GIBCO® Human Astrocytes are derived from normal brain tissue and stain positive for the astrocyte marker glial fibrillary acid protein (GFAP) [37,38]. The cells were cultured at 37 °C in GIBCO Astrocyte Medium [A1261301, Dulbecco’s modified Eagle’s medium (DMEM), N-2 Supplement, fetal bovine serum (FBS), a n d 1 0 U p e n i c i l l i n / 1 0 μg s t r e p t o m y c i n / 2 5 n g amphotericin (LONZA)] according to the manufacturer’s instructions. Cell plates were Geltrex matrix-coated (A14132) and were seeded 4 × 10⁴ cells/cm²in 96-well plates. The cultures were incubated at 37 °C in a humidi- fied atmosphere containing 5% CO2, and the culture medium was changed three times a week.

ScratchAssay and CM-hMSCA Treatments

In this study, our in vitro model (scratchassay) was characterized by mechanical injury as well as metabolic insult, and has been previously used to mimic an in vitro model of TBI [39–41]. Briefly, the cells were allowed to reach confluence for 72 h and then serum deprived for 6 h. Later, a denuded area was produced by scratching the inside diameter of the well with a 10-μl pipette tip [40–42]. Then, thescratchcells were immediately rinsed with phosphate-buffered saline (PBS) 1×

to remove debris and treated with different concentrations of CM-hMSCA (5–10–15% (v/v)). The experimental groups were as follows: (1)scratch± BSS0, cells underscratchand glucose-free conditions (BSS0); (2)scratch± BSS5,scratch cells plus BSS0 supplemented with 5.5 mM glucose (BSS5, control cells); (3) scratch± BSS0 ± CM,scratchcells plus BSS0 and treated with different concentrations of CM- hMSCA (5, 10, 15%); and (4)scratch ±BSS5 ± CM 5–10–

15%,scratchcells plus BSS5 and treated with different concentrations of CM-hMSCA. Glucose-free condition assay was performed as reported previously [43–45]. The composition of the balanced salt solution (BSS0) was NaCl, 116; CaCl2, 1.8; MgSO4 7H2O, 0.8; KCl, 5.4; NaH2PO4, 1; NaHCO3, 14.7; and 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), 10; pH 7.4. In this study, cells were submitted to scratch + BSS0 (mechanical injury plus glucose-free

(3)

conditions) and simultaneously received a co-treatment with different concentrations of CM-hMSCA for 30 h at 37 °C.

Determination of Viability and Nuclear Fragmentation

We determined viability through MTT assay (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) (Sigma, St Louis, MO, USA). Cells were seeded into 96-well plates in GIBCO Astrocyte Medium (A1261301, Dulbecco’s modified Eagle’s medium (DMEM), N-2 Supplement, fetal bovine serum (FBS), and 10 U penicillin/10μg streptomycin/25 ng amphotericin (LONZA)) at a seeding density of 4 × 10⁴ cells/cm², incubated for 72 h until they reached confluence, and treated. Cell viability was assessed and standardized at 24, 30, and 48 h, following injury by adding MTT solution at the final concentration of 2 mg/ml, for 3 h at 37 °C. Then, DMSO was added to the cells and the blue formazan production was evaluated in a plate reader at 595 nm. The MTT values were normalized with the experimental control group (scratch+ BSS5), which was considered as 100% survival. The experimental groups taken into account for the determination of viability, as well as for the other determinations, were presented inBScratch Assay and CM-hMSCA Treatments.^ Each viability assay was performed with a minimum of six repli- cates per condition.

Nuclear fragmentation was determined by Hoechst 33258 staining. Briefly, after treatments, the cells were washed with PBS and fixed for 20 min in 4% formaldehyde at room tem- perature (RT). Subsequently, the cells were washed and la- beled with Hoechst 33258 (5 mg/ml; Invitrogen) for 15 min.

Cell nuclei were observed and photographed using an inverted fluorescence microscope, Olympus IX-53 (UV excitation, fil- ter CKX-NU N1157600) (excitation 352 nm/emission 461 nm) with an exposure time set between 80 and 100 ms to avoid the saturation of the pixels. The number of fragmented nuclei was determined in at least eight randomly selected areas (0.03 mm²) from each experimental group. Data were expressed as a percentage of nuclear fragmented relative to the value in control cultures (scratch+ BSS5). The experiments were replicated at least in three different cell cultures.

Morphological Analysis

Digitized ×20 magnified black and white images were taken by a digital camera (Moticam CMOS 10) attached to an inverted microscope (Olympus IX-53). The images were used to evaluate the area, polarity, and percentage of the astrocytic processes in all the treatments. Cell area and polarity were calculated by randomly selecting cells from the images. The polarity index was calculated as the length of the main migration axis (parallel to the direction of movement) divided by the length of the perpendicular axis that intersects the center of the cell [41, 46]. The percentage of astrocytic processes was

calculated as described [47] by counting the processes of the cell body in 20 cells per experimental group. In addition, to complement the morphological study, the binary silhouette of a cell was reduced to its one-pixel contour to show a visual representation of the cell dimensions and complexity in the main experimental groups [48]. The cell area, polarity, and astrocytic processes of wound-border cells were calculated to determine the effect ofscratch+ BSS0 and CM-hMSCA on cell morphology relative to control cultures (scratch + BSS5). Data were obtained with a minimum of 25 cells for each condition.

Microscopy and Migration Analysis

For the wound closure assay, confluent astrocyte monolayers cultured on plastic plates were used. Before CM-hMSCA co- treatment, thescratchwas performed on the monolayer using a 10-μl pipette tip and washing once with PBS-1X to remove detached cells and debris. Immediately afterwards, treatment was started with the different experimental groups proposed in BScratchAssay and CM-hMSCA Treatments.^At the begin- ning of the experiment with 0 h of treatment (basal time), digital images were taken from scratch cells using a ×10 mag- nification lens. Subsequently, after 30 h (end time of the trial), the same region was again visualized and the photographic record was made.

To complete the evaluation of the astrocytes to migrate, in the images at base time and 30 h after co-treatments, the distance occupied by the cells from the edge of the wound was measured as indicated by the asterisk in Fig.4b. The measure- ment was performed on both sides of the wound, and the value was averaged and the analysis is presented in percentage of migration. The images were taken in black and white by a digital camera (Moticam CMOS 10) attached to an inverted microscope (Olympus IX-53). The images were then analyzed with NIH ImageJ software 1.46r (National Institutes of Health, downloadhttp://rsbweb.nih.gov/ij/) to quantify the percentage of wound closure by quantifying the remaining free area of cells in basal time (t= 0 h) and comparing it with that after 30 h. For the determination of the percentage of cell migration, the average of the distance occupied by the cells from the border of the wound in base time and then with 30 h of co-treatment was compared. Experiments were performed at least three times and measurements were performed in triplicate.

Determination of Mitochondrial Membrane Potential

Mitochondrial membrane potential was evaluated by using tetramethylrhodamine methyl ester (TMRM) and assessed by fluorescence image analysis. Cells were seeded at a density of 4 × 10⁴cells/cm²into 48-well plates in GIBCO Astrocyte Medium (previously described), and subjected to each experimental paradigm after 72 h. After treatments, the cells were

(4)

incubated with TMRM for 20 min. Finally, the cells were washed with PBS and photographed in a fluorescence microscope (Olympus IX-53). The images were processed with ImageJ software, and the mean fluorescence intensity of randomly selected cells was determined as described below. The mean was calculated with a minimum of 20 cells analyzed for each condition.

The experiments were repeated in three different cultures.

Estimation of Cellular Mean Fluorescence Intensity

The calculation of the mean fluorescence intensity of the cells for the determination of mitochondrial membrane potential and nuclear fragmentation was performed using ImageJ [49].

The microphotographs were loaded in the software and pre- processed eliminating the background. Subsequently, 20 cells were randomly selected using a numbered grid in each micro- photograph. The mean fluorescence value of the 20 cells was determined in six microphotographs for each treatment using the Measure algorithm of ImageJ and selecting each cell man- ually via ROI’s Management. The control cultures in the fluorescence determinations were the cells under the conditions of scratch+ BSS5. The cells were analyzed in an area of 0.03 mm². There were no variations in the conditions of the image processing. Each assay was performed with a minimum of six replicate wells for each condition.

Protein Extraction and Western Blotting

GIBCO® Human Astrocytes were lysed on ice with RIPA Lysis and Extraction Buffer Thermo Scientific™supplemented with Halt™Protease Inhibitor Cocktail, EDTA-free (100×) (Roche). Protein content was estimated using the Pierce™

BCA Protein Assay Kit. Equal amounts of protein were dissolved in sample buffer and separated by electrophoresis in SDS-PAGE, transferred onto PVDF membranes, and blocked in 5% skim milk dissolved in Tris-buffered saline containing 0.05% Tween 20 (TBS-T), at RT for 1 h. The membranes were incubated at 4 °C ON with antibodies againstβ-actin (Thermo Fisher) (1:3000), superoxide dismutase 2 (SOD2) (Thermo Fisher) (1:1000), and GPX1 (Thermo Fisher) (1:1500). The immunoreactivity was visualized by incubating the membrane with specific secondary antibody (IRDye® Antibodies) for 1 h and detected using Odyssey CLx Imaging System (LI- COR Biosciences). The intensity of each band was quantified using Image Studio Software 5.x. All data were normalized to control values (scratch+ BSS5) on each gel. All experiments were performed in triplicates.

Statistical Analysis

Data obtained from this study were tested for normal distribution by Kolmogorov–Smirnov test and homogeneity of variance by Levene’s test. The data were then examined by

analysis of variance (ANOVA), followed by Dunnet’s post hoc test for comparisons between controls and treatments.

Tukey’s post hoc test was used for multiple comparisons between the means of treatments and time points. Data were presented as mean ± SEM of three independent experiments.

A statistically significant difference was defined atp< 0.05.

Results

ScratchAssay Reduced Astrocytes’Viability at 30 h

Both brain injury and glucose deprivation are recognized as one of the causes of increased cognitive impairment and damage in many brain pathologies [18,50,51]. We did not find any damage or alteration in viability or at the mitochondrial level induced only byscratch(data not shown); therefore, we associated this mechanical damage (scratch) with a metabolic deterioration (glucose deprivation), as the experimental group scratch + BSS0. In this regard, after 6 h without FBS, we subject the cells toscratch+ BSS0 during increasing periods of times: 24 h (Fig.1d), 30 h (Fig.1e), and 48 h (Fig.1f). Cells cultured inscratch+ BSS5 solution (5.5 mM glucose) were used as controls. The results show that after 30 h ofscratch+ BSS0 treatment, cells reached a 50% decrease in cell viability (Fig. 1e–g) when compared to control cultures (scratch + BSS5) incubated for the same period of time (p < 0.0001;

Fig. 1c–g). We observed that 30 h of scratch+ BSS0 was the optimum time to be selected for the next experiments.

Cellular morphological changes were evident over time in cells subjected toscratch+ BSS0 (Fig.1d, e), reaching up to 40% viability at 48 h (Fig.1f).

Dose–Response of the CM-hMSCA Against Cell Death Induced byScratchin Human Astrocytes

To determine the effect of CM-hMSCA on cell survival, astrocytes were treated with increasing concentrations of CM- hMSCA (5–10–15%) and subjected toscratch + BSS0 for 30 h (Fig. 1d). Figure2 shows the viability results obtained during co-treatment with scratch+ BSS0 + CM at 5, 10, or 15%. Our results evidenced a recovery of 91.48% (p< 0.0001), 97.4% (p< 0.0001), and 103.7% (p< 0.0001) in cell viability during co-treatment withscratch+ BSS0 + CM5% (Fig.2a–d), scratch+ BSS0 + CM10% (Fig.2b–e), andscratch+ BSS0 + CM15% (Fig.2c–f), respectively, compared toscratch+ BSS0 (Fig. 2g). This confirms that low concentrations of CM- hMSCA in the range of 2–5% were able to recover astrocytes and attenuated cell death in this model of cell injury [39,41].

Furthermore, our findings suggest that CM-hMSCA provides cell protection in a concentration-dependent manner (Fig.2g).

However, the differences between CM treatment effects did not reach statistical significance (p= 0.8678) (Fig.2g).

(5)

CM-hMSCA Induces Changes in Human Astrocyte Morphology UnderScratchAssay

Changes in surface and cellular complexity over time may re- flect functional differences in reactive astrocytes [48,52].

Figure3shows changes in cell morphology for different experimental groups:scratch+ BSS5 (Fig.3a),scratch+ BSS0 (Fig.

3b), andscratch+ BSS0 + CM15% (Fig.3c). Qualitatively, we

observed changes in cellular morphology under scratch + BSS0 conditions (Fig.3b). In addition, in the visualization of the binary silhouette of a representative cell per experimental group, compared to the control cells (scratch+ BSS5) (Fig.3d), we found evident morphological changes and an increase in cells arbor under conditions of scratch + BSS0 (Fig. 3e).

Indeed, we determined the percentage of astrocytic processes (Fig.3g) and the cell area (Fig.3h) to evaluate our observations quantitatively. Figure3g shows the differences in the number of astrocytic processes per experimental group, with an increase of up to 133.7% in astrocytes underscratch+ BSS0 conditions compared toscratch+ BSS5 (p = 0.0013). All experimental groups,scratch+ BSS0 + CM5% (p< 0.0001),scratch+ BSS0 + CM10% (p < 0.0001), andscratch+ BSS0 + CM15% of CM-hMSCA (p = 0.0004) presented with time a significant decrease in the number of astrocytic processes and cellular complexity with relation to baseline. Furthermore, after 30 h of scratch, treatment with CM-hMSCA induced significant decreases in astrocytic processes in the groups when compared to controls (scratch+ BSS0): 82% (p < 0.0001) forscratch + BSS0 + CM5%, 82.19% (p< 0.0001) forscratch+ BSS0 + CM10%, and 75.57% (p < 0.0001) for scratch+ BSS0 + CM15%-CM-hMSCA. However, in the analysis of astrocytic processes, no significant differences were found between the concentrations of CM-hMSCA assayed (p> 0.9999). In con- trast, cells that were treated with CM-hMSCA at different concentrations maintained the mean cell area and showed significant differences with thescratch+ BSS0 group (Fig.3h). The area was higher at increasing CM concentration as follows:

562.4 μm² at 5% (p < 0.0001), 725.7 μm² at 10%

(p< 0.0001), and 822.2μm²at 15% (p< 0.0001). In this time, the differences between the concentrations ofscratch+ BSS0 + CM5% andscratch+ BSS0 + CM15% (p= 0.0057) of CM- hMSCA were significant (p< 0.05).

CM-hMSCA Induces Wound Closure and Cell Migration in Human Astrocytes Subjected toScratch

Wound closure and cell migration were evaluated under control conditions (scratch+ BSS5;scratch+ BSS0) (Fig.4b) and in the groups treated with scratch + BSS0 + CM5%

(Fig.4c),scratch+ BSS0 + CM10% (Fig.4d), andscratch + BSS0 + CM15% (Fig.4e). The results show that treatment withscratch+ BSS0 + CM5 increased wound closure in a direct proportion to the concentration of CM used. After 30 h of incubation, an increase in wound closure of 22.45%

(p < 0.0001; Fig. 4c), 31.92% (p < 0.0001; Fig.4d), and 35.48% (p< 0.0001, Fig.4e) was observed when cells were treated withscratch+ BSS0 + CM5%, scratch+ BSS0 + CM10%, andscratch+ BSS0 + CM15%, respectively.

Figure 4f shows the percentage of cell migration, which correlates with wound closure. Cells treated with scratch+ BSS0 + CM showed a higher percentage of cell migration than Fig. 1 Effects ofscratch assayon human astrocytes. The figure shows

representative images of astrocytes treated withscratch+ BSS5 at different times 24 h (a), 30 h (b), and 48 h (c).d–fChanges in the growth and cellular morphology of astrocytes at different times after scratch+ BSS0 were observed. The graph of lines (g) show cellular viability at 24, 30, and 48 h after 6 h without FBS. Number sign, p< 0.0001 betweenscratch + BSS0 from 24 and 30 h. Asterisk, p< 0.0001, betweenscratch+ BSS5 andscratch+ BSS0 at 30 h. ns (p= 0.9949) in the percentage of cell viability between the groups with and withoutscratchassay. All data in these figures are presented as mean ± SEM of three individual experiments. Error bars indicate SEM.

Scale bar, 50μm

(6)

controls, and this was dependent upon the concentration of CM-hMSCA. After 30 h, cell migration was 4.3 times higher in cells that were treated withscratch+ BSS0 + CM15% (Fig.

4e, f) than control cells treated with scratch + BSS0 (p < 0.0001) (Fig.4b–f). To confirm whether the increase in wound closure was induced by cell migration, we evaluated the index of polarity of the cells in each of the experimental groups as the index of polarity has been reported to indicate the state of migration of the cells [46]. In this context, it was found that the polarity index of the cells subjected toscratch+ BSS0 was 1.7 times lower than that of cells subjected toscratch+ BSS5 cells (p< 0.0001; Fig.5a–d). However, after 30 h of treatments, CM- treated cells showed in comparison to control cells (scratch+ BSS5) a 2.3 (p< 0.0001), 2.9 (p= 0.0002), and 4.7 (p< 0.0001) times increase in the polarity index in the cells that were treated withscratch+ BSS0 + CM5%,scratch+ BSS0 + CM10%, and scratch+ BSS0 + CM15%, respectively (Fig.5d).

CM-hMSCA Decreases Nuclear Fragmentation and Improves Mitochondrial Membrane Potential in Human Astrocytes

To further evaluate the protective effects of CM in astrocytes subjected toscratchassay, the effect of CM-hMSCA on nuclear fragmentation was assessed. Figure 6 shows

control cells (scratch + BSS5) (Fig. 6a) and astrocytes subjected to scratch + BSS0 with an increased nuclear fragmentation by 35.4% compared to control (p< 0.0001;

Fig. 6b). However, cells that were treated with scratch+ BSS0 + CM showed a marked reduction in nuclear fragmentation by 41.54 and 48.59% in the presence ofscratch + BSS0 + CM5% (p< 0.0001; Fig.6c) andscratch+ BSS0 + CM10% (p< 0.0001; Fig.6d), respectively. Surprisingly, a reduction of up to 53.55% with scratch + BSS0 + CM15% (p < 0.0001; Fig. 6e) with respect to scratch + BSS0 cells was observed (Fig.6f).

Similar results were observed when the mitochondrial membrane potential (Δψm) was determined. Figure 7 shows differences inΔψm in astrocytes exposed toscratch + BSS5 (Fig.7a),scratch+ BSS0 (Fig.7b), and astrocytes co-treated with scratch+ BSS0 + CM5% (Fig. 7c), 10%

(Fig.7d), and 15% (Fig.7e). Cells subjected toscratch+ BSS0 + CM maintained aΔψm close to that of the control (scratch + BSS5) when evaluated at different concentrations. Also, an increase of 2.31 times (p = 0.0013; Fig.

7e, f) protection of the Δψm in the group of cells with scratch + BSS0 + CM15% compared to the control (scratch + BSS5). Likewise, we found significant differences between thescratch+ BSS0 + CM10% andscratch + BSS0 + CM15%Δψm values (p< 0.0001) (Fig.7f).

Fig. 2 Dose-response of CM-hMSCA in human astrocytes withscratch + BSS0 at 30 h.scratch+ BSS0 + CM-MSCA at 5 (a–d), 10 (b–e), and 15% (c–f) increase cellular viability of astrocytes near to control conditions.hThis figure shows the recovery of cell viability by different concentrations of CM-hMSCA. MTT values were normalized withscratch+ BSS5 control cells.gComparison between the different

experimental groups and the response in cell viability to different concentrations of CM-hMSCA. Four asterisks,p< 0.0001 between scratch+ BSS0 and the different concentrationsscratch+ BSS0 + CM- hMSCA 5–10–15%. ns (p= 0,8678) in the viability at 5–10–15% into groupscratch+ BSS0 + CM-hMSCA. Error bars indicate SEM. Scale bar, 50μm

(7)

CM-hMSCA Expression of Antioxidant Proteins in Human Astrocytes uponScratch

Since several parameters evaluated in this work showed a better recovery of astrocytes treated withscratch+ BSS0 + CM15%, the effect of CM-hMSC on the endogenous antioxidant defense was tested (Fig.8). In this regard, it was observed that 30 h of co-treatment withscratch + BSS0 + CM15% resulted in an upregulation of antioxidant enzymes.

We observed 1.8- and 0.4-fold increases in glutathione peroxidase expression (GPX1;p< 0.0001, Fig.8a, b), and superoxide dismutase 2 (SOD2;p< 0.0001, Fig.8a–c) in astrocytes

treated withscratch+ BSS0 + CM15% when compared to cells treated withscratch+ BSS0 alone, respectively.

Discussion

In this study, we aimed to explore the effect of CM-hMSCA on human astrocytes subjected toscratchassay. Our findings indicated that CM-hMSCA improved cell viability, reduced nuclear fragmentation, and preserved mitochondrial membrane potential. These results were accompanied by morphological changes and increased the polarity index, thus Fig. 3 CM-hMSCA induces changes in the cellular morphology of

human astrocytes withscratch + BSS0 at 30 h. The figure shows representative micrographs of the morphology of astrocytes treated with scratch+ BSS5 (a),scratch+ BSS0 (b), andscratch+ BSS0+ CM- hMSCA 15% (c).eArboreal morphology of astrocytes inscratch+ BSS0 conditions in binary silhouette andfmorphological changes in astrocytes treated withscratch+ BSS0 + CM-hMSCA15% in binary silhouette.gPercentage of astrocytic processes in different groups. Two asterisks,p =0.0013scratch+ BSS5 vs.scratch+ BSS0 at 30 h; number

sign,p< 0.0001scratch+ BSS0 + CM-hMSCA 5–10–15% vs.scratch+ BSS0 at 30 h; ampersand,p< 0.0001scratch+ BSS0 + CM-hMSCA 5– 10–15% vs.scratch+ BSS0 in base time.hshows the mean cell area (μm²). Four asterisks,p< 0.0001scratch+ BSS5 vs.scratch+ BSS0 at 30 h; a,p =0.0057scratch+ BSS0 + hMSCA-CM 5% vs.scratch+ BSS0 + CM-hMSCA 15% at 30 h. No significant differences between CM-hMSCA concentrations for astrocytic processes. Error bars indicate SEM. Scale bar, 50μm

(8)

reflecting the ability of cells treated with CM-hMSCA to migrate towards the wound suggesting that CM-hMSCA may be considered as a promising therapeutic strategy for the protection of astrocytes in brain pathologies.

TBI is considered a pathology with dynamic and complex processes that involve multiple cells and trigger various cell

signaling pathways that lead to neurodegeneration [18]. The complexity of this pathology has made it difficult to understand and prevent the progress of the neuronal damage.

Recently, in order to understand the pathophysiology of TBI, several studies have been carried out using different in vitro models proposed as complementary tools for discovery in the Fig. 4 Effect of CM-hMSCA on cell migration and wound closure.a,b

Microphotographs of cell migration (p= 0.0452) treated withscratch+ BSS5 andscratch+ BSS0 at 30 h, respectively.c–eMicrophotographs of cells treated withscratch+ BSS0 + CM-hMSCA at 5, 10, and 15%, respectively.fScratch+ BSS0 + CM-hMSCA increases cell migration vs. base time in the concentration of 5 (a,p= 0.0180), 10 (b,p= 0.0180), and 15% (c,p< 0.0001). At 30 h, we found significant differences between thescratch+ BSS0 vs.scratch+ BSS0 + CM-hMSCA 10%

(two asterisks,p= 0.0031) and 15% (four asterisks,p< 0.0001). Blue arrows indicate non-migrating cells and black arrows migrating cells with

altered cellular morphology. One asterisk indicates the distance covered by the cells from the edge of the wound used to calculate the percentage of cell migration. On the other hand,gscratch+ BSS0 + CM-hMSCA increases wound closure at any concentration 5% (a,p< 0.0001), 10% (b, p< 0.0001), and 15% (c,p< 0.0001) with respect toscratch+ BSS0 at 30 h. Four asterisks,p< 0.0001scratch+ BSS5 vs.scratch+ BSS0 at 30 h. ns (p> 0.9999) without significant differences betweenscratch+ BSS0 base time and 30 h. All data in these figures are presented as mean ± SEM of three individual experiments. Error bars indicate SEM. Scale bar, 50μm

Fig. 5 CM-hMSCA increases the polarity index in human astrocytes.

Scratch+ BSS0 + CM-hMSCA increases polarity index in wound border cells.aMicrophotograph of control cells (scratch+ BSS5).

Yellow lines indicate the main migration axis (parallel to the direction of movement) and intersection the center of the cell used to determine the polarity index of the cells.b,cMicrophotographs ofscratch+ BSS0 and

scratch + BSS0 + CM-hMSCA 15%, respectively. Number sign, p< 0.0001scratch+ BSS5 vs.scratch+ BSS0 de 30 h. Four asterisks, p< 0.0001scratch+ BSS0 + CM-hMSCA 5, 10, and 15% vs.scratch+ BSS0 at 30 h. One asterisk,p= 0.0294scratch+ BSS0 + CM-hMSCA 5% vs.scratch+ BSS0 + CM-hMSCA 15%. Error bars indicate SEM.

Scale bar, 50μm

(9)

field [26,53–55]. One of these models is thescratchassay that has been widely used to assess the effects of brain injury in a controlled in vitro environment [26,56,57]. In the present

work, we used ascratchassay model associated with a metabolic dysfunction (glucose deprivation) to address the ability of astrocytes to migrate and whether glucose deprivation in Fig. 6 CM-hMSCA reduces nuclear fragmentation followingscratch+

BSS0. Upper panel shows representative micrographs of astrocytes exposed toscratch+ BSS5 (a);scratch+ BSS0 (b);scratch+ BSS0 + CM-hMSCA 5% (c);scratch+ BSS0 + CM-hMSCA 10% (d), and scratch+ BSS0 + CM-hMSCA 15% (e). The bar graph shows the

percentage of cells with fragmented nuclei (f). Four asterisks, p< 0.0001scratch+ BSS0 + CM-hMSCA 5, 10, and 15% vs.scratch + BSS0 at 30 h. Three asterisks,scratch+ BSS0 + CM-hMSCA 10% vs.

scratch+ BSS0 + CM-hMSCA 15% at 30 h. Error bars indicate SEM.

Scale bar, 50μm

Fig. 7 CM-hMSCA preserves the mitochondrial membrane potential following scratch + BSS0. Upper panel shows representative photomicrographs of astrocytes exposed toscratch+ BSS5 (a);scratch + BSS0 (b);scratch+ BSS0 + CM-hMSCA 5% (c);scratch+ BSS0 +

CM-hMSCA 10% (d), andscratch+ BSS0 + CM-hMSCA 15% (e). The bar graph shows the values of TMRM fluorescence (f). Four asterisks, p< 0.0001scratch+ BSS0 + CM-hMSCA 5, 10, and 15% vs.scratch+ BSS0 at 30 h. Error bars indicate SEM. Scale bar, 50μm

(10)

astrocytes might impair mitochondrial functionality and regeneration processes. The mechanical damage suffered during TBI and glucose withdrawal due to the interruption of blood flow lead to metabolic alterations and a direct relation with cell death pathways [58,59]. Besides, previous works have shown that glucose withdrawal may affect wound closure and contribute to an increase of the secondary lesions in various cerebral pathologies [41,60,61]. In this context, various compounds such as estrogens [43] and the use of MSCs [62,63] or their conditioned medium [64–66] are being evaluated to treat brain tissue lesions or to decrease glial reactivity. Some studies suggest that the use of conditioned medium or extracellular vesicles derived from MSCs may help to counteract neuronal injury upon a neurodegenerative event [67], decrease microgliosis induced by inflammation in conditions such as amyotrophic lateral sclerosis (ALS) [68], and prevent reactive astrogliosis after ischemic stroke [68–70].

Astrocytes become reactive upon brain injury or metabolic insult, which is characterized by increased expression of inter- mediate filament proteins such as GFAP and vimentin, and accompanied by prominent morphological changes [48,71, 72] (Fig. 3b–e). Astrocytes respond to lesions by changing their morphology and showing hypertrophy of processes [48,73]; [2]. These changes can last for days or even months or become generalized in the brain tissue [74]; [75]. In the present study, we evaluated the activation or reactivity of astrocyte changes in morphology that have been associated to brain pathologies [48,73]; [2]; [75,76]). In our model, differences in morphology (Fig. 3b) and cell area (Fig. 3h) were observed. Morphological alterations were evident in astrocytes subjected toscratch+ BSS0 (Fig.3b–e), with a significant recovery in cells treated withscratch+ BSS0 + CM 5– 10–15% (Fig.3c–f), positively correlating this with recovery in cell viability [77] and reported in previous studies in our laboratory [41, 78]. Experimental studies have reported changes in the morphology of astrocytes similar to those observed in our model [79–82] and which are associated with glial activation. Although the process of structural remodeling

of astrocytes during glial activation is still unclear, some molecules or pathways that may be mediating these changes have been reported. For example, fibroblast growth factor (FGF) through theβ-adrenergic receptors (β-AR) [83], cAMP signaling [84–86], the heparin-binding epidermal growth factor- like growth factor (HB-EGF), which induces EGF receptor phosphorylation in Y1068, Mapk/Erk pathway activation, and the JAK-STAT3 signaling pathway altogether are key factors in astrocyte activation. Canonical signaling in astrocyte reactivity including overexpression of GFAP [87] [88, 89] and a simple subcellular distribution in most cytoskeletal proteins are factors that can participate in the morphological and physiological changes of astrocytes.

It is well known that astrocytes perform important functions for the maintenance of tissue structure not only in nervous developmental stages but also during brain adulthood [90]. According to Faber-Elman et al. [91], one of the prob- lems in axonal regeneration is the lack of astrocytes to repop- ulate the site of injury in a balanced environment [91–94]. It is widely described that the amniotic fluid-derived conditioned medium and adipose-derived stem cells significantly enhance wound healing [95] and cell migration [96–98]. Firstly, we aimed to investigate the percentage of cell migration and wound closure (Fig.4f, g) and the polarity index (Fig.5d) in scratch astrocytes subjected to conditioned medium. Our evi- dences showed a significant increase in both parameters in astrocytes that were exposed toscratch + BSS0 + CM 5–

10–15% compared toscratch + BSS0. The increase in the migration of cells treated withscratch+ BSS0 + CM 5–10–

15% suggests an effect probably related to the production and secretion of molecules such as transforming growth factorβ1 (TGF-β1) [95], basic fibroblast growth factor (bFGF) [99, 100], epidermal growth factor (EGF) [95], laminin [101], fi- bronectin [56,101], and some heparan sulfate proteoglycans (HSPG) [102–104]. These molecules have been identified in the secretome of different MSCs [29,30,63,96,98,104,105]

and have been reported as fundamental factors to maintain the migration of astrocytes and other cells such as fibroblasts [56, Fig. 8 CM-hMSCA slightly induced the expression of antioxidant

proteins in human astrocytes. Western blot analysis indicated that the expression of GPX1 (a, b) and SOD2 (a–c) was 1.8- and 0.4-fold increase inscratch+ BSS0 + CM-hMSCA15%-treated cells compared

to untreated cells underscratch+ BSS0 at 30 h.β-Actin was used as loading control. Number sign,p< 0.0001scratch+ BSS5 vs.scratch+ BSS0 at 30 h. Four asterisks,p< 0.0001scratch+ BSS0 + CM-hMSCA 15% vs.scratch+ BSS0 at 30 h

(11)

91,101,105]. It is possible that the presence of molecules in CM-hMSCA involved in cell migration is regulating both the migration and closure of the wound in our model through the activation or mediation of different pathways such as ERK1/2 [106], Rac1 [107], ERK and JNK signaling pathways [108], and Rho GTPases [109], or perhaps with the direct action of TGF-β1 mediated through the activation of CysLT1R present in astrocytes (X. Q. [110]), and control of cell polarity regu- lated by the Cdc42 integrin [111] or EBI2-mediated cell signaling [112]. However, further research on the role of each signaling cascade is deserved.

Finally, deprived or low-glucose conditions can also induce the release of free fatty acids, impair ionic homeostasis, increase intracellular calcium flow, alter mitochondrial metabo- lism, and increase the production of free radicals [113,114].

These changes subsequently lead to DNA and mitochondrial membrane damage [44,78]. Several studies have reported that the CM-hMSC from different sources including adipose tissue exerts a protective effect both at the neuronal and astrocytic levels, maintaining neural networks, reducing neuronal

apoptosis and astrogliosis, thus providing neuroprotection [64,67]; [65,66,69,70]; [68]. Our findings show a decrease in the number of fragmented nuclei (Fig.6f) suggesting that apoptosis linked to energy depletion caused by the absence of glucose was diminished by the effect of CM-hMSCA. The above finding is probably related to the hypothesis that the CM-hMSCA contributes to maintaining mitochondrial membrane potential (Δψm), which was confirmed through our assays. As observed in Fig.7f,Δψm was recovered in astrocytes subjected toscratch+ BSS0 and treated with CM5–10–

15%. Maintenance ofΔψm suggests that CM-hMSCA may have a direct effect on mitochondrial function, as suggested by our previous studies [39]. This may also be related to the maintenance of ATP levels or the activation of other metabolic pathways important for cell recovery. However, these mechanisms require further investigation in this cell type and model.

The above discussion suggests that CM-MSCA can over- come and neutralize the negative effect ofscratchassay, act- ing in synergy with other proteins or through the activation of mechanisms that may promote cell migration and/or survival.

Fig. 9 Proposed model of the effect of CM-hMSCA on the protection of human astrocytes subjected toscratchassay.Scratchassay in association with glucose-free conditions leads to cell death, morphological changes, alterations in the removal of ROS that can be correlated with loss of Δψm, and DNA fragmentation, thus causing unrepairable cellular damage. This damage induced byscratch+ BSS0 on human astrocytes

was attenuated by CM-hMSCA at 5–10–15% for 30 h by preserving cell viability, reversing morphological changes, increasing polarity index and cell migration, decreasing nuclear fragmentation, and conservingΔψm in addition to favoring cellular antioxidant capacity with the expression of SOD2 and GPX1

(12)

This cellular survival may also be related to the control of ROS by means of antioxidant proteins. In this case, we found a slight increase in the expression of antioxidant proteins like GPX1 (Fig.8a, b) and SOD2 (Fig.8a–c) in cells treated with scratch+ BSS0 + CM15%, as these enzymes are expressed by astrocytes that help scavenge the production of ROS [115–117]. This result confirms our previous studies performed using T98G cells [39]. However, further research is required to fully evaluate the effect of CM-hMSCA on the activity of these enzymes, as well as other proteins of biolog- ical importance such as neuroglobin (Ngb), a protein which was reported as mediating the protective effect of CM-MSCA in the T98G astrocyte model [39]. A summary of the main findings found in the present manuscript is displayed in Fig.9.

In conclusion, our results demonstrate that CM-hMSCA exerts protective effects on human astrocytes exposed to scratch assay, confirming that the adipose-derived mesenchymal stem cells may participate in protective actions of astrocytes and contribute to the maintenance of their neurosupportive functions in different brain pathologies.

Acknowledgements The authors thank Dr. Camilo Prieto and the staff of the cosmetic surgery in Bogota, Colombia, for the adipose tissue sam- ples. This work was supported by PUJ ID 6260 to GEB and scholarship for doctoral studies awarded by the Vicerrectoría Académica of PUJ to Baez-Jurado E.

Abbreviations BSS0, balanced salt solution;bFGF, basic fibroblast growth factor;CNS, central nervous system;CMhMSCA, conditioned medium of human adipose mesenchymal stem cells;DMSO, dimethyl sulfoxide;DMEM, Dulbecco’s modified Eagle’s medium;EGF, epidermal growth factor;FBS, fetal bovine serum;GFAP, glial fibrillary acid protein;GPX1, glutathione peroxidase;HB-EGF, heparin-binding epidermal growth factor-like growth factor;HSPG, heparan sulfate proteoglycans;hMSCA, human mesenchymal stem cells from adipose tissue;

MSCs, mesenchymal stem cells; mesenchymal stem cells;MTT, 3-(4,5- dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; Ngb, neuroglobin;PBS, phosphate-buffered saline;ROS, reactive oxygen species;SOD2, superoxide dismutase 2; TMRM, tetramethylrhodamine methyl ester;TGFbeta 1, transforming growth factor beta 1;TBI, traumatic brain injury

References

1. Barreto G, White RE, Ouyang Y, Xu L, Giffard RG (2011) Astrocytes: targets for neuroprotection in stroke. [Research Support, N.I.H., Extramural Review]. Cent Nerv Syst Agents Med Chem 11(2):164–173

2. Sun D, Jakobs TC (2012) Structural remodeling of astrocytes in the injured CNS. [Research Support, N.I.H., Extramural Research Support, Non-U.S. Gov'tReview]. Neuroscientist 18(6):567–588.

https://doi.org/10.1177/1073858411423441

3. Myer DJ, Gurkoff GG, Lee SM, Hovda DA, Sofroniew MV (2006) Essential protective roles of reactive astrocytes in traumatic brain injury. [Comparative Study Research Support, N.I.H., Extramural Research Support, Non-U.S. Gov't]. Brain 129(Pt 10):2761–2772.https://doi.org/10.1093/brain/awl165

4. Panickar KS, Norenberg MD (2005) Astrocytes in cerebral ischemic injury: morphological and general considerations. Glia 50(4):

287–298.https://doi.org/10.1002/glia.20181

5. Steele ML, Robinson SR (2012) Reactive astrocytes give neurons less support: implications for Alzheimer's disease. [Research Support, Non-U.S. Gov't]. Neurobiol Aging 33(2):423 e421- 413.https://doi.org/10.1016/j.neurobiolaging.2010.09.018 6. Hamby ME, Sofroniew MV (2010) Reactive astrocytes as thera-

peutic targets for CNS disorders. [Research Support, N.I.H., Extramural Research Support, Non-U.S. Gov't Review].

Neurotherapeutics 7(4):494–506.https://doi.org/10.1016/j.nurt.

2010.07.003

7. Pekny M, Nilsson M (2005) Astrocyte activation and reactive gliosis. [Research Support, Non-U.S. Gov't Review]. Glia 50(4):

427–434.https://doi.org/10.1002/glia.20207

8. Sofroniew MV (2009) Molecular dissection of reactive astrogliosis and glial scar formation. [Research Support, N.I.H., Extramural Research Support, Non-U.S. Gov't Review]. Trends Neurosci 32(12):638–647.https://doi.org/10.1016/j.tins.2009.08.

002

9. Pekny M, Pekna M, Messing A, Steinhauser C, Lee JM, Parpura V et al (2016) Astrocytes: a central element in neurological diseases. [Research Support, N.I.H., Extramural Research Support, Non-U.S. Gov't Review]. Acta Neuropathol 131(3):323–345.

https://doi.org/10.1007/s00401-015-1513-1

10. Bardehle S, Kruger M, Buggenthin F, Schwausch J, Ninkovic J, Clevers H et al (2013) Live imaging of astrocyte responses to acute injury reveals selective juxtavascular proliferation.

[Research Support, Non-U.S. Gov't]. Nat Neurosci 16(5):580– 586.https://doi.org/10.1038/nn.3371

11. Pekny M, Pekna M (2016) Reactive gliosis in the pathogenesis of CNS diseases. Biochim Biophys Acta 1862(3):483–491.https://

doi.org/10.1016/j.bbadis.2015.11.014

12. Sofroniew MV (2005) Reactive astrocytes in neural repair and protection. [Research Support, N.I.H., Extramural Research Support, Non-U.S. Gov't Research Support, U.S. Gov't, P.H.S.

Review]. Neuroscientist (11, 5):400–407.https://doi.org/10.

1177/1073858405278321

13. Su J, Wang L (2012) Research advances in neonatal hypoglycemic brain injury. [Review]. Transl Pediatr 1(2):108–115.https://doi.

org/10.3978/j.issn.2224-4336.2012.04.06

14. Qiu C, Sigurdsson S, Zhang Q, Jonsdottir MK, Kjartansson O, Eiriksdottir G et al (2014) Diabetes, markers of brain pathology and cognitive function: the age, gene/environment susceptibility- Reykjavik study. [Research Support, N.I.H., Extramural Research Support, N.I.H., Intramural Research Support, Non-U.S. Gov't].

Ann Neurol 75(1):138–146.https://doi.org/10.1002/ana.24063 15. Langlois JA, Rutland-Brown W, Wald MM (2006) The epidemi-

ology and impact of traumatic brain injury: a brief overview. J Head Trauma Rehabil 21(5):375–378

16. Mathers CD, Loncar D (2006) Projections of global mortality and burden of disease from 2002 to 2030. [Research Support, Non- U.S. Gov't]. PLoS Med 3(11):e442.https://doi.org/10.1371/

journal.pmed.0030442

17. Tagliaferri F, Compagnone C, Korsic M, Servadei F, Kraus J (2006) A systematic review of brain injury epidemiology in Europe. [Research Support, Non-U.S. Gov't Review]. Acta Neurochir (Wien) 148(3):255–268; discussion 268.https://doi.

org/10.1007/s00701-005-0651-y

18. Prins M, Greco T, Alexander D, Giza CC (2013) The pathophysiology of traumatic brain injury at a glance. [Research Support, N.I.H., Extramural Research Support, Non-U.S. Gov't]. Dis Model Mech 6(6):1307–1315.https://doi.org/10.1242/dmm.

011585

19. Seaquist ER (2015) The impact of diabetes on cerebral structure and function. [Research Support, N.I.H., Extramural Review].

(13)

Psychosom Med 77(6):616–621. https://doi.org/10.1097/PSY.

0000000000000207

20. Stiles MC, Seaquist ER (2010) Cerebral structural and functional changes in type 1 diabetes. [Review]. Minerva Med 101(2):105–114 21. Dennis EL, Faskowitz J, Rashid F, Babikian T, Mink R, Babbitt C et al (2017) Diverging volumetric trajectories following pediatric traumatic brain injury. Neuroimage Clin 15:125–135.https://doi.

org/10.1016/j.nicl.2017.03.014

22. Meadows EA, Owen Yeates K, Rubin KH, Taylor HG, Bigler ED, Dennis M et al (2017) Rejection sensitivity as a moderator of psychosocial outcomes following pediatric traumatic brain injury.

J Int Neuropsychol Soc:1–9. https://doi.org/10.1017/

S1355617717000352

23. Calviello LA, Donnelly J, Zeiler FA, Thelin EP, Smielewski P, Czosnyka M (2017) Cerebral autoregulation monitoring in acute traumatic brain injury: what’s the evidence? Minerva Anestesiol.

10.23736/S0375-9393.17.12043-2

24. Yasen AL, Howell DR, Chou LS, Pazzaglia AM, Christie AD (2017a) Cortical and physical function after mild traumatic brain injury. Med Sci Sports Exerc 49(6):1066–1071.https://doi.org/10.

1249/MSS.0000000000001217

25. Yasen AL, Howell DR, Chou LS, Pazzaglia AM, Christie AD (2017b) Cortical and physical function following mild traumatic brain injury. Med Sci Sports Exerc.https://doi.org/10.1249/MSS.

0000000000001217

26. Janowitz T, Menon DK (2010) Exploring new routes for neuroprotective drug development in traumatic brain injury. [Research Support, Non-U.S. Gov't]. Sci Transl Med 2(27):27rv21.https://

doi.org/10.1126/scitranslmed.3000330

27. Burda JE, Bernstein AM, Sofroniew MV (2016) Astrocyte roles in traumatic brain injury. [Research Support, N.I.H., Extramural Research Support, Non-U.S. gov't Review]. Exp Neurol 275(Pt 3):305–315.https://doi.org/10.1016/j.expneurol.2015.03.020 28. Karve IP, Taylor JM, Crack PJ (2016) The contribution of astro-

cytes and microglia to traumatic brain injury. [Review]. Br J Pharmacol 173(4):692–702.https://doi.org/10.1111/bph.13125 29. Salgado, A. J., Sousa, J. C., Costa, B. M., Pires, A. O., Mateus-

Pinheiro, A., Teixeira, F., . . . Sousa, N. (2015). Mesenchymal stem cells secretome as a modulator of the neurogenic niche: basic insights and therapeutic opportunities. Front Cell Neurosci,9 30. Teixeira FG, Carvalho MM, Sousa N, Salgado AJ (2013)

Mesenchymal stem cells secretome: a new paradigm for central nervous system regeneration? [Research Support, Non-U.S. Gov't Review]. Cell Mol Life Sci 70(20):3871–3882.https://doi.org/10.

1007/s00018-013-1290-8

31. Yamazaki H, Jin Y, Tsuchiya A, Kanno T, Nishizaki T (2015) Adipose-derived stem cell-conditioned medium ameliorates antidepression-related behaviors in the mouse model of Alzheimer's disease. Neurosci Lett 609:53–57.https://doi.org/10.

1016/j.neulet.2015.10.023

32. Cho YJ, Song HS, Bhang S, Lee S, Kang BG, Lee JC et al (2012) Therapeutic effects of human adipose stem cell-conditioned medium on stroke. [Research Support, Non-U.S. Gov't]. J Neurosci Res 90(9):1794–1802.https://doi.org/10.1002/jnr.23063 33. Egashira Y, Sugitani S, Suzuki Y, Mishiro K, Tsuruma K,

Shimazawa M et al (2012) The conditioned medium of murine and human adipose-derived stem cells exerts neuroprotective effects against experimental stroke model. Brain Res 1461:87–95.

https://doi.org/10.1016/j.brainres.2012.04.033

34. Wei X, Zhao L, Zhong J, Gu H, Feng D, Johnstone BH et al (2009) Adipose stromal cells-secreted neuroprotective media against neuronal apoptosis. Neurosci Lett 462(1):76–79.https://doi.org/10.

1016/j.neulet.2009.06.054

35. Zhou Z, Chen Y, Zhang H, Min S, Yu B, He B, Jin A (2013) Comparison of mesenchymal stromal cells from human bone mar- row and adipose tissue for the treatment of spinal cord injury.

Cytotherapy 15(4):434–448.https://doi.org/10.1016/j.jcyt.2012.

11.015

36. Mytych J, Lewinska A, Zebrowski J, Wnuk M (2015) Gold nano- particles promote oxidant-mediated activation of NF-kappaB and 53BP1 recruitment-based adaptive response in human astrocytes.

Biomed Res Int 2015:304575. https://doi.org/10.1155/2015/

304575

37. Zhang L, Wang H, Ding K, Xu J (2015a) FTY720 induces autophagy-related apoptosis and necroptosis in human glioblasto- ma cells. Toxicol Lett 236(1):43–59.https://doi.org/10.1016/j.

toxlet.2015.04.015

38. Zhang L, Yin JC, Yeh H, Ma NX, Lee G, Chen XA et al (2015b) Small molecules efficiently reprogram human astroglial cells into functional neurons. Cell Stem Cell 17(6):735–747.https://doi.org/

10.1016/j.stem.2015.09.012

39. Baez-Jurado E, Vega GG, Aliev G, Tarasov VV, Esquinas P, Echeverria V, Barreto GE (2017) Blockade of neuroglobin reduces protection of conditioned medium from human mesenchymal stem cells in human astrocyte model (T98G) under a scratch assay. Mol Neurobiol.https://doi.org/10.1007/s12035-017-0481-y

40. Bourguignon LY, Gilad E, Peyrollier K, Brightman A, Swanson RA (2007) Hyaluronan-CD44 interaction stimulates Rac1 signaling and PKN gamma kinase activation leading to cytoskeleton function and cell migration in astrocytes. J Neurochem 101(4):

1002–1017.https://doi.org/10.1111/j.1471-4159.2007.04485.x 41. Torrente D, Avila MF, Cabezas R, Morales L, Gonzalez J,

Samudio I, Barreto GE (2014) Paracrine factors of human mesenchymal stem cells increase wound closure and reduce reactive oxygen species production in a traumatic brain injury in vitro model. Hum Exp Toxicol 33(7):673–684.https://doi.org/10.

1177/0960327113509659

42. Loov C, Shevchenko G, Geeyarpuram Nadadhur A, Clausen F, Hillered L, Wetterhall M, Erlandsson A (2013) Identification of injury specific proteins in a cell culture model of traumatic brain injury. [Research Support, Non-U.S. Gov't]. PLoS One 8(2):

e55983.https://doi.org/10.1371/journal.pone.0055983

43. Avila-Rodriguez M, Garcia-Segura LM, Hidalgo-Lanussa O, Baez E, Gonzalez J, Barreto GE (2016) Tibolone protects astrocytic cells from glucose deprivation through a mechanism involv- ing estrogen receptor beta and the upregulation of neuroglobin expression. Mol Cell Endocrinol 433:35–46.https://doi.org/10.

1016/j.mce.2016.05.024

44. Avila Rodriguez M, Garcia-Segura LM, Cabezas R, Torrente D, Capani F, Gonzalez J, Barreto GE (2014) Tibolone protects T98G cells from glucose deprivation. [Research Support, Non-U.S.

Gov't]. J Steroid Biochem Mol Biol 144(Pt B):294–303.https://

doi.org/10.1016/j.jsbmb.2014.07.009

45. Ouyang YB, Xu LJ, Emery JF, Lee AS, Giffard RG (2011) Overexpressing GRP78 influences Ca2+ handling and function of mitochondria in astrocytes after ischemia-like stress.

Mitochondrion 11(2):279–286.https://doi.org/10.1016/j.mito.

2010.10.007

46. Lamers ML, Almeida ME, Vicente-Manzanares M, Horwitz AF, Santos MF (2011) High glucose-mediated oxidative stress impairs cell migration. PLoS One 6(8):e22865

47. Paquet M, Kuwajima M, Yun CC, Smith Y, Hall RA (2006) Astrocytic and neuronal localization of the scaffold protein Na+/

H+ exchanger regulatory factor 2 (NHERF-2) in mouse brain. J Comp Neurol 494(5):752–762.https://doi.org/10.1002/cne.20854 48. Barreto GE, White RE, Xu L, Palm CJ, Giffard RG (2012) Effects of heat shock protein 72 (Hsp72) on evolution of astrocyte activation following stroke in the mouse. Exp Neurol 238(2):284–296.

https://doi.org/10.1016/j.expneurol.2012.08.015

49. Gralinski LE, Bankhead A 3rd, Jeng S, Menachery VD, Proll S, Belisle SE et al (2013) Mechanisms of severe acute respiratory