Intranasal Stem Cell Treatment as a Novel Therapy for Subarachnoid Hemorrhage

Abstract

Subarachnoid hemorrhage (SAH) represents a major health problem in Western society due to high mortality and morbidity, and the relative young age of patients. Currently, efficacious therapeutic options are very limited. Mesenchymal stem cell (MSC) administration has been shown to improve functional outcome and lesion size in experimental models of stroke and neonatal hypoxic-ischemic brain injury. Here, we studied the therapeutic potential of intranasally administered bone marrow-derived MSCs relatively late postinsult using a rat endovascular puncture model for SAH. Six days after induction of SAH, rats were treated with MSCs or vehicle through nasal administration. Intranasal MSC treatment significantly improved sensorimotor and mechanosensory function at 21 days after SAH. Gray and white matter loss was significantly reduced by MSC treatment and the number of NeuN⁺ neurons around the lesion increased due to MSC treatment. Moreover, intranasal MSC administration led to a sharp decrease in SAH-induced activation of astrocytes and microglia/macrophages in the lesioned hemisphere, especially of M2-like (CD206⁺) microglia/macrophages. Interestingly, MSC administration also decreased SAH-induced depression-like behavior in association with a restoration of tyrosine hydroxylase expression in the substantia nigra and striatum. We show here for the first time that intranasal MSC administration reverses the devastating consequences of SAH, including regeneration of the cerebral lesion, functional recovery, and treatment of comorbid depression-like behavior.

Introduction

The mortality rate of subarachnoid hemorrhage (SAH) in humans is approximately 40%–50% and surviving patients are confronted with a great loss in quality of life [1,2]. During SAH, blood flows into the subarachnoid space of the brain, which induces several physiological changes including an increase in cerebral pressure, a decrease in oxygen tension, and an influx of peripheral immune cells, all of which contribute to cerebral brain damage. After the initial injury, which is caused by transient cerebral ischemia and direct effects of presence of blood in the subarachnoid space, secondary brain injury can develop due to delayed cerebral ischemia (DCI). DCI is considered to be a major contributor to increased mortality and morbidity after SAH and can be evoked through multiple pathophysiological mechanisms such as formation of cerebral edema, vasospasm, cortical spreading depolarization, and microthromboemboli [3 –5].

The long-term outcome of SAH includes motor, speech, and cognitive dysfunction [6]. Furthermore, there is evidence for sensory symptoms in the subgroup of convexity SAH patients [7].

Moreover, depression is reported by one-third of the patients even 12 months after the insult [8]. The limited treatment options available at present are mainly aimed at preventing vasospasm to reduce DCI. However, the efficacy of these therapies is currently debated [5,9 –11]. In view of the great need for novel efficacious therapies for SAH, also with a longer therapeutic time window, we studied the potential of mesenchymal stem cells (MSCs) as a regenerative strategy in the endovascular puncture rodent model of SAH.

Several studies using models of ischemic brain injury, such as stroke and neonatal hypoxia–ischemia (HI), have shown that MSC administration improves functional neurological recovery and decreases brain lesion volume [12 –16]. MSC treatment up to even 10 days after HI in neonatal mice has significant beneficial effects on functional outcome and brain lesion size [12,17]. Furthermore, our recent data in the neonatal HI model clearly indicate that transplanted MSCs stimulate endogenous repair by stimulation of the differentiation of neural progenitors into adult neurons [18 –20].

In several experimental studies, MSCs have been administered either intracranially or systemically. Intravenous or intra-arterial injection of MSCs, however, results in a high accumulation of the cells in peripheral organs where they are not needed [21]. Moreover, when administered systemically, a higher number of MSCs is needed to obtain the same functional effect as we found with intranasal administration. For clinical application, the intracranial route is too invasive [22].

To target the brain, stem cells have also been applied through the nasal route [17,22 –26]. We previously demonstrated that in neonatal mice with HI brain injury, intranasal treatment with MSCs significantly decreased brain damage and improved functional outcome [17]. MSCs were capable of migrating selectively from the cribriform plate toward the ipsilateral cerebral lesion site in 2–24 h. Furthermore, recent data show that xenogeneic or allogeneic transplantation (with human or allogeneic murine MSCs, respectively) has similar regenerative capacity as mouse MSCs in a neonatal HI mouse model [22].

The goal of this study was to determine whether intranasal administration of MSCs at 6 days after experimental SAH reduces cerebral lesion volume and neuroinflammation, and improves functional outcome measured at 21 days after induction of SAH.

Materials and Methods

Animals and experimental procedure SAH

Experiments were performed in accordance with Dutch regulations, the European international guidelines (Directive 86/609, ETS 123, Annex II), and approved by the Experimental Animal Committee of the University Medical Center Utrecht (license Nos. 2012.I.04.057 and 2013.I.02.023). The experiments are reported in compliance with the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines. Adult male Wistar rats (Charles River, Maastricht, The Netherlands) of 300–350 g were housed in groups of four in Eurostandard Type IV cages (Tecniplast, Milan, Italy) with chip bedding and rat polycarbonate retreat (Bio-Serv/Plexx, Elst, The Netherlands) on reversed light/dark cycle (12 h light cycle) with ad libidum food access. Rats were anesthetized by mechanical ventilation with 2% isoflurane in air/oxygen (2:1). An intramuscular injection of 5 mg/kg gentamicin (Centrafarm, Ettenleur, The Netherlands) was administered to prevent infections. Core temperature was maintained at 37.5°C using a temperature-controlled heating pad. SAH was induced as described previously [27,28]. In short, a sharpened 3.0 Prolene suture was introduced into the right internal cerebral artery (ICA) up to the bifurcation of the ICA and the middle cerebral artery where the Circle of Willis was perforated. Animals received 0.03 mg/kg buprenorphine (Reckitt & Colman, Kingston-Upon-Hill, UK) for pain relief. Sham-operated rats underwent exactly the same procedure and anesthetic ventilation duration, without endovascular puncture. For magnetic resonance imaging (MRI) studies of the brain during the acute phase after SAH in this model, we refer to our earlier work [29,30]. Daily weight loss was monitored and weight loss >10 g per 24 h was compensated by subcutaneous injection of 5 mL Ringer's lactate (Baxter, Utrecht, The Netherlands). Animals were randomly assigned to four different experimental groups: SAH animals treated with either vehicle n = 10 or MSCs n = 13, and sham-operated rats treated with either vehicle n = 13 or MSCs n = 9. Group sizes were determined based on variability in our previous experiments using the same techniques [27]. We could not include some of the rat brains because the quality of the slices in the region of interest was not sufficient. The decision to exclude those samples was made before unblinding the groups.

MSC administration

Bone marrow-derived MSCs of Sprague Dawley rats were purchased from Invitrogen [GIBCO Rat (SD) MSCs; Life Technologies] and cultured according to the manufacturer's instructions. At 6 days post-SAH, MSCs or vehicle [phosphate buffered saline (PBS)] was delivered intranasally in awake animals. Thirty minutes before MSC or vehicle administration, two doses of 6 μL hyaluronidase in PBS (total 100 U; Sigma-Aldrich, Steinheim, Germany) were applied to each nostril and spontaneously inhaled [17,24]. Hyaluronidase degrades hyaluronic acid, thereby increasing tissue permeability, which is necessary for administration of larger particles like cells. Subsequently, a total of 1.5 × 10⁶ MSCs in 24 μL PBS or 24 μL vehicle solution was administered in two doses of 6 μL applied to each nostril. The method of intranasal delivery of cells has been extensively described by Danielyan et al. [23].

Functional outcome

All behavioral experiments were performed in a blinded manner.

Neurological score

Acute neurological deficits to define the severity of the insult were measured before SAH (day 0) and at days 1−−7 post-SAH by an adaptation of the method described by Sugawara et al. [31] and described earlier by our group [27]. The scoring includes measurement of spontaneous activity, spontaneous movements of all limbs, outstretching while held by tail, climbing wall of wire cage, and response to vibrissae touch. The maximum score was 15 in absence of any deficits.

We recently showed that there is a wide variety in neurological score and hemorrhage size in the endovascular puncture model due to the unpredictable degree of bleeding [27]. We showed that SAH animals with a neurological score >12 (“mild SAH”) at day 1 do not develop a clear cerebral lesion nor a deficit in sensorimotor function [27]. To be able to monitor changes in cerebral lesion volume and functional outcome, we, therefore, selected the group of animals with severe neurological disabilities of ≤12 (“severe SAH”).

Sensorimotor function

Sensorimotor function was tested by the adhesive removal task (ART) at 21 days post-SAH. Adhesive labels (Tough-Spots; Diversified Biotech, Boston, MA) were placed on the left and right forepaw, and the latency to removal was recorded. The mean time until complete removal of three stickers per forepaw was recorded. Sticker placement on left and right forepaws was alternated between and within animals. The ART was performed in the dark by a trained observer blinded to treatment.

Mechanosensory function

Mechanosensory function was measured at 21 days post-SAH using von Frey hairs [32]. Rats were placed in individual plastic enclosures on a mesh floor and allowed to acclimate for 30 min. A series of calibrated von Frey filaments (bending force range from 1 to 15 g; Stoelting, Wood Dale, IL) was applied perpendicularly to the plantar surface of the hindpaw with sufficient force to bend the filaments for 6 s, and brisk paw withdrawals, shaking, licking, or flinching were considered as a positive response. In the absence of a response, the filament of next greater force was applied. If a response occurred, the filament of next lower force was applied. The force in grams producing a 50% likelihood of withdrawal (50% withdrawal threshold) was calculated using the up-and-down method as described [33].

Depression-like behavior

To determine depression-like behavior, we used the sucrose preference test [34]. At 21 days post-SAH, animals were solitary housed with access to food, normal tap water, and 1% sucrose water. During the dark period, water and 1% sucrose were weighed every 4 h. The first day was used for acclimation of the animals and these results were not used for data analysis. During the next 2 days, water and 1% sucrose water intake were measured every 4 h. To determine whether changes in sucrose intake respond to the antidepressant effect of ketamine, we administered ketamine (Narketan 10; Vetoquinol S.A., Lure Cedex, France) intraperitoneally after the last measurement on day 3 at a dose of 10 mg/kg [35]. After administration of ketamine, sucrose preference was measured for 1 more day (on day 4). The sucrose preference test was performed in the dark by an observer blinded to the treatment.

Histology

Rats were euthanized at 21 days post-SAH by pentobarbital overdose and perfused with PBS followed by 4% paraformaldehyde. Brains were postfixed, embedded in paraffin, and coronal sections of 8 μm were cut at bregma. Deparaffinized sections were incubated with mouse-antimicrotubule-associated protein 2 (MAP2; 1:1,000; Sigma-Aldrich), mouse-antimyelin basic protein (MBP; 1:1,600; Sternberger Monoclonals, Lutherville, MD), or rabbit antityrosine hydroxylase (TH; 1:1,000; Abcam, Cambridge, UK) followed by biotin-labeled horse-antimouse or goat-antirabbit antibodies (Vector Laboratories, Burlingame, CA). Staining was revealed using Vectastain ABC kit (Vector Laboratories) and diaminobenzamidine with an addition of ammonium nickel sulfate for the TH staining. The MAP2 and MBP data were analyzed by determining the ipsi-/contralateral positive area × 100% using Adobe Photoshop CS6 and ImageJ software (http://imagej.nih.gov/ij/index.html), respectively. TH-positive pixels were analyzed using ImageJ software (http://imagej.nih.gov/ij/index.html) with threshold and histogram plugins. By using this software, background staining was subtracted. The same threshold was used for all animals. For the substantia nigra, we captured images of three microscopic fields on one slide for each animal in the same area of the substantia nigra, both ipsilateral and contralateral, for all slides and quantified the percentage area staining positively for TH. We did not observe overt tissue damage in this area (no loss of MAP2 staining, no cysts, and no apparent gliosis). In addition, we captured images of the entire striatum in each hemisphere on one slide per animal and assessed the number of positive pixels normalized for striatal area. In the striatum, we did observe clear lesions in at least some of the areas and, therefore, we normalized for total striatal area.

For immunofluorescent stainings, sections were incubated with mouse-anti-NeuN (1:100; Chemicon), mouse-antiglial fibrillary acidic protein (GFAP; 1:200; Cymbus Biotechnology, Southampton, UK), rabbit-anti-ionized calcium binding adaptor molecule 1 (Iba-1; 1:500; Wako), rabbit-anti-CD86 (1:200; AbD Serotec, Oxford, UK), or goat-anti-CD206 (1:200; R&D Systems, Minneapolis, MN), followed by incubation with secondary antibodies AlexaFluor-488 goat-antimouse IgG, AlexaFluor-594 goat-antirabbit IgG, AlexaFluor-488 donkey-antigoat IgG, or AlexaFluor-488 goat-antirabbit IgG (all Molecular Probes, Eugene, OR). Sections were counterstained with DAPI and were examined with an Axio Observer Microscope with Axiovision Rel 4.6 software (Carl Zeiss, Sliedrecht, The Netherlands). For quantifying the number NeuN⁺ cells surrounding the lesion site, NeuN⁺ cells were manually counted in two microscopic fields surrounding the lesion site or, if no lesion was observed, at similar locations in the cortex. Iba-1 and GFAP were analyzed by determining the percentage Iba-1- or GFAP-positive expression per whole hemisphere per animal. CD86 and CD206 expression levels were assessed by measuring the number of CD86- or CD206- and DAPI-positive cells per mm². CD86 and CD206 cell counts were assessed in three microscopic fields in intact cortical areas surrounding the infarct area both ipsilateral and contralateral so six microscopic fields per animal. For all analyses, software of ImageJ was used (http://imagej.nih.gov/ij/index.html). All analyses were performed in a blinded manner.

Statistical analysis

Data were analyzed by one- or two-way analysis of variance with Bonferroni post hoc tests. Data are presented as mean ± standard error of the mean. Gray and white matter damage (MAP2, MBP stainings), sensorimotor behavior, and neuroinflammation (Iba-1, GFAP stainings) were measured as primary outcome parameters. Other markers were assessed as secondary outcome parameters.

Results

Acute neurological score after SAH

A total of 105 rats were subjected to the endovascular puncture procedure (or sham operation) and severity of the insult was determined by scoring neurological deficits from days 1 to 7 after the surgical procedure using a neurological scoring system of 1–15. A score of 15 represents “no deficits” [27,31]. At day 6 after surgery, the neurological score of all surviving SAH animals had returned to levels of sham-control animals (Fig. 1A). In this study, only SAH animals with a neurological score of ≤12 at day 1 (in total n = 46), which were still alive on day 6 (n = 23), were used, as we have shown before that these animals (designated as “severe SAH”) develop a cerebral lesion and functional deficits [27]. The mortality rate after severe SAH was 50% (23 out of 46 animals in this group survived) and all mortality occurred within the first 6 days, before we started the MSC administration (Fig. 1B). The neurological scores at day 1 of SAH animals that received MSC administration (n = 13) were similar to those of animals that received vehicle treatment (n = 10). None of the sham-operated animals (n = 22) died (Fig. 1B). “Severe SAH” rats showed significant weight loss in comparison with “mild SAH” (a total of n = 37 that were not followed-up) and sham-control rats at 1–6 days post-SAH (Fig. 1C).

FIG. 1.

Neurological scores, survival, and weight loss after SAH. (A) Acute neurological scores from days 1 to 7 after SAH are displayed for all rats that survived for 21 days [n = 22 sham (green circles), n = 23 mild SAH (black squares), n = 23 severe SAH (red triangles)]. Stratification in mild or severe SAH was based on the neurological score at day 1 using neurological score >12 as “mild SAH” and neurological score ≤12 as “severe SAH.” (B) Survival curve from days 1 to 21 after induction of SAH. Survival rate is shown for all animals used in this study. Sham n = 22 (green), mild SAH n = 37 (black), severe SAH n = 46 (red). (C) Weight loss from days 1 to 6 post-SAH compared with weight before SAH at day 0. Sham n = 22 (green circles), mild SAH n = 23 (black squares), severe SAH n = 23 (red triangles). *P < 0.05, **P < 0.01, ***P < 0.001 versus sham. Data represent mean ± SEM. SAH, subarachnoid hemorrhage; SEM, standard error of the mean. Color images available online at www.liebertpub.com/scd

Intranasal treatment with MSCs and sensorimotor behavior

Animals with “severe SAH” and sham-control animals were treated intranasally with 1.5 × 10⁶ MSCs (sham-control: n = 9; SAH: n = 13) or vehicle (sham: n = 13; SAH: n = 10). To investigate whether treatment with MSCs improved sensorimotor function, we subjected all rats to the ART at 21 days post-SAH. During the ART, the time is measured that the animal needs to remove a tiny adhesive patch from its forepaw (“total removal time”). The “total removal time” can be split in the latency to start removal of the adhesive patch (“latency to start removal”) as a measure of sensory function, and the time it takes the animal to remove the patch after sensing it (“effective adhesive removal time”) as a measure of fine motoric function [36].

Vehicle-treated SAH animals (n = 10) showed a significant increase in “total removal time” of the adhesive patch from their left (impaired) forepaw compared with their right (unimpaired) forepaw or compared with vehicle-treated sham-control rats (n = 13; Fig. 2A). In sham-control rats, the “total removal time” of the adhesive patch between left and right forepaw did not differ (Fig. 2A). In addition, “latency to start removal” and “effective adhesive removal time” were both increased in the left (impaired) forepaw in vehicle-treated SAH rats (Fig. 2B, C).

FIG. 2.

Long-term sensorimotor and mechanosensory behavior deficits after SAH are attenuated after MSC treatment. (A–C) Sensorimotor function was tested using the ART at 21 days post-SAH. (A) The total removal time for both left and right forepaws is presented in seconds for all experimental groups. (B) The latency to start removal is presented in seconds. (C) The effective adhesive removal time is presented in seconds. (D) Sensitivity to mechanical stimulation was determined using the von Frey test at 21 days post-SAH. Data are expressed as 50% withdrawal threshold in grams. **P < 0.01 ***P < 0.001 versus sham; ^### P < 0.001 SAH+vehicle versus SAH+MSC. Data represent mean ± SEM. Sham n = 13, sham+MSC n = 9, SAH+vehicle n = 10, SAH+MSC n = 13. ART, adhesive removal task; MSC, mesenchymal stem cell.

MSC treatment at 6 days post-SAH significantly decreased “total removal time” of the left (impaired) forepaw (n = 13; Fig. 2A), “latency to start removal” (Fig. 2B), and “effective adhesive removal time” (Fig. 2C) compared with vehicle treatment. Sham-operated animals treated with MSCs (n = 9) showed similar removal times in the ART as vehicle-treated sham-operated animals (Fig. 2A–C).

In the same group of rats, we measured mechanosensory function after SAH by testing the sensitivity to touch of the hind paws using von Frey hairs [32,33]. This test uses nylon filaments of various strengths to apply different but accurate forces to the hind paw to determine the mechanical threshold to touch. The sensitivity to touch was significantly decreased in vehicle-treated SAH animals compared with sham-operated animals at 21 days after SAH. Administration of MSCs at 6 days after SAH significantly increased the sensitivity to touch represented by a decrease in the 50% threshold, indicating an improved mechanosensory response after MSC treatment in SAH animals (Fig. 2D).

Brain lesion size after SAH and MSC treatment

To determine whether treatment with MSCs at 6 days after SAH improves structural brain damage, we assessed gray and white matter loss by MAP2 and MBP staining, respectively, at 21 days after injury.

After SAH, rats showed an extensive brain lesion represented by a large decrease in ipsilateral MAP2 expression, mainly in the cortex (sham-control vehicle: n = 13; sham-control MSC: n = 9; SAH vehicle: n = 8; SAH MSC: n = 11; Fig. 3A, B). Treatment with MSCs at 6 days after SAH resulted in significantly less ipsilateral MAP2 loss compared with vehicle treatment at 21 days after SAH (Fig. 3A, B). MSC treatment in sham-operated animals did not have any effect on MAP2 expression compared with vehicle treatment in sham-operated animals (Fig. 3B). To assess neuronal repopulation around the SAH-induced lesion, the density of NeuN⁺ cells in the lesion area was quantified. Figure 3C and D shows that the number of NeuN⁺ neurons surrounding the lesion site was reduced in vehicle-treated SAH animals in comparison with sham animals. Importantly, MSC treatment significantly increased the number of NeuN⁺ cells in the area surrounding the lesion site (sham-control vehicle: n = 6; SAH vehicle: n = 5; SAH MSC: n = 7; Fig. 3C, D).

FIG. 3.

MSC treatment reduces gray and white matter loss and increases the number of neuronal cells populating the lesion site after SAH. (A) Effect of MSC treatment on gray matter damage at 21 days after SAH determined by neuronal-specific MAP2 staining. Photographs show representable pictures of MAP2 staining in sham-operated (top), vehicle-treated SAH (middle), and MSC-treated SAH animals (bottom). (B) Bar graph showing ipsi-/contralateral MAP2-positive area × 100%. (C) Effect of MSCs on number of NeuN⁺ neurons (green) surrounding the lesion site (L) at 21 days post-SAH. Nuclei are stained with DAPI (blue). Scale bar: 100 μm. (D) The SAH-induced decrease in density of NeuN⁺ cells surrounding the lesion is partly restored after MSC treatment. (E) Effect of MSC treatment on white matter damage at 21 days after SAH determined by myelin-specific MBP staining. Photographs show representable pictures of MBP staining in sham-operated (top), vehicle-treated SAH (middle), and MSC-treated SAH animals (bottom). (F) Bar graph showing ipsi-/contralateral MBP-positive area × 100%. ***P < 0.001, ****P < 0.0001 sham versus SAH+vehicle, ^# P < 0.05 and ^## P < 0.01 SAH+vehicle versus SAH+MSC. (A, B, E, F) Sham+vehicle n = 13; sham+MSC n = 9; SAH+vehicle n = 8; SAH+MSC n = 11. (C, D) Sham+vehicle n = 6; SAH+vehicle n = 5; SAH+MSC n = 7. Data represent mean ± SEM. MAP2, microtubule-associated protein 2; MBP, myelin basic protein. Color images available online at www.liebertpub.com/scd

SAH also induced white matter loss, represented by a significant decrease in MBP staining in the ipsilateral cortex (Fig. 3E, F). MSC treatment reduced ipsilateral white matter loss as shown by an increased ratio of ipsilateral/contralateral MBP expression compared with vehicle treatment after SAH (sham-control vehicle: n = 13; sham-control MSC: n = 9; SAH vehicle: n = 8; SAH MSC: n = 11; Fig. 3E, F). MSC treatment in sham-operated animals did not affect MBP expression (Fig. 3F).

Because we did not observe any effects of MSC treatment in sham-control animals on behavioral outcome and/or gray and white matter damage, we did not include this experimental group in the further analyses.

Cerebral inflammation and MSC treatment

We recently showed that SAH induces a neuroinflammatory response in the brain in rats that persists until at least 21 days post-SAH [27]. To evaluate the effect of MSC treatment on this inflammatory response, expression of GFAP in the ipsilateral hemisphere as a marker for astrocyte activation was determined. At 21 days post-SAH, GFAP expression was increased in vehicle-treated SAH animals compared with sham-operated animals, particularly at the border of the lesion (sham-control vehicle: n = 6; SAH vehicle: n = 6; SAH MSC: n = 8; Fig. 4A, B). MSC treatment after SAH significantly reduced GFAP expression at the lesion border compared with vehicle treatment (Fig. 4A, B).

FIG. 4.

Long-term macrophage/microglia and astrocyte activation after SAH is reduced after MSC administration. (A) Immunofluorescent photographs showing macrophage/microglia and astrocyte activation by Iba-1 staining (red) and GFAP staining (green) respectively, in the ipsilateral hemisphere of five representative sham-operated, vehicle-treated SAH and MSC-treated SAH animals. Photographs are taken at 21 days post-SAH. Insets show a higher magnification; scale bar in inset represents 300 μm. (B) Effect of MSC treatment on GFAP expression at 21 days after SAH. Bar graph shows percentage of GFAP expression in the ipsilateral hemisphere. (C) Effect of MSC treatment on Iba-1 expression at 21 days after SAH. Bar graph shows percentage of Iba-1 expression in the ipsilateral hemisphere. Sham n = 6, SAH+vehicle n = 6, SAH+MSC n = 8. *P < 0.05, ***P < 0.001 sham versus SAH+vehicle; ^# P < 0.05 SAH+vehicle versus SAH+MSC. Data represent mean ± SEM. GFAP, glial fibrillary acidic protein; Iba-1, ionized calcium binding adaptor molecule 1. Color images available online at www.liebertpub.com/scd

Macrophage/microglia activation, measured by Iba-1 expression in the ipsilateral hemisphere, was 2.2-fold increased in vehicle-treated SAH rats compared with brains of sham-operated animals (Fig. 4A). Increased Iba-1 expression after SAH was mainly observed in the cortex and striatum (Fig. 4A). Furthermore, the morphology of Iba-1-expressing macrophages/microglia in the ipsilateral cortex and striatum changed from small, finely branched, quiescent microglia observed in sham-operated animals toward denser, less-branched, amoeboid-like activated microglia in vehicle-treated SAH animals (Fig. 4A, insets). MSC treatment after SAH dramatically decreased ipsilateral Iba-1 expression compared with vehicle treatment (Fig. 4A, C). Moreover, after MSC treatment, the morphology of the Iba-1-positive cells appeared as “resting” macrophages/microglia with typical ramified appearance (Fig. 4A, insets).

Next we evaluated microglia polarization toward a more proinflammatory M1-like state or an anti-inflammatory/regenerative M2-like state. We determined CD86 expression as a proinflammatory M1-like phenotype and CD206 expression as a M2-like phenotype. CD206-positive macrophages/microglia have been shown to exert a debris-scavenging, prohealing function [37]. Compared with sham-control rats, the number of CD86-positive cells was not altered after SAH or after MSC treatment (sham-control vehicle: n = 10; SAH vehicle: n = 8; SAH MSC: n = 12; Fig. 5). In contrast, the number of CD206-positive cells was highly increased at 21 days after SAH (Fig. 5B, D). Interestingly, MSC treatment reduced the number of CD206-positive cells to sham level (Fig. 5B, D).

FIG. 5.

Effect of MSC treatment on polarization of microglia. (A) Representative immunofluorescent photographs of CD86 staining (green) in contra- and ipsilateral hemispheres of sham-operated, vehicle-treated SAH, and MSC-treated SAH animals. Nuclear staining (DAPI) is depicted in blue. (B) Bar graph showing number of CD86-positive cells per mm² in contra- and ipsilateral hemispheres of all experimental groups. (C) Representative immunofluorescent photographs of CD206 staining (green) in contra- and ipsilateral hemispheres of sham-operated, vehicle-treated SAH, and MSC-treated SAH animals. Nuclear staining (DAPI) is depicted in blue. (A, C) Photographs are taken at 21 days post-SAH. Scale bar represents 40 μm. (D) Bar graph showing number of CD206-positive cells per mm² in contra- and ipsilateral hemispheres of all experimental groups. ***P < 0.001 sham-ipsi versus SAH+vehicle-ipsi and ^# P < 0.05 SAH+vehicle-ipsi versus SAH+MSC-ipsi. Data are presented as mean ± SEM. Sham n = 10; SAH+vehicle n = 8; SAH+MSC n = 12. Color images available online at www.liebertpub.com/scd

The effect of MSC treatment on SAH-induced depression-like behavior

One of the long-term consequences of SAH in patients is development of major depression with a prevalence of 33% at 12 months post-SAH [8]. To determine whether depression-like behavior is present in rats after SAH, we performed the sucrose preference test at 21 days post-SAH. This test has been extensively used as a measure of depression-like behavior in rodents and captures lack of motivation and anhedonia [38,39]. Sweetening of water by adding sucrose elicits a natural preference over normal tap water in healthy rodents. Anhedonia has been shown to reduce the preference for sucrose water [40]. In our study, sham-operated animals showed a sucrose preference of 88%, whereas vehicle-treated SAH animals showed no preference for sucrose, indicating anhedonia (sham-control vehicle: n = 13; sham-control MSC: n = 9; SAH vehicle: n = 10; SAH MSC: n = 13; Fig. 6).

FIG. 6.

MSC administration attenuates SAH-induced depression-like behavior. Depression-like (anhedonic) behavior was tested using the sucrose preference test. The intake of normal tap water and sucrose water was measured every 4 h and preference for the sucrose water (%) was calculated. *P < 0.05, **P < 0.01, ***P < 0.001 versus no preference (50%) (dotted line). Data are presented as mean ± SEM. Sham+vehicle n = 13, sham+MSC n = 9, SAH+vehicle n = 10, SAH+MSC n = 13, SAH+ketamine n = 10.

To further define whether anhedonia represents depression-like behavior, rats were treated with the pharmacological antidepressant, ketamine [41,42]. Ketamine has recently been described to have an acute antidepressive effect in rodents as well as in humans when given in a low dose [43]. Figure 6 shows that a single dose of ketamine significantly reversed sucrose preference in vehicle-treated SAH animals.

Importantly, MSC treatment at 6 days after SAH completely restored the sucrose preference of SAH animals to the level of sham-operated animals. MSC treatment did not affect sucrose preference in sham-operated animals (Fig. 6). The total fluid intake did not differ between all treatment groups (data not shown).

Influence of SAH and role of MSCs on the dopaminergic system

Major depression in humans is often associated with a widespread loss of dopaminergic tonus and decreased dopamine synthesis [38]. Moreover, dopaminergic neurons are very sensitive to cerebral injury signals. TH converts tyrosine to l-DOPA, a precursor of dopamine, and a decrease in TH staining can be considered as a decrease in dopamine production.

At 21 days post-SAH, TH staining was decreased in the nerve endings in the striatum (Fig. 7A, B) and in the substantia nigra (Fig. 7C, D) of the ipsilateral hemisphere of vehicle-treated SAH rats compared with sham animals (sham-control vehicle: n = 11; SAH vehicle: n = 9; SAH MSC: n = 12). TH staining in the contralateral substantia nigra in SAH animals was indistinguishable of that in sham animals (Fig. 7C, D). MSC treatment increased ipsilateral TH staining in cell bodies of the substantia nigra and nerve endings in the striatum to levels in sham animals (Fig. 7A–D).

FIG. 7.

MSC administration increases TH expression after SAH. Effect of MSC administration on dopaminergic tonus at 21 days after SAH determined by TH expression in striatum (A, B) and substantia nigra (C, D). (A) Representative photographs of TH staining at the contralateral (contra) and ipsilateral (ipsi) striatal level of sham-operated, vehicle-treated SAH, and MSC-treated SAH animals. Scale bar represents 50 μm. (B) Bar graph showing effect of MSC treatment on TH expression in the contra- and ipsilateral hemisphere at striatal level. (C) Representative photographs of TH staining in the contralateral (contra) and ipsilateral (ipsi) substantia nigra of sham-operated, vehicle-treated SAH, and MSC-treated SAH animals. Scale bar represents 50 μm. (D) Bar graph showing effect of MSC treatment on TH expression in the contra- and ipsilateral hemisphere in the substantia nigra as quantified in the area indicated by the square in the overview images. ^## P < 0.01 SAH+vehicle-contra and -ipsi versus SAH+MSC-contra and -ipsi, *P < 0.05 SAH+vehicle-ipsi versus sham-ipsi and versus SAH+MSC-ipsi. Data represent mean ± SEM. Sham n = 11, SAH+vehicle n = 9, SAH+MSC n = 12. TH, tyrosine hydroxylase.

Discussion

We demonstrated for the first time that intranasal administration of MSCs at 6 days after severe SAH improves sensorimotor function and decreases lesion size in rats. Moreover, neuroinflammation associated with SAH, assessed by astrocyte and microglia/macrophage activation, was significantly reduced by MSC treatment [27]. Finally, intranasal administration of MSCs reversed the depression-like state of rats induced by SAH, which was associated with a restoration of TH expression in the ipsilateral substantia nigra and the striatum.

The advantage of using MSCs over current pharmacological strategies to combat the negative consequences of SAH is the capacity of MSCs to repair already damaged tissue as we also have shown for MSC treatment in our mouse model of hypoxic-ischemic brain injury [17,22]. In that model, the lesion size was reduced by approximately 50%–60% in response to intranasal administration of MSCs. Khalili et al. showed a beneficial effect of intravenous MSC transplantation in the single endogenous blood injection SAH model [44,45]. MSCs were administered at 24 h after SAH in this model and showed an improvement in neurological score from 7.75 to 3.5 at 14 days post-SAH. A comparison between this study and the study performed by Khalili et al. is difficult since different readouts were used and different animal models of SAH were applied using different administration routes of the MSCs [44]. Instead of intravenous administration, we used a noninvasive and easy route of intranasal administration. Although we did not assess infiltration of the MSCs into the brain in this study, recent studies by Donega et al. and van Velthoven et al. showed in a neonatal mouse model of HI that nasal administration of MSCs allows migration of the MSCs specifically to the damaged brain areas [17,22,24], probably through a chemokine-mediated pathway. Interestingly in this model of unilateral brain damage, the nasally administered MSCs accumulate predominantly in the hemisphere in which the damage was present, even though they were administered through both nostrils [17,24]. The effective intranasal delivery of stem cells to the rodent brain has also been shown in several other adult models of neurodegeneration such as Parkinson's disease, stroke, multiple sclerosis, Alzheimer's disease, and brain tumors and is now widely used and accepted [23,25,26,46 –48]. Recent unpublished data showed that besides the migration toward the damaged brain area in small rodents, intranasal administered MSCs are also capable of migrating to the lesioned area in a newborn baboon subjected to carotid occlusion (article in preparation). These data indicate that even when the migration route is considerably longer in the newborn monkey, the MSCs still arrive at the damaged brain area. The latter data are of translational importance when this intranasal route will be used for patients with SAH.

We started our MSC treatment relatively late, that is, at 6 days after the insult, to allow for full establishment of the lesion, including damage caused by possible delayed cerebral ischemia (DCI) which have an onset at days 1–3 after the endovascular puncture. Previous MRI studies by us and others in this SAH model have shown that increased tissue perfusion, leakage of blood–brain barrier, edema formation, and further tissue degeneration, including gliosis and necrosis, lead to a well-established lesion at day 7 after SAH [29,30,49]. We, therefore, suggest that the beneficial effects of MSC transplantation at day 6 after SAH are the result of regenerative repair mechanisms activated by MSCs. This is in line with our earlier findings showing that MSC transplantation alters the gene expression profile and induces new neuronal precursor formation in the subventricular zone (SVZ) in a model of neonatal brain damage [19,50]. We propose that activation of endogenous repair mechanisms underlies the beneficial effects of MSC treatment. Indeed our present data indicate that MSC treatment induces an increase in NeuN⁺ cells surrounding the lesion area when compared with vehicle treatment in the current SAH model (Fig. 3). Even though it remains to be determined what the exact mechanism is that underlies the increase in NeuN⁺ cells, our data point toward promotion of repair rather than a protective effect of MSCs. This is clinically relevant because a regenerative mechanism allows for an extended therapeutic window, which is urgently needed for SAH patients.

We observed unilateral brain damage after SAH in the hemisphere ipsilateral to the endovascular puncture, which is in line with other studies using the endovascular puncture model. Only in cases of very severe bleeding, some studies have shown involvement of the contralateral hemisphere [27,51]. Delayed ischemia, possibly resulting from vasospasms, is an important secondary process that might contribute to unilateral brain damage after experimental SAH [52 –54]. Vasospasms are most prominent near the site of the aneurysm [3], this would be a possible explanation for the unilateral brain damage that is observed.

One of the striking consequences of SAH is persistent abundant microglia activation in the damaged hemisphere as represented by an increase in Iba-1 staining that was even present at 21 days after SAH [27]. However, although the microglia showed a morphology associated with an activated state, we only observed an increase in CD206-positive staining of microglia after SAH and not of the CD86-positive phenotype. CD206-expressing microglia/macrophages have been associated with tissue regeneration and repair [37]. CD206-positive cells typically phagocytose red blood cells or tissue debris, recruit anti-inflammatory T cells, and cause a downregulation of proinflammatory cytokine production [37]. Moreover, CD206 itself is involved in the binding of apoptotic and necrotic cells [55,56]. Therefore, we suggest that the overwhelming presence of CD206-positive microglia in the brains of SAH animals is probably a late response to the presence of erythrocytes, apoptotic cells, and tissue debris.

Jose et al. showed that MSCs reduce microglia proliferation in vitro [57], which is in accordance with the decrease in Iba-1 expression after MSC administration post-SAH as we show in Fig. 4. In this respect, it is of interest that MSC treatment prevented the expression of CD206-positive microglia/macrophages, whereas the expression of CD86-positive microglia/macrophages remained at a similar level as after vehicle treatment. This apparent contradiction of a reduced number of regenerative microglia/macrophages after MSC treatment may be explained by the fact that MSCs when given at day 6 reduce the proinflammatory state of the brain, which prevents further deterioration of brain damage and thereby abolishes the need for a late polarization of microglia to the M2-like state.

The beneficial effects of treatment with MSCs also include an improvement of sensorimotor function. The increased time to sense the presence of the patch in the ART in SAH animals indicates a loss of sensory function. This conclusion is further strengthened by our data in the von Frey test that showed higher threshold to sense light sensory stimuli applied to the paws of SAH rats.

We observed that brain damage after SAH predominantly occurs at the cortical level and spreads, depending on the intensity of insult, from layer I in the S1/M1 cortex to layer VI [27]. The clear reduction in cortical damage after MSC treatment may possibly explain the improvement in sensorimotor behavior after MSC transplantation.

In humans, depression is a frequent symptom after SAH. In a prospective study, Kreiter et al. showed that 38% of SAH patients have depressive symptoms at 3 months after the insult and 33% of patients are still depressed at 12 months postinsult [8]. Moreover, up to 50% of SAH patients encounter cognitive impairments, which is thought to contribute to poor recovery and reduced quality of life after SAH. A reduction in depressive symptoms would most likely improve the quality of life and overall recovery of SAH patients.

In the rodent endovascular puncture model of SAH, we observed depression-like behavior at 21 days post-SAH. Although this might be a relative early time point after SAH, rats lost their preference for sucrose water, and this anhedonic state has been defined as depression-like behavior in multiple animal studies [58,59]. Structural brain damage evoked by SAH may cause a loss of sucrose preference by a damage-induced loss of taste and/or olfactory function. However, a single acute treatment with a low dose of ketamine as an antidepressant drug reversed SAH-induced anhedonia, indicating that the depression-like behavior is not the result of loss of taste/olfactory function.

The mechanisms underlying the depression-like state after SAH and the beneficial effects of MSC treatment remain unclear however. We investigated changes in the dopaminergic system and show a unilateral decrease in TH staining in the substantia nigra and a decrease in TH in nerve endings in the striatum after SAH. These findings indicate a reduction in dopamine production that may contribute to the anhedonia as observed in the sucrose preference test. MSC treatment restored TH staining and normalized sucrose preference, supporting the notion that loss of dopamine output and impaired sucrose preference may be related.

Early release of proinflammatory cytokines possibly as a response to the acute insult may induce depression-like behavior by upregulation of the enzyme indoleamine 2,3-dioxygenase, leading to an increase in quinolinic acid, a neurotoxic NMDA receptor agonist known to induce depression-like behavior [43,60,61].

A decrease in BDNF expression has also been proposed as a mechanism for establishing a depressed state in animals and humans [62]. In the double-hemorrhage model of SAH, Jiang et al. showed an increased ratio between pro-BDNF and mature BDNF expression in the brain, indicating a decreased expression of mature BDNF [63]. Our group has shown previously that MSCs increase their BDNF production when encountering an ischemic brain milieu [19,48]. MSCs likely also increase BDNF levels in the brain damaged by SAH and this is another potential mechanism for suppression of the observed anhedonia.

In conclusion, we demonstrate for the first time that a single intranasal delivery of MSCs has the potential to become an effective therapeutic strategy with a longer therapeutic time window than currently available therapies to combat the devastating effects of SAH. Moreover, intranasal MSC therapy does not only repair the brain structure and improve sensorimotor function after the insult, but also reverses the depression-like state induced by SAH.

Footnotes

Acknowledgments

This study was funded, in part, by the “Dirkzwager-Assink Foundation” and by the “Friends of UMC Utrecht fund.” The authors thank S. Versteeg for excellent technical assistance with the von Frey test.

Author Disclosure Statement

No competing financial interests exist for any of the authors.

References

The ACROSS group. (2000). Epidemiology of aneurysmal subarachnoid hemorrhage in Australia and New Zealand: incidence and case fatality from the Australasian Cooperative Research on Subarachnoid Hemorrhage Study (ACROSS). Stroke, 31:1843–1850.

van Gijn

, Kerr

and Rinkel

. (2007). Subarachnoid haemorrhage. Lancet, 369:306–318.

Al-Tamimi

, Orsi

, Quinn

, Homer-Vanniasinkam

and Ross

. (2010). A review of delayed ischemic neurologic deficit following aneurysmal subarachnoid hemorrhage: historical overview, current treatment, and pathophysiology. World Neurosurg, 73:654–667.

Laskowitz

and Kolls

. (2010). Neuroprotection in subarachnoid hemorrhage. Stroke, 41:S79–S84.

Macdonald

. (2014). Delayed neurological deterioration after subarachnoid haemorrhage. Nat Rev Neurol, 10:44–58.

Rinkel

and Algra

. (2011). Long-term outcomes of patients with aneurysmal subarachnoid haemorrhage. Lancet Neurol, 10:349–356.

Beitzke

, Gattringer

, Enzinger

, Wagner

, Niederkorn

and Fazekas

. (2011). Clinical presentation, etiology, and long-term prognosis in patients with nontraumatic convexal subarachnoid hemorrhage. Stroke, 42:3055–3060.

Kreiter

, Rosengart

, Claassen

, Fitzsimmons

, Peery

, Du

, Connolly

and Mayer

. (2013). Depressed mood and quality of life after subarachnoid hemorrhage. J Neurol Sci, 335:64–71.

Dankbaar

, Slooter

, Rinkel

and Schaaf

. (2010). Effect of different components of triple-H therapy on cerebral perfusion in patients with aneurysmal subarachnoid haemorrhage: a systematic review. Crit Care, 14:R23.

10.

, Pan

, Wang

, Rong

, Yin

, Milgrom

, Shi

, Tang

and Peng

. (2013). Clipping versus coiling for ruptured intracranial aneurysms: a systematic review and meta-analysis. Stroke, 44:29–37.

11.

Priebe

. (2007). Aneurysmal subarachnoid haemorrhage and the anaesthetist. Br J Anaesth, 99:102–118.

12.

Donega

, van Velthoven

CTJ

, Nijboer

, Kavelaars

and Heijnen

. (2013). The endogenous regenerative capacity of the damaged newborn brain: boosting neurogenesis with mesenchymal stem cell treatment. J Cereb Blood Flow Metab, 33:625–634.

13.

Kocsis

and Honmou

. (2012). Bone marrow stem cells in experimental stroke. Prog Brain Res, 201:79–98.

14.

Onda

, Honmou

, Harada

, Houkin

, Hamada

and Kocsis

. (2007). Therapeutic benefits by human mesenchymal stem cells (hMSCs) and Ang-1 gene-modified hMSCs after cerebral ischemia. J Cereb Blood Flow Metab, 28:329–340.

15.

Scheibe

, Ladhoff

, Huck

, Grohmann

, Blazej

, Oersal

, Baeva

, Seifert

and Priller

. (2012). Immune effects of mesenchymal stromal cells in experimental stroke. J Cereb Blood Flow Metab, 32:1578–1588.

16.

van Velthoven

CTJ

, Kavelaars

, van Bel

and Heijnen

. (2010). Repeated mesenchymal stem cell treatment after neonatal hypoxia-ischemia has distinct effects on formation and maturation of new neurons and oligodendrocytes leading to restoration of damage, corticospinal motor tract activity, and sensorimotor function. J Neurosci, 30:9603–9611.

17.

Donega

, van Velthoven

, Nijboer

, van Bel

, Kas

, Kavelaars

and Heijnen

. (2013). Intranasal mesenchymal stem cell treatment for neonatal brain damage: long-term cognitive and sensorimotor improvement. PLoS One, 8:e51253.

18.

Mahmood

, Lu

and Chopp

. (2004). Marrow stromal cell transplantation after traumatic brain injury promotes cellular proliferation within the brain. Neurosurgery, 55:1185–1193.

19.

van Velthoven

CTJ

, Kavelaars

, van Bel

and Heijnen

. (2011). Mesenchymal stem cell transplantation changes the gene expression profile of the neonatal ischemic brain. Brain Behav Immun, 25:1342–1348.

20.

Wakabayashi

, Nagai

, Sheikh

, Shiota

and Narantuya

. (2010). Transplantation of human mesenchymal stem cells promotes functional improvement and increased expression of neurotrophic factors in a rat focal cerebral ischemia model. J Neurosci Res, 88:1017–1025.

21.

Hao

, Zou

, Tian

, Zhang

, Zhou

and Liu

. (2014). Stem cell-based therapies for ischemic stroke. Biomed Res Int, 2014:468748.

22.

Donega

, Nijboer

, Braccioli

, Slaper-Cortenbach

, Kavelaars

, van Bel

and Heijnen

. (2014). Intranasal administration of human MSC for ischemic brain injury in the mouse: in vitro and in vivo neuroregenerative functions. PLoS One, 9:e112339.

23.

Danielyan

, Schafer

, von Ameln-Mayerhofer

, Buadze

, Geisler

, Klopfer

, Burkhardt

, Proksch

, Verleysdonk

, et al. (2009). Intranasal delivery of cells to the brain. Eur J Cell Biol, 88:315–324.

24.

van Velthoven

, Kavelaars

, van Bel

and Heijnen

. (2010). Nasal administration of stem cells: a promising novel route to treat neonatal ischemic brain damage. Pediatr Res, 68:419–422.

25.

, Li

, Lesniak

and Balyasnikova

. (2017). Intranasal delivery of therapeutic stem cells to glioblastoma in a mouse model. J Vis Exp Jun, 4:124.

26.

Mangraviti

, Tzeng

, Gullotti

, Kozielski

, Kim

, Seng

, Abbadi

, Schiapparelli

, Sarabia-Estrada

, et al. (2016). Non-virally engineered human adipose mesenchymal stem cells produce BMP4, target brain tumors, and extend survival. Biomaterials, 100:53–66.

27.

Kooijman

, Nijboer

, van Velthoven

, Mol

, Dijkhuizen

, Kesecioglu

and Heijnen

. (2014). Long-term functional consequences and ongoing cerebral inflammation after subarachnoid hemorrhage in the rat. PLoS One, 9:e90584.

28.

Prunell

, Mathiesen

, Diemer

and Svendgaard

. (2003). Experimental subarachnoid hemorrhage: subarachnoid blood volume, mortality rate, neuronal death, cerebral blood flow, and perfusion pressure in three different rat models. Neurosurgery, 52:165–175.

29.

Tiebosch

, van den Bergh

, Bouts

, Zwartbol

, van der Toorn

and Dijkhuizen

. (2013). Progression of brain lesions in relation to hyperperfusion from subacute to chronic stages after experimental subarachnoid hemorrhage: a multiparametric MRI study. Cerebrovasc Dis, 36:167–172.

30.

Tiebosch

, Dijkhuizen

, Cobelens

, Bouts

, Zwartbol

, van der Meide

and van den Bergh

. (2013). Effect of interferon-β on neuroinflammation, brain injury and neurological outcome after experimental subarachnoid hemorrhage. Neurocrit Care, 18:96–105.

31.

Sugawara

, Ayer

, Jadhav

and Zhang

. (2008). A new grading system evaluating bleeding scale in filament perforation subarachnoid hemorrhage rat model. J Neurosci Methods, 167:327–334.

32.

Willemen

, Eijkelkamp

, Wang

, Dantzer

, Dorn

2nd, Kelley

, Heijnen

and Kavelaars

. (2010). Microglial/macrophage GRK2 determines duration of peripheral IL-1beta-induced hyperalgesia: contribution of spinal cord CX3CR1, p38 and IL-1 signaling. Pain, 150:550–560.

33.

Chaplan

, Bach

, Pogrel

, Chung

and Yaksh

. (1994). Quantitative assessment of tactile allodynia in the rat paw. J Neurosci Methods, 53:55–63.

34.

Dantzer

, Heijnen

, Kavelaars

, Laye

and Capuron

. (2014). The neuroimmune basis of fatigue. Trends Neurosci, 37:39–46.

35.

, Liu

, Dwyer

, Banasr

, Lee

, Son

, Li

, Aghajanian

and Duman

. (2011). Glutamate N-methyl-d-aspartate receptor antagonists rapidly reverse behavioral and synaptic deficits caused by chronic stress exposure. Biol Psychiatry, 69:754–761.

36.

Schallert

, Fleming

, Leasure

, Tillerson

and Bland

. (2000). CNS plasticity and assessment of forelimb sensorimotor outcome in unilateral rat models of stroke, cortical ablation, parkinsonism and spinal cord injury. Neuropharmacology, 39:777–787.

37.

David

and Kroner

. (2011). Repertoire of microglial and macrophage responses after spinal cord injury. Nat Rev Neurosci, 12:388–399.

38.

Krishnan

and Nestler

. (2011). Animal models of depression: molecular perspectives. Curr Top Behav Neurosci, 7:121–147.

39.

Strekalova

, Couch

, Kholod

, Boyks

, Malin

, Leprince

and Steinbusch

. (2011). Update in the methodology of the chronic stress paradigm: internal control matters. Behav Brain Funct, 7:9.

40.

Papp

, Willner

and Muscat

. (1991). An animal model of anhedonia: attenuation of sucrose consumption and place preference conditioning by chronic unpredictable mild stress. Psychopharmacology (Berl), 104:255–259.

41.

Tokita

, Yamaji

and Hashimoto

. (2012). Roles of glutamate signaling in preclinical and/or mechanistic models of depression. Pharmacol Biochem Behav, 100:688–704.

42.

Wang

, Goffer

, Xu

, Tukey

, Shamir

, Eberle

, Zou

, Blanck

and Ziff

. (2011). A single subanesthetic dose of ketamine relieves depression-like behaviors induced by neuropathic pain in rats. Anesthesiology, 115:812–821.

43.

Walker

, Budac

, Bisulco

, Lee

, Smith

, Beenders

, Kelley

and Dantzer

. (2013). NMDA receptor blockade by ketamine abrogates lipopolysaccharide-induced depressive-like behavior in C57BL/6J mice. Neuropsychopharmacology, 38:1609–1616.

44.

Khalili

, Anvari

, Hekmati-Moghadam

, Sadeghian-Nodoushan

, Fesahat

and Miresmaeili

. (2012). Therapeutic benefit of intravenous transplantation of mesenchymal stem cells after experimental subarachnoid hemorrhage in rats. J Stroke Cerebrovasc Dis, 21:445–451.

45.

Khalili

, Sadeghian-Nodoushan

, Fesahat

, Mir-Esmaeili

, Anvari

and Hekmati-Moghadam

. (2014). Mesenchymal stem cells improved the ultrastructural morphology of cerebral tissues after subarachnoid hemorrhage in rats. Exp Neurobiol, 23:77–85.

46.

Danielyan

, Schafer

, von Ameln-Mayerhofer

, Bernhard

, Verleysdonk

, Buadze

, Lourhmati

, Klopfer

, Schaumann

, et al. (2011). Therapeutic efficacy of intranasally delivered mesenchymal stem cells in a rat model of Parkinson disease. Rejuvenation Res, 14:3–16.

47.

Danielyan

, Beer-Hammer

, Stolzing

, Schäfer

, Siegel

, Fabian

, Kahle

, Biedermann

, Lourhmati

, et al. (2014). Intranasal delivery of bone marrow-derived mesenchymal stem cells, macrophages, and microglia to the brain in mouse models of Alzheimer's and Parkinson's disease. Cell Transplant, 23:S123–S139.

48.

Frey

, 2nd. (2015). Noninvasive intranasal stem cells bypass the blood-brain barrier to target the brain to treat Parkinson's disease, stroke, MS, brain tumors, cerebral ischemia, Alzheimer's and other CNS disorders. J Nat Sci, 1:e23.

49.

Schroeter

, Franke

, Stoll

and Hoehn

. (2001). Dynamic changes of magnetic resonance imaging abnormalities in relation to inflammation and glial responses after photothrombotic cerebral infarction in the rat brain. Acta Neuropathol, 101:114–122.

50.

Donega

, Nijboer

, van Tilborg

, Dijkhuizen

, Kavelaars

and Heijnen

. (2014). Intranasally administered mesenchymal stem cells promote a regenerative niche for repair of neonatal ischemic brain injury. Exp Neurol, 261:53–64.

51.

Silasi

and Colbourne

. (2009). Long-term assessment of motor and cognitive behaviours in the intraluminal perforation model of subarachnoid hemorrhage in rats. Behav Brain Res, 198:380–387.

52.

Dziurdzik

, Krawczyk

, Jalowiecki

, Kondera-Anasz

and Menon

. (2004). Serum interleukin-10 in ICU patients with severe acute central nervous system injuries. Inflamm Res, 53:338–343.

53.

Jeon

, Ai

, Sabri

, Tariq

and Macdonald

. (2010). Learning deficits after experimental subarachnoid hemorrhage in rats. Neuroscience, 169:1805–1814.

54.

Ostrowski

, Colohan

and Zhang

. (2006). Neuroprotective effect of hyperbaric oxygen in a rat model of subarachnoid hemorrhage. Acta Neurochir Suppl, 96:188–193.

55.

Kigerl

, Gensel

, Ankeny

, Alexander

, Donnelly

and Popovich

. (2009). Identification of two distinct macrophage subsets with divergent effects causing either neurotoxicity or regeneration in the injured mouse spinal cord. J Neurosci, 29:13435–13444.

56.

Nauta

, Raaschou-Jensen

, Roos

, Daha

, Madsen

, Borrias-Essers

, Ryder

, Koch

and Garred

. (2003). Mannose-binding lectin engagement with late apoptotic and necrotic cells. Eur J Immunol, 33:2853–2863.

57.

Jose

, Tan

, Ooi

, Ramasamy

and Vidyadaran

. (2014). Mesenchymal stem cells exert anti-proliferative effect on lipopolysaccharide-stimulated BV2 microglia by reducing tumour necrosis factor-alpha levels. J Neuroinflammation, 11:149.

58.

Strekalova

, Spanagel

, Bartsch

, Henn

and Gass

. (2004). Stress-induced anhedonia in mice is associated with deficits in forced swimming and exploration. Neuropsychopharmacology, 29:2007–2017.

59.

Willner

. (1997). Validity, reliability and utility of the chronic mild stress model of depression: a 10-year review and evaluation. Psychopharmacology, 134:319–329.

60.

Dantzer

. (2009). Cytokine, sickness behavior, and depression. Immunol Allergy Clin North Am, 29:247–264.

61.

Walker

, Kavelaars

, Heijnen

and Dantzer

. (2014). Neuroinflammation and comorbidity of pain and depression. Pharmacol Rev, 66:80–101.

62.

Autry

and Monteggia

. (2012). Brain-derived neurotrophic factor and neuropsychiatric disorders. Pharmacol Rev, 64:238–258.

63.

Jiang

, Liu

, Han

, Dong

, Gao

, Du

, Meng

and Shi

. (2012). Neuroprotective effects of anti-tumor necrosis factor-alpha antibody on apoptosis following subarachnoid hemorrhage in a rat model. J Clin Neurosci, 19:866–872.