Human umbilical cord blood mesenchymal stem cells protect mice brain after trauma*

Human umbilical cord blood mesenchymal stem cells protect mice brain after trauma* Elisa R. Zanier, MD; Mery Montinaro, BS; Mariele Vigano, PhD; Pia Villa, PhD; Stefano Fumagalli, BS; Francesca Pischiutta, BS; Luca Longhi, MD; Matteo L. Leoni, MD; Paolo Rebulla, MD; Nino Stocchetti, MD; Lorenza Lazzari, PhD; Maria-Grazia De Simoni, PhD Objective: To investigate whether human umbilical cord blood mesenchymal stem cells, a novel source of progenitors with multilineage potential: 1) decrease traumatic brain injury sequelae and restore brain function; 2) are able to survive and home to the lesioned region; and 3) induce relevant changes in the environment in which they are infused. Design: Prospective experimental study. Setting: Research laboratory. Subjects: Male C57Bl/6 mice. Interventions: Mice were subjected to controlled cortical impact/sham brain injury. At 24 hrs postinjury, human umbilical cord blood mesenchymal stem cells (150,000/5 ␮L) or phosphatebuffered saline (control group) were infused intracerebroventricularly contralateral to the injured side. Immunosuppression was achieved by cyclosporine A (10 mg/kg intraperitoneally). Measurements and Main Results: After controlled cortical impact, human umbilical cord blood mesenchymal stem cell transplantation induced an early and long-lasting improvement in sensorimotor functions assessed by neuroscore and beam walk tests. One month postinjury, human umbilical cord blood mesenchymal stem cell mice showed attenuated learning dysfunction at the Morris water maze and reduced contusion volume compared with controls. Hoechst positive human umbilical cord blood mesenchymal stem

T

cells homed to lesioned tissue as early as 1 wk after injury in 67% of mice and survived in the injured brain up to 5 wks. By 3 days postinjury, cell infusion significantly increased brain-derived neurotrophic factor concentration into the lesioned tissue, restoring its expression close to the levels observed in sham operated mice. By 7 days postinjury, controlled cortical impact human umbilical cord blood mesenchymal stem cell mice showed a nonphagocytic activation of microglia/macrophages as shown by a selective rise (260%) in CD11b staining (a marker of microglia/macrophage activation/ recruitment) associated with a decrease (58%) in CD68 (a marker of active phagocytosis). Thirty-five days postinjury, controlled cortical impact human umbilical cord blood mesenchymal stem cell mice showed a decrease of glial fibrillary acidic protein positivity in the scar region compared with control mice. Conclusions: These findings indicate that human umbilical cord blood mesenchymal stem cells stimulate the injured brain and evoke trophic events, microglia/macrophage phenotypical switch, and glial scar inhibitory effects that remodel the brain and lead to significant improvement of neurologic outcome. (Crit Care Med 2011; 39:2501–2510) KEY WORDS: human cord blood mesenchymal stem cells; traumatic brain injury; microglia; transplantation; functional recovery; brain protection

raumatic brain injury (TBI) is the leading cause of mortality and disability among young individuals in high-income countries and constitutes a major health and socioeconomic problem throughout the world (1). The sequelae of TBI may be divided into primary injury (caused by

the biomechanical impact) and secondary events initiated minutes after TBI and lasting for months. These injury cascades (acting concurrently and often with synergizing effects) include neurotransmitter release, free-radical generation, calciummediated damage, gene activation, mitochondrial dysfunction, and inflammatory

*See also p. 2577. From the Department of Neuroscience (ERZ, MM, PV, SF, FP, L. Longhi, MGDS), Mario Negri Institute for Pharmacological Research; Cell Factory (MV, PR, L. Lazzari), Center of Transfusion Medicine, Cellular Therapy and Cryobiology, Department of Regenerative Medicine, Milano, Italy; CNR (PV), Institute of Neuroscience, Milano, Italy; and the University of Milano (L. Longhi, MLL, NS), Department of Anesthesia and Critical Care Medicine, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico Milano, Italy. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and

are provided in the HTML and PDF versions of this article on the journal’s Web site (www.ccmjournal.com). Partially supported by fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico Milano, Italy. Mr. Fumagalli is recipient of a fellowship from Fondazione Monzino. The authors have not disclosed any potential conflicts of interest. For information regarding this article, E-mail: [email protected] Copyright © 2011 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins

Crit Care Med 2011 Vol. 39, No. 11

DOI: 10.1097/CCM.0b013e31822629ba

responses that lead to delayed cellular dysfunction and death and determine the extent of secondary brain damage. The extended nature of these events and the multiplicity of targets offer the possibility for therapeutic interventions. Recent preclinical studies have revealed that TBI induces neurogenesis, axonal sprouting, and angiogenesis (2, 3), which may contribute to the spontaneous recovery of function suggesting that the promotion of neurorestorative processes may be an additional potential therapy for TBI. Today, however, no single-agent pharmacologic treatment has been successfully translated to the clinical setting (4 – 6) underlining the need to focus on strategies that affect simultaneously multiple injury mechanisms. In this context, mesenchymal stem cells (MSC) represent promising candidates as increasing evidence shows the multipotency of MSC 2501

and their capability to exert a protective effect after injury in different organs through paracrine production of mitogenic, antiapoptotic, and trophic factors through immunomodulatory action (7, 8) and through their ability to efficiently scavenge reactive species (9). Furthermore, MSC have been shown to express neuronal and glial markers in vitro and in vivo (10 –12), thus possessing a specific reparative potential after brain injury. Among MSC, umbilical cord blood (CB)– MSC hold selective advantages. CB-MSC share with bone marrow (BM)–MSC similar expression of a considerable number of transcripts but express trophic factors (13) and matrix remodelling genes to a higher extent (14, 15). Furthermore, compared with other postnatal MSC sources, CB provides stem cells with a lower immunogenic potential in vitro and in vivo (16). This evidence, associated with the noninvasive collection after birth and the lack of donor-site morbidity, make them an interesting therapeutic tool. In the setting of acute brain injury, protective effects of CB-MSC have been reported in brain ischemia (13, 17); however, as a result of the fact that TBI and ischemia differ in the nature of the primary insult show selected cellular vulnerability and involve activation of different pathogenetic cascades, translation of the results across these two types of injury is not possible. In TBI, early studies show that MSC obtained from BM may partially improve sensorimotor functions in rodents (18 –21). To our best knowledge, the effects of CB-MSC after TBI have not been explored so far. In the present study, we administered human CB-MSC intracerebroventricularly (ICV) to adult TBI mice to explore whether CB-MSC are able to decrease TBI sequelae and restore brain function, to survive and home to the lesioned region, and to induce relevant changes in the environment in which they are infused. We investigated whether the CB-MSC protective effect is mediated by the induction of brain-derived neurotrophic factor (BDNF) by alternative activation of microglia/macrophages (M/M) and by a damping of gliotic scar. The present study shows for the first time that CBMSC injected 24 hrs post-TBI into the contralateral brain ventricle induce a sensorimotor improvement detectable as early as 1 wk after injury that persist for the whole duration of the study. Furthermore, we also observed a significant re2502

duction of cognitive deficits paralleled by a structural neuroprotective effect 1 month after injury.

MATERIALS AND METHODS Preparation of CB-MSC. Human CB was collected after written informed consent of the mother from full-term newborns and isolation of MSC was performed as previously described (14). Briefly, within 12 hrs, CB was centrifuged, plasma discarded, and an enrichment protocol was performed by using Rosette Sep enrichment cocktail (StemCell Technologies, Vancouver, Canada). The blood was diluted, laid on density gradient Lympholyte-H (1.077 g/mL; Cedarlane, Ontario, Canada), and centrifuged (1200 rpm for 25 mins). The lowdensity cell fraction was collected and washed twice. Cell suspension was seeded in the presence of ␣-Modified Eagle Medium (Life Technologies, Carlsdad, CA) supplemented with 20% fetal bovine serum (Biochrom, AG, Berlin, Germany) and 2 mM L-glutammine (Life Technologies). After overnight incubation, the medium was replaced with fresh medium. Colonies were observed after 2–3 wks. Cells were detached using trypsin-EDTA (Life Technologies) when subconfluent (80%), counted, and split 1:3. CB-MSC from third to sixth passages (approximately between days 15 and 30 from the beginning of culture) were used for the in vivo experiments. The cumulative population doublings rate was determined using the following formula: PD ⫽

Log10 共N兲 Log10 共2兲

N is the ratio between the harvested cells and the seeded cells. The PD for each passage was calculated and added to the PD of the previous passages to generate data for cumulative number of population doublings. Flow Cytometry Analysis. CB-MSC were characterized by flow cytometry during culture. Cells (106) were washed in phosphatebuffered saline (PBS) for 20 mins at room temperature and incubated in the dark with the following directly conjugated mouse– antihuman antibodies: CD90-PE, CD73-PE, CD44-FITC, CD10-FITC, HLA-class II-PE, and CD34-PE (Becton Dickinson, BD, San Jose`, CA), CD105-PE (ImmunoTools, Friesoythe, Germany), ␣-SMA FITC (Sigma-Aldrich, St Louis, MO), CD45-APC, HLA-class I-FITC (Beckman Coulter, Fullerton, CA), CD133-PE (Miltenyi Biotec, Gladbach, Germany), and CD146-FITC (BioCytex, Marseille, France). The isotype-matched immunoglobulins IgG1PE-FITC (Chemicon, Temecula, CA) and IgG1PC7 (Beckman Coulter) were used as negative controls under the same conditions. At least 50,000 events were acquired with a cytomics FC500 flow cytometer (Beckman Coulter) and plots were generated using the CXP analysis software (Beckman Coulter).

RNA Isolation, Reverse Transcriptase– Polymerase Chain Reaction and Real-Time Reverse Transcriptase–Polymerase Chain Reaction Analysis. Total RNA was extracted from 4 ⫻ 105 to 1 ⫻ 106 CB-MSC using the RNeasy Mini Kit (Qiagen AG, Hilden, Germany). The contaminating genomic DNA was further digested by DNase (Qiagen). RNA quality was assessed by spectrophotometric and electrophoretic analysis. Samples (800 ng) were retrotranscripted to cDNA using iScriptTM cDNA Syntesis Kit (Bio-Rad Laboratories Ltd, Hercules, CA). cDNA (40 ng) was used for each polymerase chain reaction (PCR) assay using GoTaq (Promega Corporation, Madison, WI). Primers used (400 nM) constructed on the basis of published human sequences (Primer3 software 0.4.0; Whitehead Institute for Biomedical Research, Cambridge, MA) are listed in Supplementary Table 1 (see Supplemental Digital Content 1, http://links.lww.com/CCM/ A276). Each set of oligonucleotides was designed to span two different exons and controlled using Blast software. Positive controls were obtained from the corresponding fetal tissues. Glyceraldehyde-3-phosphate-dehydrogenase was used as a housekeeping gene. Samples were loaded on 2% agarose gels. Real-time reverse transcriptase– PCR was performed by using the CFX96 RealTime PCR detection system (Bio-Rad). Amplifications were carried out in a final reaction volume of 20 L with the SsoFast Eva Green Supermix (Bio-Rad), 20 ng cDNA template, and 400 nM primers. The melting curves were acquired after PCR to confirm the specificity of the amplified products. The relative amount of gene expression for each CB-MSC line (n ⫽ 3), normalized to two housekeeping genes, glyceraldehyde-3-phosphate-dehydrogenase and ␤-2microglobulin, was compared with that of MCF7 (a human breast cancer cell line) as a positive control sample. Animals. All procedures were conducted conform to the institutional guidelines that are in compliance with national (D.L. n.116, G.U. suppl 40, February 18, 1992) and international laws and policies (EEC Council Directive 86/609, OJL 358,1; December 12, 1987; National Institutes of Health Guide for the Care and Use of Laboratory Animals, US National Research Council 1996) (22, 23). The study was reviewed and approved by the local ethics committee of the Mario Negri Institute. Male C57Bl/6 mice (20 –24 g) were housed in a SPF vivarium at a constant temperature (21 ⫾ 1°C) with a 12-hr light– dark cycle and ad libitum access to food and water. Study Design and Blinding of In Vivo Studies. A total of 232 mice, divided into four experimental conditions, were used for the study: 1) controlled cortical impact (CCI) mice receiving CB-MSC (150,000/5 ␮L ICV) 24 hrs after injury; 2) sham mice receiving CB-MSC (150,000/5 ␮L ICV) 24 hrs after surgery; 3) CCI mice receiving PBS (5 ␮L ICV) 24 hrs after injury; and 4) sham mice receiving PBS (5 ␮L ICV) 24 hrs after surgery. The following sets of mice were used: 1) 18 mice per group

Crit Care Med 2011 Vol. 39, No. 11

Figure 1. Experimental design. Umbilical cord blood-mesenchymal stem cell (CB-MSC) were infused intracerebroventricularly (icv) in the contralateral ventricle 24 hrs after controlled cortical impact (CCI)/sham surgery. All mice were treated with cyclosporine A (CSA). Behavioral tests, histology, immunohistochemistry, and Western blot experiments were performed at the time points indicated. ip, interperitoneal.

were used for behavioral analysis up to 5 wks postinjury; after euthanasia, brain tissues were processed for subsequent evaluation of contusion volume (n ⫽ 12), CB-MSC distribution (n ⫽ 12), anti-CD11b, anti-CD68, and antiglial fibrillary acidic protein (GFAP) immunohistochemistry (n ⫽ 8); 2) eight mice per group were euthanized at 7 days postinjury for evaluation of contusion volume, CB-MSC distribution, anti-CD11b, and anti-CD68 immunohistochemistry; 3) eight mice per group were euthanized 3 days postinjury for cortical BDNF measurements; 4) 12 mice per group injected with CB-MSC obtained from two additional donors (B-CB-MSC and C-CB-MSC) were tested behaviorally at 7 days to assess if the observed protection was significantly influenced by a specific blood donor. All surgeries and injuries were performed by the same investigator who was blinded to the treatment allocation. Mice were assigned to surgery and treatment groups with surgery and treatment distributed equally across cages and days. Investigators who performed behavioral assessments and those who measured contusion volume, CB-MSC distribution, CD11b, CD68, and glia fibrillary-associated protein counts and Western blot analysis were blinded to the treatment. The experimental design is illustrated in Figure 1. Experimental Brain Injury. Mice were anesthetized with intraperitoneal administration of sodium pentobarbital, 65 mg/kg, and placed in a stereotaxic frame. They were subjected to craniectomy followed by induction of CCI brain injury as previously described (23–25). Our model of injury uses a 3-mm rigid impactor driven by a pneumatic piston, rigidly mounted at an angle of 20° from the vertical plane, and applied perpendicularly to the exposed dura ma-

Crit Care Med 2011 Vol. 39, No. 11

ter over the left parietotemporal cortex at a velocity of 5 m/s and depth of 1 mm. The craniotomy was then covered with a cranioplasty and the scalp sutured. During all surgical procedures, mice were maintained at the body temperature of 37°C. Sham-injured mice received identical anesthesia and surgery without brain injury. The mortality rate was 5%. Immunosuppression. For the first 2 wks after CCI/sham brain injury, all animals received a daily injection of cyclosporine A (Sandimmun; Novartis Farma S.p.A., Origgio, Italy; 10 mg/kg, intraperitoneally, the first dose was given 1 hr after surgery). Thereafter, cyclosporine A was administered three times a week up to the time of euthanasia. Hoechst Staining, Cell Preparation, and Transplantation. Before transplantation, CB-MSC were stained with 5 ␮g/mL Hoechst-33258 (Sigma-Aldrich) for 1 hr at 37°C and subsequently resuspended in PBS. Cell number was evaluated by light microscopy. Viability of CB-MSC after Hoechst staining was evaluated by the Trypan blue exclusion test and cell concentration was adjusted to 150,000 cells/5 ␮L PBS. Twenty-four hrs after CCI, a hole was drilled at the scalp contralateral to the injured side at coordinates 0 mm caudal to bregma, ⫺1 mm lateral to the midline, and ⫺3 mm beneath the dura mater in anesthetized mice. CB-MSC were infused ICV over 5 mins and the needle was left in place afterward for additional 5 mins. Control mice were infused with PBS alone (5 ␮L) following the same procedures. No animals died after transplantation. Behavioral Tests. Sensory motor deficits were evaluated by neuroscore (23) and beam walk test (26) the day after surgery and then weekly up to 4 wks after CCI. For neuroscore, animals were scored from 4 (normal) to 0

(severely impaired) for each of the following indices: 1) forelimb function; 2) hind limb function; and 3) resistance to lateral pulsion as previously described (23, 24). The maximum score per animal is 12. The beam walk test (27) measures the number of foot faults of a trained mouse walking twice on an elevated and narrow wooden beam (5-mm wide and 100 cm length). The best score is 0. Evaluation of cognitive function was performed using the Morris water maze (27). A circular pool (1-m diameter) filled with water (18 –20°C) made opaque by a nontoxic white paint and a fixed submerged platform (1 cm below the water surface) was used. The learning task consisted of eight trials/day for 3 consecutive days for a total of 24 trials. Latencies to reach and climb onto the platform were recorded for each trial with a maximum of 60 secs per trial. Cognitive performance was obtained by averaging the latencies of 24 trials over 3 days (23). Five days after the learning task, mice were tested for their ability to remember the location of the submerged platform. Animals were allowed to swim for 60 secs with the platform removed, and their swim paths were recorded using a computerized video analysis system (Ethovision XT 5.0; Noldus Information Technology, Wageningen, The Netherlands). A memory score was calculated to grade memory retention deficits in the different groups (28). Contusion Volume. At 1 and 5 wks, perfused brains were obtained (23, 24) and cryosectioned at 20 ␮m. Eight sections (bregma ⫹0.6; 0; ⫺0.8; ⫺1.5; ⫺2.25; ⫺2.65; ⫺3.25; and ⫺4 mm) were stained with Neutral Red (Sigma-Aldrich). Images were acquired on a computer using the image analyzer Analytical Image System (Imaging Research Inc, Brock University, St Catharines, Ontario, Canada) and contusion volume was calculated as previously described (23). Western Blot Analysis. At day 3, mice ipsilateral cortical areas (including all the tissue above the rhinal fissure) (29) were dissected out, rapidly frozen on dry ice, and stored at ⫺80°C until analysis. Samples were homogenized by sonication in lysis buffer (10 mM Tris, 10 mM EDTA, 10 mM EGTA, 0.1 mM phenylmethylsulfonyl fluoride, cocktail of protease inhibitors, F complete; Roche, Basel, Switzerland). Homogenates were centrifuged at 14,000 g for 15 mins at 4°C. Protein concentration of each sample was determined by Bradford method (Protein Assay; Bio-Rad). Equal amounts of protein (20 ␮g/sample) were electrophoresed on a 14% sodium dodecyl sulphate–polyacrylamide gel and transferred to polyvinyl difluoride membranes. Incubation with primary antibody was performed using anti-BDNF polyclonal antibody (1:500; Santa Cruz Biotechnology, Santa Cruz, CA) for 3 hrs followed by 1-hr incubation with horseradish peroxidase-conjugated secondary antibody (1: 1000; Santa Cruz Biotechnology). Immunocomplexes were visualized by chemiluminescence using the ECL Western blot substrate

2503

(Pierce, Thermo Scientific, Rockford, IL). Results were standardized using ␤-actin as the control protein (primary antibody anti-␤-actin 1:15,000; Chemicon, Millipore, Billerica, MA; horseradish peroxidase-conjugated secondary antibody 1:20,000; Santa Cruz Biotechnology). Quantification of the prosurvival 14 kDa mature isoform of BDNF was carried out using Quantity One Software (Bio-Rad). Immunohistochemistry. Immunohistochemistry was performed on 20-␮m brain coronal sections using rat anti-CD11b (1:1000, kindly provided by Dr Doni [30]), rat antiCD68 (1:200; Serotec, Kidlington, UK), or mouse GFAP (1:2000; Chemicon) to measure M/M activation, phagocytic activity, and astrogliosis, respectively. Positive CD11b, CD68, or GFAP cells were stained by reaction with 3,3diaminobenzidine-tetrahydrochloride (Vector Laboratories, Burlingame, CA) as previously described (31). For each reaction, adequate negative controls were performed. Brain coronal sections per mouse (at 0.4, 1.6, and 2.8 mm posterior to bregma), were used to quantify CD11b and CD68-stained area. See distribution of frames in Supplemental Figure S1 (see Supplemental Digital Content 1, http://links.lww.com/CCM/A276). Quantitative analysis was thus performed in defined anatomic boundaries, namely contusion edge and cortical boundary zone, and by acquiring the same focal plan throughout the samples (see Fig. S1 [see Supplemental Digital Content 1, http://links.lww.com/CCM/A276] [32]). For quantification of GFAP-positive area, two regions of interest with distinct GFAP-positive cell morphology were identified, namely a scar region, from the contusion edge and extending up to 500 ␮m, and a periscar region adjacent to the scar region and extending for further 500 ␮m. Based on this observation, five fields in the scar and five in the periscar region at each selected stereotactic coordinate were acquired (Fig. S1 [see supplemental Digital Content 1, http://links.lww.com/CCM/A276]). An Olympus BX61 microscope equipped with a motorized stage and managed with AnalySIS software (Olympus, Tokyo, Japan) was used for image acquisition. Immunostained area for each marker was measured using ImageJ software (http://rsbweb.nih.gov/ij/) and expressed as positive pixels/total assessed pixels and indicated as staining percentage area. Immunofluorescence and Confocal Analysis. Immunofluorescence was performed on 20-␮m coronal sections as previously described (22). Primary antibodies used were: antimouse neuronal nuclei (1:250) and antimouse NG-2 (1:200; Chemicon), antimouse BDNF (1:800; Santa Cruz Biotechnology), antimouse CD11b (1:1000), anti-GFAP (1:2000), anti-CD68 (1:200). Fluoroconjugated secondary antibodies used were: Alexa 546 antirat, Alexa 594 antirabbit, and Alexa 488 antimouse (all 1:500; Invitrogen). Biotinylated antirabbit or antirat (1:200; Vector Laboratories) was

2504

also used followed by fluorescent signal coupling with streptavidin Alexa 647 conjugated (Invitrogen) or TSA amplification kit (Perkin Elmer, Waltham, MA). Immunofluorescence was acquired using a scanning sequential mode by an IX81 microscope equipped with a confocal scan unit FV500 with three laser lines: Ar-Kr (488 nm), He-Ne red (646 nm), and He-Ne green (532 nm) (Olympus) and a ultraviolet diode. Statistical Analysis. For statistical analysis, we used standard software packages (GraphPad Prism version 4.00 or JMP version 7.0). All data are presented as mean and SD. For beam walk, neuroscore, and learning latencies, the comparison between groups was performed using two-way analysis of variance for repeated measurements followed by Tukey post hoc test. For memory score, contusion volumes, and BDNF concentrations, the comparison between groups was performed using two-way analysis of variance followed by Tukey post hoc test. Two-tailed unpaired t test was used for analysis of CD11b, CD68, and GFAP data. Efficacy of CB-MSC obtained from three different cord blood donors was assessed by one-way analysis of variance. Probability values ⬍.05 were considered statistically significant. Assumptions of normality were checked

using Kolmogorov-Smirnov test. Where the normality assumption was violated (as for beam walk, neuroscore, and memory score), Kruskal-Wallis/Friedman tests followed by appropriate post hoc tests were performed to crosscheck the parametric analysis: significance remained identical. Thus, only the results of parametric analysis are reported throughout the text.

RESULTS CB-MSC, isolated by mononuclear cells after lineage-negative depletion, were obtained after 2–3 wks of culture and a homogeneous population of fibroblastic-like cells with a spindle-shaped morphology was observed during passages (Fig. 2A). The growth rate was calculated as cumulative number of population doublings (in Fig. S2A a representative proliferation rate [see Supplemental Digital Content 1, http://links.lww.com/CCM/A276]). By quantitative PCR, the expression of senescence-related genes (p53, c-myc, and human telomerase reverse transcriptase– hTERT) was detected in CB-MSC at

Figure 2. CB-MSC culture and characterization. A, Representative image of human umbilical cord blood-mesenchymal stem cell (CB-MSC) showing their typical spindle-shape morphology during passages, bar ⫽ 100 ␮m. B, CB-MSC were analyzed by reverse transcriptase–polymerase chain reaction for their potential to express neuronal and glial genes. Lanes represent: positive control obtained from fetal brain tissue (lane 1); negative control for neural genes obtained from mononuclear cells (lane 2); CB-MSC (lane 3) and negative control for reverse transcriptase–polymerase chain reaction with the reaction mix without the complementary DNA (lane 4). C, Representative CB-MSC immunophenotype profile at passage 4 showed the expression of the typical cell-surface antigens of MSC: CD90, CD73, CD44, CD105, ␣ smooth muscle actin, CD10, HLA I, and CD146. D, Immunofluorescence analysis revealed that CB-MSC in culture are able to express brain-derived neurotrophic factor (BDNF) (purple). Nuclei are counterstained with diamidino phenyl indole (DAPI) (blue). As a negative control, cells were incubated with DAPI and fluorescent secondary antibody in the absence of anti-BDNF antibody (E). Bar ⫽ 50 ␮m.

Crit Care Med 2011 Vol. 39, No. 11

passages 3 and 6. As shown in Fig. S2B–D (see Supplemental Digital Content 1, http://links.lww.com/CCM/A276), the expression level of p53 was stable between passages, whereas hTERT and c-myc did not show any measurable value. We then determined if CB-MSC were able to express neuronal and glial genes (Fig. 2B). CB-MSC were positive for nestin, neuronal cell adhesion molecule, ␤-tubulin III, neuron-specific enolase, and microtubuleassociated protein 2 as assessed by reverse transcriptase–PCR. At passage 4, CB-MSC immunophenotype profile showed that they expressed typical cell-surface antigens of MSC as CD90, CD73, CD44, CD105, ␣ smooth muscle actin, CD10, HLA I, and CD146, whereas they were negative for hematopoietic and endothelial markers as CD45, CD34, CD133, and HLA II (Fig. 2C). Sensorimotor Deficits. Seven days after CCI, mice receiving PBS injection showed a significantly higher number of foot faults (Fig. 3A) and a lower neuroscore (Fig. 3B) compared with sham mice, indicating an impairment in sensorimotor activity measured in two independent tests (beam walk: p ⬍ .0001, neuroscore: p ⬍ .0001). These deficits were significantly attenuated in CCI mice infused with CB-MSC showing an early effect of CB-MSC on recovery of function (Fig. 3A–B) (beam walk: p ⬍ .001, neuroscore: p ⬍ .001). To understand if CBMSC-protective effects were persistent, CCI mice receiving PBS or CB-MSC were exposed to both sensorimotor tests every week for the entire duration of the study. The score of all groups of CCI mice recovered slightly in both tests over the 4-wk observation period. Similarly to what was observed at 7 days, CCI CB-MSC mice performed significantly better compared to those injected with PBS for the entire duration of the study (Fig. 3A–B) (4 wks postinjury beam walk: p ⬍ .001, neuroscore: p ⬍ .001). We further assessed whether this protective effect was dependent on a specific cord blood donor. Therefore, we compared the effects of CB-MSC obtained from three different cord blood donors (code: A-CBMSC, B-CB-MSC, and C-CB-MSC) on sensorimotor deficits. The tested cell lines were equally able to significantly improve both beam walk (Fig. S3A [see Supplemental Digital Content 1, http://links.lww.com/ CCM/A276]) and neuroscore (Fig. S3B [see Supplemental Digital Content 1, http:// links.lww.com/CCM/A276]) at 7 days (beam walk: A, B, and C-CB-MSC vs. PBS p ⬍ Crit Care Med 2011 Vol. 39, No. 11

Figure 3. Protective effects of umbilical cord blood-mesenchymal stem cell (CB-MSC) on sensorimotor and cognitive functions. Brain-injured mice receiving phosphate-buffered saline (pbs) (controlled cortical impact [CCI] pbs) showed a significant motor function deficit during the entire duration of the study assessed by beam walk (A, mean ⫾ SD, n ⫽ 16 –18) and neuroscore (B, mean ⫾ SD, n ⫽ 16 –18). In both tests, brain-injured mice receiving stem cells (CCI CB-MSC) showed a significant attenuation of neurologic motor deficits already at 7 days postinjury that persisted up to 28 days postinjury compared with the CCI pbs group. Effect of CB-MSC on learning was assessed as latency to locate the hidden platform in the Morris water maze conducted 28 –30 days after surgery (C, mean ⫾ SD, n ⫽ 16 –18). CB-MSC significantly improved cognitive performance after CCI leading to a shorter escape to platform on day 3 of training (p ⬍ .001). Five days after learning, CCI pbs but not CCI CB-MSC mice showed a clear memory retention deficit (D, mean ⫾ SD, n ⫽ 16 –18). Beam walk: two-way analysis of variance (ANOVA) for repeated measurements: p ⬍ .0001, p interaction ⬍.001; neuroscore: two-way for repeated measurements: p ⬍ .0001, p interaction ⬍.0001. latency: two-way ANOVA for repeated measurements: p ⬍ .001, p interaction ⫽ .013, memory score: two-way ANOVA for repeated measurements: p ⫽ .001, p interaction ⫽ .47, post hoc Tukey’s test *p ⬍ .05, **p ⬍ .01, ***p ⬍ .001 CCI cells compared with CCI pbs mice, ##p ⬍ .01 CCI cells compared with sham mice.

.001, neuroscore: A, B, and C-CB-MSC vs. PBS p ⬍ .001). Cognitive Function. Four weeks after brain injury, mice were evaluated for their ability to learn the position of a hidden platform in the Morris water maze. All animals were able to swim without visible alterations in swimming ability, as reflected by decreasing latencies to find the platform over the 3-day period (Fig. 3C). However, although CCI PBS mice showed a robust learning dysfunction, CCI CB-MSC performed significantly better than CCI PBS mice (mean latency over 3 days: 48.98 ⫾ 5.18 and 41.38 ⫾ 7.51 sec respectively, twoway analysis of variance p ⬍ .0001, interactions p ⬍ .001, Tukey posttest p ⬍ .05). Notably, this effect was particularly evident on day 3 (Fig. 3C, p ⬍ .01). Likewise, 5 days after learning, CCI PBS mice showed a

clear memory retention deficit that was not present in CCI CB-MSC mice (Fig. 3D), although the difference between CB-MSC and PBS groups did not reach significance (p for interaction ⫽ .47). Cell Distribution. Because infused Hoechst-positive cells were often packed or clustered preventing cell count, we performed a qualitative assessment of CB-MSC presence and distribution in the brain of sham-operated and CCI mice. Seven days after surgery, Hoechstpositive cells were mostly located in the contralateral ventricle (injection site) of sham mice with only few cells located along the ICV injection route (Fig. S4A–D [see Supplemental Digital Content 1, http://links.lww.com/CCM/A276]). Notably none of them could be found into the parenchyma. Conversely, in CCI mice, 2505

Figure 4. Protective effect of umbilical cord blood-mesenchymal stem cell (CB-MSC) on anatomic damage. Contusion volume at 7 and 35 days after injury: Controlled cortical impact phosphate-buffered saline (CCI pbs) mice showed a significant increase in contusion volume at 35 days compared with 7 days after injury. Conversely, no significant increase in the contusion volume was detected in CCI CB-MSC mice over the 35 days, leading to a significant difference at this time point between the two groups. Data are expressed as mean ⫾ SD, n ⫽ 7–12, p ⫽ .0006; p interaction ⫽ .0348, two-way analysis of variance followed by Tukey’s test. *p ⬍ .05, **p ⬍ .01.

CB-MSC were juxtaposed to the ventricular walls in both hemispheres, and some cells were clearly diffusing from the ventricles into the injured cortex where they could be found in 67% of mice (Fig. S4E–H [see Supplemental Digital Content 1, http://links.lww.com/CCM/A276]). Thirty-five days after sham or CCI injury, the general distribution of CB-MSC was similar to that observed at day 7. CB-MSC were still present into the ventricles of sham-injured mice, but again no cells could be found into the sham-injured hemisphere. At this time point, 45% of mice showed CB-MSC in the contused tissue. To assess whether the presence of infused cells in the contused tissue was related to behavioral and anatomical protection, mice were divided in two groups based on the presence (n ⫽ 5) or absence (n ⫽ 6) of injected cells in the contused tissue. No difference on functional deficits (Fig. S5A–B [see Supplemental Digital Content 1, http://links.lww.com/ CCM/A276]) and anatomic damage (Fig. S5C [see Supplemental Digital Content 1, http://links.lww.com/CCM/A276]) was detected in these two groups 5 wks after injury. Anatomical Damage. No effect of CBMSC on contusion volume was observed 7 days after CCI (Fig. 4). As expected (23), over 35 days postinjury, we observed a progressive increase in contusion volume in CCI PBS mice (p ⬍ .01). Conversely, no significant increase in the contusion volume was detected in CCI CB-MSC mice over time, leading to a significant difference at this time point between CB-MSC (13.72 ⫾ 1.78 mm3) and PBS- (17.33 ⫾ 1.35 mm3, p ⬍ .05) treated mice, thus indicating 2506

Figure 5. Umbilical cord blood-mesenchymal stem cell (CB-MSC) increase brain-derived neurotrophic factor (BDNF) production after controlled cortical impact (CCI). Western blot analysis on cortical tissue extracts 3 days after injury revealed a significant increase of BDNF in CCI CB-MSC compared with CCI phosphate-buffered saline (pbs) mice. A, Typical blots; (B) mean ⫾ SD, n ⫽ 8. p ⫽ .0006, p interaction ⫽ .0395, two-way analysis of variance followed by Tukey’s test *p ⬍ .05, ***p ⬍ .001.

that CB-MSC were able to rescue pericontusional tissue or to prevent the progression of the lesion. CB-MSC Increase BDNF Expression After TBI. Immunofluorescence analysis showed that CB-MSC in culture were already positive for BDNF (Fig. 2D). We examined whether injected cells could induce an increased BDNF expression early after injury. By 3 days postinjury, Western blot analysis of the cortical contusion core and bordering regions revealed that CCI induced a significant decrease of the mature isoform of BDNF in CCI PBS mice compared with sham animals (72.88 ⫾ 36.10 and 185.70 ⫾ 35.50 arbitrary units, respectively, p ⬍ .001). More importantly, cell infusion after CCI significantly increased BDNF concentration in CCI CB-MSC compared with CCI PBS mice (131.50 ⫾ 49.10 arbitrary units, p ⬍ .05) and restored BDNF expression close to the levels observed in sham operated animals (Fig. 5A–B). By 7 days postinjury, double-label confocal microscopy demonstrated that a part of CB-MSC injected cells expressed BDNF and that among resident cells, within the pericontusional region, neuronal nuclei and GFAP-positive cells were also positively stained for BDNF (Fig. S6A–O [see Supplemental Digital Content 1, http://links.lww.com/CCM/A276]).

Stem Cells Infusion Affects CD68/ CD11b Ratio in Injured Brain Tissue. Our previous work had shown that after brain ischemia, M/M activation is required for stem cell protective action (30). To understand whether a similar mechanism is relevant after CCI, we analyzed CD11b staining, a marker of active M/M. Being interested to determine parameters of glial activation which is related to morphology, we quantified stained areas of positive cells. By 7 days postinjury, when the protective effect on sensorimotor functions began to be apparent, we observed a 260% rise in CD11b-positive cells of CCI CB-MSC compared with CCI PBS mice (0.13 ⫾ 0.02 and 0.05 ⫾ 0.02, p ⬍ .001; Fig. 6E). In addition, we could observe that in CCI CB-MSC mice CD11b-positive cells displayed a highly ramified morphology (Fig. 6B), differently from CCI PBS mice in which CD11bpositive cells had a prevalent globular morphology (Fig. 6A). We then assessed whether M/M activation rise was the result of an increase in phagocytic activity. Quantification of CD68, a M/M phagocytic marker, showed a 42% decrease in CD68-positive-stained area in CCI CB-MSC compared with CCI PBS mice (0.07 ⫾ 0.03 and 0.12 ⫾ 0.03, respectively, p ⬍ .01) (Figs. 6C–D and 6F), thus leading to a decrease in the CD68/CD11b ratio in CCI CB-MSC compared with CCI PBS mice (p ⬍ .01, Fig. 6G). The confocal analysis revealed that the high number of ramified CD11bpositive cells of CCI CB-MSC mice were associated with low levels of CD68 (Fig. 6K–M), whereas globular CD11b cells in CCI PBS mice were prevalently highly positive to CD68 (Fig. 6H–J). Five weeks after injury, the positive stained area in the CCI CB-MSC group was still higher for CD11b (by 125%) and lower for CD68 (by 44%) compared with CCI PBS mice, but significant changes were no longer detected at this time point (data not shown). Attenuation of Gliotic Scar. Thirtyfive days after injury, in the scar region, GFAP-positive cells showed a stretched morphology and no positivity for BDNF, suggestive of scarring astrocytes with no trophic function (Fig. 7A). In this region, CCI CBMSC mice showed a selective decrease in GFAP positively stained area compared with CCI PBS mice (p ⬍ .01, Fig. 7B). Conversely, no significant difference in GFAP expression could be observed in the two groups in the periscar region where GFAP cells showed a ramiCrit Care Med 2011 Vol. 39, No. 11

Figure 6. Expression of CD11b and CD68 in mice receiving Umbilical cord blood-mesenchymal stem cell (CB-MSC) or phosphate-buffered saline (pbs) infusion. Micrographs represent CD11b (A–B) or CD68 (C–D) immunostaining 7 days after injury on pericontusional cortex of mice infused with pbs (A, C) or CB-MSC (B, D). Bar: 50 ␮m. Quantification of microglia/macrophage activation states by anti-CD11b immunostaining and of phagocytic activity by anti-CD68 immunostaining is shown in E and F, respectively. Controlled cortical impact (CCI) mice infused with CB-MSC showed an increase in CD11b staining and a decrease in CD68 phagocytic activity compared with those infused with pbs leading to the drop of CD68/CD11b ratio (G). Data are expressed as percentage of stained area over the total sampled area and reported as mean ⫾ SD of 33 frames/mouse (n ⫽ 8). **p ⬍ .01, ***p ⬍ .001, unpaired t test. Confocal microscopy for CD11b (H, K) and CD68 (I, L) showed that in CCI pbs mice, CD11b cells were prevalently globular (H) and highly positive to CD68 (J), whereas CCI CB-MSC mice displayed a high number of ramified CD11b-positive cells (K) expressing low levels of CD68 (M). BDNF, brain-derived neurotrophic factor.

fied morphology and BDNF positivity (Fig. 7C–D).

DISCUSSION In the present study, we demonstrated that infused CB-MSC were able to induce a significant attenuation of neurobehavioral deficits as early as 1 wk after injury that persisted for the whole period of the study. Notably we also observed a delayed structural neuroprotective effect paralleled by a glial scar-inhibitory action, which we attributed to a trophic mechanism and to the activation of selected aspects of the inflammatory response occurring soon after cell transplantation. Crit Care Med 2011 Vol. 39, No. 11

To our best knowledge, this is the first study to show long-lasting functional recovery, including sensorimotor and cognitive as well as anatomical protection after CB-MSC infusion in TBI, thus providing evidence that CB-MSC can represent a novel and attractive strategy for TBI therapy. Indeed, in the field of regenerative medicine, MSC are one of the most promising cell target. Over the adult MSC counterpart, CB shows distinctive advantages, including ease of procurement, immediate availability, lower risk to donors, greater immunotolerance, and lower incidence of severe graft-vs.-host disease (33). Further-

more CB represents an ethically noncontroversial, clinically relevant, and easily accessible source that can be stored in banks and become available to a wide population. Notably, to date, there are ⬎300,000 units stored worldwide (34). So far the effects of CB on TBI have been explored in the study by Lu and collaborators (35) using the whole blood. They could show an improvement on sensorimotor function, although no investigation on cognitive function, anatomic damage, or any mechanistic insight was done. Because CB includes hematopoietic, mesenchymal, and endothelial progenitors as well as autoimmune T cells, mononuclear, and mast cells, they investigated the effects of a mixed cell population, thus not providing any insight on the relevance of a selected cell population for TBI treatment. We show that CB-MSC induce an early and persistent beneficial effect on sensorimotor deficits as well as attenuate both cognitive deficits and anatomical damage 35 days after CCI suggesting a stronger and broader effect of CB-MSC compared with previous observations with BM-MSC (18 –21). Our group has recently performed a direct comparison between CB-MSC and human BM-MSC in a model of acute kidney injury in mice (14), showing that the infusion of CB-MSC improved animal survival to a higher extent than human BM-MSC. A possible explanation for the greater potential of CB-MSC to confer protection after injury may rest on the evidence that CBMSC express trophic and matrix remodeling genes at a higher extent compared with BM-MSC (15). Assessment of expression of genes related to senescence of MSC is a relevant issue not only for basic research, but also in a translational perspective. In particular, p53 functions as a longevity assurance gene and as a regulator of aging (36); c-myc expression is related to the process of cellular transformation (37); hTERT is the telomerase catalytic subunit and is involved in the telomereshortening process in cell senescence (38). In this context and to be confident for future clinical applications, the expression of these crucial markers was evaluated during passages and no relevant expression was found. An additional question addressed in the study has been to assess whether the observed protection is influenced by a specific blood donor. We therefore compared the effects of CB-MSC obtained from three different cord blood donors on 2507

Figure 7. Umbilical cord blood-mesenchymal stem cell (CB-MSC) attenuate gliotic scar formation. Immunoreactivity for glial fibrillary acidic protein (GFAP) 35 days after injury in mice receiving CB-MSC or phosphate-buffered saline (pbs) infusion. Representative confocal montages of the scar (A) and periscar (C) region. In the periscar tissue, most GFAP positive cells were also brain-derived neurotrophic factor (BDNF)-positive cells were also BDNF-positive cells, whereas all GFAP-positive cells at the contusion edge were BDNF-negative (A). CB-MSC infusion resulted in a decrease of GFAP positively stained area in the scar (B) but not in the periscar (D) region, mean ⫾ SD, n ⫽ 8. *p ⬍ .05, unpaired t test. Bars: 20 ␮m.

sensorimotor deficits. All cell lines were equally able to improve sensorimotor functions assessed by two independent tests at 7 days, thus indicating that CB-MSC’s protective effect may be independent from specific cord blood donors. This is a relevant finding in view of a possible clinical development. In the same translational perspective, we choose the ventricular administration. Intraventricular cannulation in human TBI is invasive and may have significant complications; however, it is recommended by authoritative guidelines for intracranial pressure monitoring of severe TBI admitted to the intensive care unit (39, 40). In these patients, this site of injection would therefore be free from additional surgical complications. CB-MSC injected 24 hrs after CCI were capable of homing to lesioned tissue as early as 1 wk after injury in 67% of mice. In sham-operated mice, conversely, the infused cells did not diffuse into the parenchyma. This finding is consistent with our own and literature data showing greater stem cell numbers in the injured brain (18, 41). Local disruption of the blood– brain barrier, along with various mechanisms promoting homing to areas of inflammation such as upregulation of chemokines (42, 43), may be responsible for this phenomenon. Transplanted cells were able to survive at least for 5 wks postinjury in the contused tissue. In this time window, however, CB-MSC did not express cell markers for neurons, astrocytes, and oligodendrocytes. Furthermore, the presence of cell clusters into the con2508

tused tissue was not related to a stronger functional or anatomic effect. These findings, along with evidence that these transplanted cells induce an early functional improvement, allow to hypothesize that the beneficial effects are owing to stimulation of endogenous neuroprotective processes through a paracrine action. To explore the trophic effect of CB-MSC after injury, we focused on BDNF because accumulating data reveal that BDNF has a critical role in axonal regeneration, synaptic formation, and brain plasticity (44), mechanisms that may contribute to behavioral recovery after TBI (45). Previous studies (46, 47) reported that BM-MSC express mRNA of BDNF and secrete this neurotrophic factor into culture medium. Similarly, in the present study, we observed a remarkable amount of BDNF produced by CB-MSC in culture, suggesting that CBMSC may promote recovery by paracrine secretion of BDNF. Furthermore, in vivo we found that CB-MSC treatment induced a significant increase in BDNF protein expression in CCI mice compared with untreated animals at 3 days postinjury. These findings support the paradigm that growth factors locally delivered by stem cells and/or synthesized by resident brain cells in response to CB-MSC could act in concert to mitigate brain injury and hasten repair. A further relevant finding of our work is the evidence of an involvement of M/M in the protective effect driven by CB-MSC. These cells hold a prominent role in tissue surveillance and response to altered central nervous system conditions. Depend-

ing on the signal participating at their activation, M/M cells exert a double, antithetic function. On one hand, they could behave as phagocytic cells, polishing tissue from cellular debris and activating apoptotic pathways (48); on the other, they can switch toward protective functions (49, 50). In the present study, we examined whether the protective action of CB-MSC occurs through M/M cell activation and is associated with a reduction of phagocytic activity. We have previously shown that infusion of neurosphere-derived stem cells in ischemic mice trigger an important activation of M/M and that this activation is required for stem cell-mediated protective action. We now show a similar M/M functional activation in a different model of acute brain injury and using stem cells from a different source. Seven days after CCI and CB-MSC infusion, we observed a selective rise in CD11b-positive cells when compared with CCI PBS mice. In addition, CB-MSC administration significantly reduced phagocytosis in the pericontusional tissue when compared with CCI PBS mice. Confocal analysis revealed that the high number of ramified CD11bpositive cells in CCI CB-MSC mice were associated with low levels of CD68, whereas globular CD11b cells in CCI PBS mice were prevalently highly positive to CD68. Thus, our data indicate that the protective effect of infused stem cells after CCI is associated with a decreased CD68/CD11b ratio and suggest that two different M/M populations are acting at the site of injury in the two experimental conditions. The decreased CD68/CD11b ratio also indicates that the presence of CB-MSC does not induce any phagocytic activity. Increasing evidence indicates that in the subacute, chronic phase after injury, recovery necessitates axonal regrowth and reconnection, events that are both inhibited by the scar tissue (51). Recent studies show that stem cell infusion is able to dampen gliotic scar formation after cortical injury (52, 53) and brain ischemia (54). In the present study, we demonstrated a delayed structural protective effect that included the attenuation of CCI-related contusion volume and inhibition of glial scar 35 days after CB-MSC infusion. CB-MSC-treated mice showed a significant decrease of GFAP-positive stained area around the cortical contusion indicating that infused cells restrict the transformation of host astrocytes to reactive astrocytes and, therefore, limit glial Crit Care Med 2011 Vol. 39, No. 11

scar formation around the traumatized area. This finding suggests that CB-MSC can render the injured tissue permissive to regeneration and repair, thus fostering recovery.

10.

11.

CONCLUSIONS CB-MSC, administered 24 hrs after TBI, effectively reduced brain damage in an in vivo animal model of CCI brain injury in mice. Namely, CB-MSC attenuated sensorimotor activity impairments starting as early as 1 wk after injury and lasting up to 35 days post-CCI. At this time point, mice receiving cells exhibited also cognitive improvement and anatomic damage reduction. Multiple mechanisms are involved in the reported longlasting protection induced by CB-MSC, including trophic events, M/M phenotypical switch, and glial scar inhibitory effects that remodel the brain and lead to significant improvement of neurologic outcome. Based on these results, we propose that CB-MSC may be regarded as a novel therapeutic strategy for TBI.

12.

13.

14.

15.

16.

REFERENCES 17. 1. Lingsma HF, Roozenbeek B, Steyerberg EW, et al: Early prognosis in traumatic brain injury: From prophecies to predictions. Lancet Neurol 2010; 9:543–554 2. Xiong Y, Mahmood A, Chopp M: Neurorestorative treatments for traumatic brain injury. Discov Med 2010; 10:434 – 442 3. Xiong Y, Mahmood A, Meng Y, et al: Delayed administration of erythropoietin reducing hippocampal cell loss, enhancing angiogenesis and neurogenesis, and improving functional outcome following traumatic brain injury in rats: Comparison of treatment with single and triple dose. J Neurosurg 2010; 113:598 – 608 4. Loane DJ, Faden AI: Neuroprotection for traumatic brain injury: Translational challenges and emerging therapeutic strategies. Trends Pharmacol Sci 2010; 31:596 – 604 5. Faden AI, Stoica B: Neuroprotection: Challenges and opportunities. Arch Neurol 2007; 64:794 – 800 6. Maas AI, Stocchetti N, Bullock R: Moderate and severe traumatic brain injury in adults. Lancet Neurol 2008; 7:728 –741 7. Salem HK, Thiemermann C: Mesenchymal stromal cells: Current understanding and clinical status. Stem Cells 2010; 28:585–596 8. Walker PA, Shah SK, Harting MT, et al: Progenitor cell therapies for traumatic brain injury: Barriers and opportunities in translation. Dis Model Mech 2009; 2:23–38 9. Valle-Prieto A, Conget PA: Human mesenchymal stem cells efficiently manage oxida-

Crit Care Med 2011 Vol. 39, No. 11

18.

19.

20.

21.

22.

23.

24.

tive stress. Stem Cells Dev 2010; 19: 1885–1893 Munoz-Elias G, Woodbury D, Black IB: Marrow stromal cells, mitosis, and neuronal differentiation: Stem cell and precursor functions. Stem Cells 2003; 21:437– 448 Chen J, Li Y, Chopp M: Intracerebral transplantation of bone marrow with BDNF after MCAo in rat. Neuropharmacology 2000; 39: 711–716 Yano S, Kuroda S, Shichinohe H, et al: Do bone marrow stromal cells proliferate after transplantation into mice cerebral infarct? A double labeling study. Brain Res 2005; 1065: 60 – 67 Koh SH, Kim KS, Choi MR, et al: Implantation of human umbilical cord-derived mesenchymal stem cells as a neuroprotective therapy for ischemic stroke in rats. Brain Res 2008; 1229:233–248 Morigi M, Rota C, Montemurro T, et al: Life-sparing effect of human cord blood– mesenchymal stem cells in experimental acute kidney injury. Stem Cells 2010; 28: 513–522 Panepucci RA, Siufi JL, Silva WA Jr, et al: Comparison of gene expression of umbilical cord vein and bone marrow-derived mesenchymal stem cells. Stem Cells 2004; 22: 1263–1278 Jager M, Zilkens C, Bittersohl B, et al: Cord blood—An alternative source for bone regeneration. Stem Cell Rev 2009; 5:266 –277 Liao W, Zhong J, Yu J, et al: Therapeutic benefit of human umbilical cord derived mesenchymal stromal cells in intracerebral hemorrhage rat: Implications of anti-inflammation and angiogenesis. Cell Physiol Biochem 2009; 24:307–316 Mahmood A, Lu D, Yi L, et al: Intracranial bone marrow transplantation after traumatic brain injury improving functional outcome in adult rats. J Neurosurg 2001; 94:589 –595 Mahmood A, Lu D, Lu M, et al: Treatment of traumatic brain injury in adult rats with intravenous administration of human bone marrow stromal cells. Neurosurgery 2003; 53:697–702; discussion 702–793 Mahmood A, Lu D, Qu C, et al: Human marrow stromal cell treatment provides longlasting benefit after traumatic brain injury in rats. Neurosurgery 2005; 57:1026 –1031; discussion 1026 –1031 Kim HJ, Lee JH, Kim SH: Therapeutic effects of human mesenchymal stem cells on traumatic brain injury in rats: Secretion of neurotrophic factors and inhibition of apoptosis. J Neurotrauma 2010; 27:131–138 Gesuete R, Storini C, Fantin A, et al: Recombinant C1 inhibitor in brain ischemic injury. Ann Neurol 2009; 66:332–342 Longhi L, Perego C, Ortolano F, et al: C1inhibitor attenuates neurobehavioral deficits and reduces contusion volume after controlled cortical impact brain injury in mice. Crit Care Med 2009; 37:659 – 665 Ortolano F, Colombo A, Zanier ER, et al: c-Jun N-terminal kinase pathway activation

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

in human and experimental cerebral contusion. J Neuropathol Exp Neurol 2009; 68: 964 –971 Longhi L, Gesuete R, Perego C, et al: Longlasting protection in brain trauma by endotoxin preconditioning. J Cereb Blood Flow Metab 2011 Apr 6 [Epub ahead of print] Fox GB, Fan L, Levasseur RA, et al: Sustained sensory/motor and cognitive deficits with neuronal apoptosis following controlled cortical impact brain injury in the mouse. J Neurotrauma 1998; 15:599 – 614 Fujimoto ST, Longhi L, Saatman KE, et al: Motor and cognitive function evaluation following experimental traumatic brain injury. Neurosci Biobehav Rev 2004; 28:365–378 Smith DH, Okiyama K, Thomas MJ, et al: Evaluation of memory dysfunction following experimental brain injury using the Morris water maze. J Neurotrauma 1991; 8:259 –269 Paxinos G, Franklin K: The Mouse Brain in Stereotaxic Coordinates. New York, Academic Press, 2008 Capone C, Frigerio S, Fumagalli S, et al: Neurosphere-derived cells exert a neuroprotective action by changing the ischemic microenvironment. PLoS ONE 2007; 2:e373 Storini C, Bergamaschini L, Gesuete R, et al: Selective inhibition of plasma kallikrein protects brain from reperfusion injury. J Pharmacol Exp Ther 2006; 318:849 – 854 Donnelly DJ, Gensel JC, Ankeny DP, et al: An efficient and reproducible method for quantifying macrophages in different experimental models of central nervous system pathology. J Neurosci Methods 2009; 181:36 – 44 Broxmeyer HE: Biology of cord blood cells and future prospects for enhanced clinical benefit. Cytotherapy 2005; 7:209 –218 Reimann V, Creutzig U, Kogler G: Stem cells derived from cord blood in transplantation and regenerative medicine. Dtsch Arztebl Int 2009; 106:831– 836 Lu D, Sanberg PR, Mahmood A, et al: Intravenous administration of human umbilical cord blood reduces neurological deficit in the rat after traumatic brain injury. Cell Transplant 2002; 11:275–281 Vigneron A, Vousden KH: p53, ROS and senescence in the control of aging. Aging (Albany NY) 2010; 2:471– 474 Cheng H, Qiu L, Ma J, et al: Replicative senescence of human bone marrow and umbilical cord derived mesenchymal stem cells and their differentiation to adipocytes and osteoblasts. Mol Biol Rep 2010 Dec 29 [Epub ahead of print] Ksiazek K: A comprehensive review on mesenchymal stem cell growth and senescence. Rejuvenation Res 2009; 12:105–116 Bratton SL, Chestnut RM, Ghajar J, et al: Guidelines for the management of severe traumatic brain injury. VII. Intracranial pressure monitoring technology. J Neurotrauma 2007; 24(Suppl 1):S45–54 Stocchetti N, Longhi L, Zanier ER: Intracranial pressure monitoring for traumatic brain injury: Available evidence and clinical impli-

2509

41.

42.

43.

44.

45.

cations. Minerva Anestesiol 2008; 74: 197–203 Harting MT, Jimenez F, Xue H, et al: Intravenous mesenchymal stem cell therapy for traumatic brain injury. J Neurosurg 2009; 110:1189 –1197 Belmadani A, Tran PB, Ren D, et al: Chemokines regulate the migration of neural progenitors to sites of neuroinflammation. J Neurosci 2006; 26:3182–3191 Son BR, Marquez-Curtis LA, Kucia M, et al: Migration of bone marrow and cord blood mesenchymal stem cells in vitro is regulated by stromal-derived factor-1-CXCR4 and hepatocyte growth factor-c-met axes and involves matrix metalloproteinases. Stem Cells 2006; 24:1254 –1264 Brock JH, Rosenzweig ES, Blesch A, et al: Local and remote growth factor effects after primate spinal cord injury. J Neurosci 2010; 30:9728 –9737 Griesbach GS, Sutton RL, Hovda DA, et al: Controlled contusion injury alters molecular

2510

46.

47.

48.

49.

50.

systems associated with cognitive performance. J Neurosci Res 2009; 87:795– 805 Yamaguchi S, Kuroda S, Kobayashi H, et al: The effects of neuronal induction on gene expression profile in bone marrow stromal cells (BMSC)—A preliminary study using microarray analysis. Brain Res 2006; 1087: 15–27 Gu W, Zhang F, Xue Q, et al: Transplantation of bone marrow mesenchymal stem cells reduces lesion volume and induces axonal regrowth of injured spinal cord. Neuropathology 2010; 30:205–217 Block ML, Zecca L, Hong JS: Microgliamediated neurotoxicity: Uncovering the molecular mechanisms. Nat Rev Neurosci 2007; 8:57– 69 Davalos D, Grutzendler J, Yang G, et al: ATP mediates rapid microglial response to local brain injury in vivo. Nat Neurosci 2005; 8:752–758 Lambertsen KL, Clausen BH, Babcock AA, et al: Microglia protect neurons against isch-

51.

52.

53.

54.

emia by synthesis of tumor necrosis factor. J Neurosci 2009; 29:1319 –1330 Wanner IB, Deik A, Torres M, et al: A new in vitro model of the glial scar inhibits axon growth. Glia 2008; 56:1691–1709 Makri G, Lavdas AA, Katsimpardi L, et al: Transplantation of embryonic neural stem/ precursor cells overexpressing BM88/Cend1 enhances the generation of neuronal cells in the injured mouse cortex. Stem Cells 2010; 28:127–139 Heile AM, Wallrapp C, Klinge PM, et al: Cerebral transplantation of encapsulated mesenchymal stem cells improves cellular pathology after experimental traumatic brain injury. Neurosci Lett 2009; 463: 176 –181 Bacigaluppi M, Pluchino S, Peruzzotti Jametti L, et al: Delayed post-ischaemic neuroprotection following systemic neural stem cell transplantation involves multiple mechanisms. Brain 2009; 132:2239 –2251

Crit Care Med 2011 Vol. 39, No. 11