University of Amsterdam
Psychobiology Bachelor’s Thesis
The efficacy of mesenchymal stem cell (MSC) therapy and
docosahexaenoic acid (DHA)-enriched nutrition in dampening
neuroinflammation in neonatal hypoxic-ischemic
encephalopathy (HIE)
Author
Eline Groeneveld
11595396
Supervisors
Caroline de Theije, Dr.
Myrna Brandt, PhD
22 January 2021
Abstract
Neonatal hypoxic-ischemic encephalopathy (HIE) due to perinatal asphyxia is a major cause of death
and neurological impairments, while its treatment options are very limited. Following hypoxic-ischemic
(HI) insult, inflammation is rapidly induced in the brain and is known to be a key player in the mediation
of brain damage. Lately, mesenchymal stem cell (MSC) therapy and docosahexaenoic acid
(DHA)-enriched nutrition have shown promising effects in the treatment of HI-induced brain injury. However,
the underlying mechanisms of action remain to be elucidated. Here, we proposed dampening
neuroinflammation as a mechanism of action of MSC treatment and DHA-enriched nutrition. HI was
induced in postnatal day 9 (P9) mice which either received MSC treatment at P12 or DHA-enriched
nutrition throughout the whole experiment. Inflammatory markers (i.e. astrocyte reactivity and
microglia activation) were analyzed at 4 weeks and 15 weeks after HI for MSC treatment and
DHA-enriched nutrition, respectively. We show that HI resulted in increased astrocyte reactivity at both 4
and 15 weeks. Moreover, MSC treatment and DHA-enriched nutrition effectively reduced astrocyte
reactivity in HI-animals in specific regions. Microglia activation was assessed by morphological analysis.
Only the Feret’s diameter of microglia was reduced in HI mice and no treatment effects on microglia
activation were detected. Overall, our data suggests MSC therapy and DHA-enriched nutrition to be
promising strategies to target neuroinflammation in HIE.
1.
Introduction
Oxygen deprivation around birth, also known as perinatal asphyxia, is one of the leading causes of death and severe brain injury in children1,2. If not lethal, it typically results in mild to severe neonatal hypoxic-ischemic
encephalopathy (HIE)1. This subtype of neonatal encephalopathy has an incidence ranging from 1 to 8 per 1000
live births3 and is associated with various sequalae such as cerebral palsy, epilepsy and cognitive impairments4.
The burden is high for patients, their family and society as HIE comes with lower quality of life, lost years of life and high financial costs4,5.
The neurological outcome following the hypoxic-ischemic (HI) insult depends on the severity and duration of the insult, gestational age at time of the insult and timing of intervention1. Timing of intervention plays a particularly
important role, since HIE is rather an ongoing process than a single event (Fig. 1)6. This process is characterized by
several distinct phases. The primary energy failure occurs immediately after the insult and is identified by the depletion of the cell’s high-energy stores resulting in anaerobic metabolism, cytotoxic edema and excitotoxicity which eventually leads to primary neuronal death7,8. Reperfusion is followed by a latency phase (~6 hours) during
which recovery from oxidative stress and inflammation take place7,9. Depending on the severity of brain damage
following primary energy failure, secondary energy failure commences and ensues in another failure of oxidative metabolism, accumulation of excitotoxins, renewed cytotoxic edema, chronic inflammation and ultimately secondary neuronal death9–11. This phase normally occurs between 6 and 48 hours after HI insult and can last for
days9,11. In some infants, active mechanisms – such as inflammation and epigenetic changes - become persistent
and complicate repair and regeneration of neurons in the tertiary phase. These active processes might persist for months to years and therefore exacerbate brain damage and predispose infants to further injury9,12.
Figure 1. Schematic overview of hypoxic-ischemic encephalopathy as defined by its different phases. Image retrieved from Douglas-Escobar & Weiss (2015).
Inflammation is a key mediator of HI-induced brain injury and is characterized by the multifaceted interplay of immunomodulatory cells and processes8. Microglia, the main resident immune cells of the brain, constantly
surveillance the brain’s microenvironment and play a critical role in the initiation of inflammation13–15. In response
to HI, microglia become activated, increase in number and take on an amoeboid morphology characterized by an enlarged cell body and fewer branches16. Activated microglia act like macrophages through phagocytosis,
secretion of pro- and anti-inflammatory cytokines and the release of matrix metalloproteinases (MMPs), which contribute to breakdown of the blood-brain barrier (BBB)8,17. As a consequence, peripheral immune cells – such
as leukocytes - infiltrate the brain and worsen inflammatory processes17. Not only microglia, but astrocytes too,
become rapidly activated after HI-injury. Astrocytes fulfill several regulatory functions, but are mostly important in the support of neurons through mechanisms like energy delivery and neurotransmitter uptake18. In response to
HI-insult, astrocytes become reactive and increase significantly in number, a process known as reactive astrogliosis15,19. This innate defensive mechanism initially aims to limit brain damage and restore homeostasis19.
However, reactive astrocytes are also known to produce chemokines and pro-inflammatory cytokines8,15.
Since the extent of damage increases tremendously with time after asphyxia it is of great importance to both recognize and treat HIE as soon as possible. Currently, the only treatment for perinatal asphyxia is therapeutic hypothermia (TH)20 during which the body temperature is lowered to 33-34°C for 48-72 hours21. Although the
neuroprotective function of TH is well established22, its therapeutic window of opportunity is severely limited.
Ideally, TH is started within the latency phase to achieve its optimal neuroprotective effect21,23. However, even
when infants have undergone TH within the first three hours, they still face serious neurological difficulties later on in life21. Hence, novel treatment options for HIE are urgently needed.
In the past few years, mesenchymal stem cell (MSC) therapy emerged as an effective strategy to treat HIE. These non-embryonic stem cells have great therapeutical potential mainly due to their differentiation capacity, ready availability (i.e. from bone marrow or umbilical cord) and rapid in vitro expansion24. Moreover, MSCs are known
to secrete certain trophic factors which positively affect immune responses, cell survival and cell proliferation24,25.
Treatment with MSCs improved motor behavior in rodent models of neonatal HIE and induced neuronal and oligodendrocyte regeneration and synaptogenesis in the brain26–28. Intranasal MSC therapy also showed to reduce
lesion size and improve cognitive behavior in neonatal HI injured mice29,30. Additionally, Donega et al.
demonstrated that intranasally administered MSCs migrate specifically to the site of lesion where they promote neuroregeneration31.
It has long been known that optimal nutrition is of great importance for the growth of term and preterm infants. Only recently, medical nutrition is being seen as a promising strategy to treat brain injury. Also, medical nutrition is often considered to be relatively safe, free of side effects and it is easily implemented into the clinic32. Lately,
long-chain n-3 polyunsaturated fatty acids (PUFAs) have gained the interest as neuroprotective agents. These PUFAs are known to play a significant role in brain development, regulation of oxidative stress and composition of neuronal membranes33. Especially the fatty acid docosahexaenoic acid (DHA) seems to be a promising candidate34.
Nutritional enrichment of DHA is thought to provide neuroprotection in brain injury33,35,36. Moreover, some studies
also addressed its beneficial effects on regenerative processes, such as synaptogenesis37–39. In a model of perinatal
hypoxia-ischemia, DHA treatment was shown to improve functional outcome and reduce brain damage40,41, while
depletion of DHA led to impaired recovery after traumatic brain injury42.
Thus, previous research has demonstrated the effectiveness of MSC and DHA therapy in reducing lesion size and improving functional outcome after HI injury26–31,40,41. However, the mechanisms of action of these therapies are
still not yet fully understood. Since neuroinflammation plays a critical role in mediating HI induced brain damage8,
we considered this a possible mechanism of action of MSC treatment and DHA-enriched nutrition. Hence, we aimed to investigate the efficacy of both MSC treatment and DHA-enriched nutrition in dampening neuroinflammation in a model of neonatal HIE. Both therapies are expected to be effective in dampening neuroinflammation by reducing both astrocyte reactivity and microglial activation. Additionally, we expect DHA-enriched nutrition to reduce synapse loss in neonatal HI-injured mice.
In this study, we used the well-validated Rice-Vannucci model (RVM) as a model for neonatal HIE43. In this model,
HIE is induced by unilateral occlusion of the common carotid artery followed by a period of hypoxia performed in mice at the age of 9 days43. This procedure results in hippocampal and thalamic brain damage extending toward
cortical areas and neurological impairments corresponding to term infants with severe HIE43. In two distinct
experiments, mice either received MSC treatment or DHA-enriched nutrition. The level of inflammation was examined by analyzing astrocyte reactivity and microglia activation using immunohistochemical staining. Additionally, we investigated the effects of DHA-enriched nutrition on synapse loss as assessed by synaptophysin staining.
2. Material and Methods
Ethics statement
During all procedures Dutch and European international guidelines (Directive 86/609, ETS 123, Annex II) were followed. The Animal Experiment Committee (Animal Welfare Body Utrecht, Utrecht, Netherlands) approved all animal experiments. All efforts were made to minimize animal suffering.
Animals
C57BL/6 mice were bred and housed under standard housing conditions at the shared animal facility in Utrecht. All mice were exposed to a 12 hour light/dark cycle and had ad libitum access to food and water. At the day of birth (P0) pups were randomly assigned to predefined groups in such a manner that both sexes were evenly represented within every group.
Surgery and hypoxia
According to the RVM, 9-days old pups were anesthetized with isoflurane (5% induction and 3% maintenance) and underwent unilateral ligation of the right carotid artery. Afterwards, pups were put back in their home cage for 1-1.5 hour recovery. Bupivacaine 0.5% (Actavis, New Jersey, USA) and Xylocaine 2% (AstraZeneca, Den Haag, The Netherlands) were used for pain relief before, during and after surgery. After recovery, pups were put in a temperature controlled box (ThermaCage) at 36.0- 36.5°C and exposed to an hypoxic environment (10% O2) for
45 minutes. Sham animals underwent anesthesia and isolation of the right carotid artery only.
MSC treatment
After birth, pups were randomly assigned to three different groups: sham operated animals (n = 15), HI vehicle treated animals (n = 15) and HI MSC treated animals (n = 11). MSCs retrieved from umbilical cord (Chiesi, Parma, Italy) were cultured in basal medium (Chiesi, Parma, Italy) containing 5% platelet Lysate supplement (Lonza, Basel, Switzerland) and 0.2% Heparin (APP Pharmaceuticals, Schaumburg, IL, USA). MSCs were cultured under normoxic conditions (37°C, 5% CO2, 21% O2) and passaged once prior to administration. Three days after HI-induction, pups
were treated intranasally with 2.0x106 MSCs (3 drops of 2µl per nostril). Thirty minutes before intranasal
administration of MSCs mice were treated with hyaluronidase (100 U, Sigma-Aldrich, St. Louis, MO, USA) in order to pretreat the connective tissue of the nasal cavity. At 28 days post-surgery, unilateral sensorimotor impairments were measured using the cylinder rearing test (CRT). Mice were sacrificed after completing the CRT.
DHA-enriched nutrition
On P0, pups were assigned to two different groups. One group received the DHA-enriched diet (sham-operated: n = 19; HI-operated: n = 19) and the other group received an isocaloric control diet (AIN93G; Ssniff Spezialdiäten, Soest, Germany) (sham-operated: n = 20; HI-operated: n = 19) for the whole duration of the experiment. One day after surgery (P10) nests were culled to a maximum of 7 pups. Four weeks after HI-induction, mice were weaned and continuously fed the assigned diet. At 8.5 weeks after HI-surgery mice completed the Rotarod task and were then put on a reversed day-night cycle. The Novel Object Recognition task (NORT) could therefore be performed during the active phase under red light conditions at 10 weeks after HI-surgery. At 15 weeks, the CRT was performed and mice were sacrificed thereafter.
Tissue preparation, histology and immunohistochemistry.
When sacrificed, mice were anesthetized with an intraperitoneal injection (0.1ml 20% pentobarbital) and then intracardially perfused using phosphate-buffered saline (PBS) and 4% paraformaldehyde (PFA). Brains were harvested and embedded in paraffin prior to slicing. Using a microtome, brains were sliced into 8μm thick coronal sections and mounted on glass slides. After deparaffinization and antigen retrieval, slides were treated with 1% Saponin and 10% Bovine Serum Albumin (BSA) blocking solution for 1 hour. The slides were then washed and incubated overnight with primary antibodies. For astrocyte and microglia staining, mouse anti-glial fibrillary acidic protein (GFAP, 1:100, Acris, Herford Germany) and rabbit anti-ionized calcium-binding adaptor protein-1 (IBA-1, 1:500, Wako Chemicals, Richmond VA, USA) were used, respectively. For dendrite and synapse staining, mouse-anti-microtubule-associated protein 2 (MAP2, 1:500, Sigma Aldrich) and rabbit-anti-synaptophysin (1:400, Abcam, Cambridge MA, USA) were used, respectively. The next day, slides were washed, incubated 1 hour with secondary antibodies goat-anti-mouse Alexa488 (1:500, Invitrogen) and goat-anti-rabbit Alexa594 (1:500, Invitrogen) and washed again. Afterwards, slides were incubated with 4’,6-diamidino-2-phenylindole (DAPI, 1:5000, Invitrogen) for nuclei staining, washed in distilled water and finally mounted under coverslips using FluorSave reagent (VWR international, Amsterdam, The Netherlands).
Microscopy and image analysis
Fluorescence images were obtained using a Zeiss AxioCam MRM (Carl Zeiss, Breda, The Netherlands) on an Axio Observer Z1 Microscope with Zen 2 (Blue Edition, Carl Zeiss). Images for astrocyte and microglia analysis were taken in the cortical, hippocampal and perilesional regions (Fig. 2 left) at 20x magnification. For synapse analysis images of the cortex and thalamus (Fig. 2 right) were taken at 40x magnification. All images were analyzed using
FIJI 1.49u software. Grayscale images were used for astrocyte analysis and manually thresholded in order to determine the GFAP positive (GFAP+) area. Microglia morphological analysis was performed by semiautomatic selection of iba1 positive (iba1+) cells. The perimeter, circularity and Feret’s diameter of every selected microglia was then measured. Synaptophysin positive (SYN+) area and intensity were determined using greyscale images which were thresholded manually.
Figure 2. Schematic overview of the analyzed brain regions. (Left) Red, blue and green squares indicate the cortical, hippocampal and perilesional regions, respectively, where pictures were taken for GFAP and iba1 analysis. (Right) For synaptophysin analysis, red and blue squares indicate the places where pictures in the cortex and thalamus were taken, respectively.
Statistical analysis
Data analysis was performed in a blinded manner. Upon image analysis, it was decided to exclude images taken in the hippocampal area from statistical analysis due to severe hippocampal tissue loss in HI-mice. To evaluate differences between groups, one-way ANOVA was used, followed by Tukey’s multiple comparisons test. Additionally, two-way ANOVA and Tukey’s multiple comparisons test were used to investigate sex-dependent effects. Results with p < 0.05 were considered significant. Data are presented as mean ± SEM.
3. Results
The effects of MSC therapy on astrocyte reactivity
Here, we investigated the astrocytic response in mice 28 days after HI-induction. We assessed GFAP staining as a measure for astrocyte reactivity in cortical and perilesional regions (Fig. 4).
In the cortex of the contralateral hemisphere, results showed no significant difference in GFAP+ area between groups (data not shown). Our data in figure 3a show that MSC and vehicle treated animals had an increased area of GFAP+ cells in the ipsilateral cortex compared to sham-operated animals (MSC: p < 0.05; vehicle: p < 0.001), implying an effect of HI. Although graphs (Fig. 3a, b) demonstrate lower GFAP protein levels after MSC treatment compared to vehicle treatment, this was not found to be significant. In line with previous studies, we next set out to investigate whether analyzing females and males separately resulted in different outcomes44–47. In figure 3b
data reveal a trend toward decreased cortical GFAP protein levels for females only when compared to vehicle treated animals.
Around the hippocampus of the contralateral hemisphere no differences in GFAP+ area between groups was observed (data not shown). In the perilesional region of the ipsilateral hemisphere, the area of GFAP+ cells was increased in vehicle animals compared to sham (p < 0.0001), whereas GFAP+ area was reduced in MSC treated mice compared to vehicle treated mice (Fig. 3c). No sex-dependent effects were observed (Fig. 3d).
Sham Vehicle MSC 0 1 2 3 G F A P + a re a ( % ) ✱ ✱ ✱ ✱ Sham Vehicle MSC 0 1 2 3 4 G F A P + a re a ( % ) ✱ ✱ ✱ ✱ ✱ Female Male 0 1 2 3 4 5 G F A P + a re a ( % ) ✱ ✱ ✱ ✱ Female Male 0 1 2 3 4 G F A P + a re a ( % ) ✱ ✱ ✱
c
o
rt
e
x
p
e
ri
le
s
io
n
MSC Vehicle Shamc
a
b
d
Figure 3. Effect of MSC treatment on astrocyte reactivity in the ipsilateral hemisphere 28 days after HI. Mice were subjected to sham operation or HI at P9. HI mice were treated intranasally with vehicle or 2.0x106 MSCs 3 days post-HI. GFAP staining
was used to determine astrocyte reactivity in the cortex (a, b) and in the region adjacent to the lesion (c, d) in the ipsilateral hemisphere. (a, c) GFAP+ area was calculated as percentage of total area and (b, d) analyzed for each sex separately. Two-way ANOVA showed no interaction effect of treatment and sex. (a, c) n = 11 – 15 animals per group. (b, d) n = 5 – 9 animals per group. Data represent means ± SEM. * p < 0.05, *** p < 0.001, **** p < 0.0001.
c
o
rt
e
x
p
e
ri
le
s
io
n
sham
vehicle
MSC
Figure 4. Astrocyte reactivity is upregulated in the ipsilateral hemisphere 28 days after HI induction. Representative images of immunohistochemical staining for GFAP in cortical and perilesional regions. Scale bars: 50 µm.
The effects of MSC therapy on microglia activation
Next, we investigated the effects of MSC therapy on the activation of microglia 28 days after HI-insult. In order to do so, we analyzed the morphology of iba1 positive cells assessed by their perimeter, Feret’s diameter and circularity (Fig. 7).
When comparing the morphology of iba1+ cells in the contralateral cortex, no significant differences were detected between groups for all three parameters (data not shown). One-way ANOVA analyses revealed no significant differences in the ipsilateral cortex for iba1+ cell’s perimeter (Fig. 5a, b) and circularity (Fig. 5e, f). The Feret’s diameter of iba1+ cells in the ipsilateral cortex was significantly decreased in vehicle mice compared to sham-operated mice (p < 0.05), but not in MSC-treated mice (Fig. 5c). Nonetheless, graphs in figure 5 show that microglia in the ipsilateral cortex had a greater perimeter and Feret’s diameter, and lower circularity after MSC treatment compared to vehicle treatment, although these differences were not statistically significant. Two-way ANOVA and post hoc analysis on sex-dependent effects revealed no differences (Fig. 5b, d and f).
Analysis on morphological differences of iba1+ cells in the contralateral region reflecting the lesion site revealed no significant differences between groups for cell’s perimeter, Feret’s diameter and circularity (data not shown). No significant differences in iba1+ cell’s appearance in the ipsilateral perilesional region was detected between groups, although graphs show trends toward sham-like results for all three parameters after MSC treatment (Fig. 6a, c and e). We then investigated the effects of sex on morphological outcomes. These results showed a significant difference in iba1+ cell’s perimeter between MSC- and vehicle-treatment in females only (p < 0.05). Despite no further significant differences, graphs in figure 6b, d and f show positive trends in MSC-treatment outcomes in females, whereas male groups have more similar outcomes for cell’s perimeter, Feret’s diameter and circularity in the perilesional area.
Sham Vehicle MSC 300 350 400 450 500 IB A 1 + c e ll p e ri m e te r Female Male 300 350 400 450 500 IB A 1 + c e ll p e ri m e te r Sham Vehicle MSC 80 90 100 110 IB A 1 + F e re t' s d ia m e te r ✱ Female Male 80 90 100 110 IB A 1 + F e re t' s d ia m e te r Sham Vehicle MSC 0.05 0.10 0.15 0.20 IB A 1 + c ir c u la ri ty Female Male 0.05 0.10 0.15 0.20 IB A 1 + c ir c u la ri ty MSC Vehicle Sham
a
b
c
d
e
f
Figure 5. Effect of MSC treatment on microglia morphology in the ipsilateral cortex 28 days after HI. On P9, mice were subjected to either sham operation or HI. Three days post HI, mice were treated intranasally with vehicle or 2.0x106 MSCs. Iba1 was used
as a marker for microglia. The cell’s (a, b) perimeter, (c, d) Feret’s diameter and (e, f) circularity were assessed as measures for morphology analysis. (a, c, e) MSC treatment did not affect microglia morphology (b, d, e) nor were any sex-dependent effects present. Two-way ANOVAs showed no interaction effect of treatment and sex. (a, c, d) n = 11 – 15 animals per group. (b, d)
Sham Vehicle MSC 300 350 400 450 500 IB A 1 + c e ll p e ri m e te r Female Male 300 350 400 450 500 IB A 1 + c e ll p e ri m e te r ✱ Female Male 0.05 0.10 0.15 0.20 IB A 1 + c ir c u la ri ty Female Male 80 90 100 110 IB A 1 + F e re t' s d ia m e te r Sham Vehicle MSC 0.05 0.10 0.15 0.20 IB A 1 + c ir c u la ri ty Sham Vehicle MSC 80 90 100 110 IB A 1 + F e re t' s d ia m e te r MSC Vehicle Sham
a
b
e
c
d
f
Figure 6. Effect of MSC treatment on microglia morphology in the ipsilateral perilesional region 28 days after HI. On P9, mice underwent either sham or HI surgery. Three days after HI-induction, vehicle or 2.0x106 -MSCs were administered intranasally.
Iba1 was used as a immunohistochemical marker for microglia. The cell’s (a, b) perimeter, (c, d) Feret’s diameter and (e, f) circularity were assessed as measures for morphology analysis. (a, c, e) MSC treatment did not affect microglia morphology. (b, d, e) Iba1+ cell’s perimeter was only significantly different between female vehicle and MSC treated mice (p < 0.05). Two-way ANOVAs showed no interaction effect of treatment and sex. (a, c, d) n = 11 – 15 animals per group. (b, d, e) n = 5 – 9 animals per group. Data represent means ± SEM. * p < 0.05.
c
o
rt
e
x
p
e
ri
le
s
io
n
sham
vehicle
MSC
Figure 7. Microglia morphology in the ipsilateral hemisphere 28 days after HI induction. Representative fluorescent images of immunohistochemical staining for iba1 in cortical and perilesional regions. Scale bars: 50 µm.
The effects of DHA-enriched nutrition on astrocyte reactivity
Fifteen weeks after HI, immunohistochemical staining was performed to investigate the effect of DHA-enriched nutrition on astrocyte reactivity. The GFAP positive area was assessed as measure for astrocyte reactivity in cortical and perilesional regions (Fig. 9).
One-way ANOVA analyses showed no significant differences in the contralateral cortical region (data not shown). In the ipsilateral hemisphere, cortical GFAP+ area was upregulated in HI-animals fed the control diet (p < 0.0001), but not in HI-animals fed the DHA-enriched diet when compared to their corresponding control group (Fig. 8a). Additionally, the DHA-enriched diet significantly reduced the GFAP+ area in HI-animals compared to HI-animals that received the control diet (p < 0.01). No effect of sex was detected with two-way ANOVA analysis. However, post-hoc analysis revealed a significant downregulation of GFAP protein levels in HI-animals fed the DHA-enriched diet compared to HI-animals fed the control diet (p < 0.01) as seen in figure 8b. In contrast, HI-females showed no decrease in GFAP protein levels when they were fed the DHA-enriched diet (Fig. 8b).
GFAP protein levels in the perilesional area of the contralateral hemisphere revealed not to be different among groups (data not shown). Irrespective of diet, HI-animals had upregulated GFAP+ area in the ipsilateral perilesional region (Fig. 8c) when compared to the corresponding control group. The DHA-enriched diet showed not to be effective in decreasing the upregulated GFAP+ area in HI-animals (Fig. 8c). In figure 8d, graphs show no effect of sex on treatment outcomes. Interestingly, figures 8b and d show smaller GFAP+ areas for HI-females compared to HI-males, implying a possible difference in astrocytic response between sexes.
Sham HI Sham HI 0 1 2 3 4 G F A P + a re a ( % ) ✱ ✱ ✱ ✱ ✱ ✱ DHA Control Female Male 0 1 2 3 4 5 G F A P + a re a ( % ) ✱ ✱ ✱ ✱ ✱ ✱ Sham HI Sham HI 0 2 4 6 G F A P + a re a ( % ) ✱ ✱ ✱ ✱ ✱ DHA Control Female Male 0 2 4 6 8 G F A P + a re a ( % ) ✱ ✱ c o rt e x p e ri le s io n a l DHA, Sham DHA, HI Control, Sham Control, HI
a
b
c
d
Figure 8. Effect of DHA-enriched diet on astrocyte reactivity in the brain 15 weeks after HI. Mice were subjected to sham operation or HI at P9. Throughout the whole experiment, mice were fed either the DHA-enriched diet or control diet. GFAP staining was used to determine astrocyte reactivity in the cortex (a, b) and in the region adjacent to the lesion (c, d) in the ipsilateral hemisphere. (a, c) n = 19 – 20 animals per group. (b, d) n = 8 – 11 animals per group. Data represent means ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.
control, sham control, HI DHA, sham DHA, HI c o rt e x p e ri le s io n
Figure 9. Astrocyte reactivity is upregulated in the ipsilateral hemisphere 15 weeks after HI induction. Representative images of cortical and perilesional regions stained for astrocyte marker GFAP. Scale bars: 50 µm.
The effects of DHA-enriched nutrition on microglia activation
To determine the effects of DHA-enriched diet on microglia activation, we performed morphological analysis on microglia 15 weeks after HI-insult. The activated state of microglia was assessed by measuring the perimeter, Feret’s diameter and circularity of Iba1 positive cells in the cortical and perilesional area (Fig. 12).
In the contralateral cortex area, no differences in iba1+ cell’s perimeter, Feret’s diameter and circularity were found among groups (data not shown). In the ipsilateral hemisphere, there was an HI-effect in animals that received the DHA-enriched diet (Fig. 10a, c and d). Here, the cell’s perimeter and Feret’s diameter were decreased in HI-animals (perimeter: p < 0.05; Feret’s diameter: p < 0.001), while circularity was increased compared to the operated animals (p < 0.01). No morphological differences were seen in microglia between HI- and sham-operated animals that received the control diet, except for the Feret’s diameter (p < 0.05) as seen in figure 10c. Interestingly, two-way ANOVA analyses and Tukey’s multiple comparisons tests revealed an HI-effect in male mice that received the control diet (perimeter: p < 0.05; Feret’s diameter: p < 0.01; circularity: p < 0.01), but not in females that received the control diet (Fig. 10b, d and f). Moreover, no effect of DHA-enriched diet was seen in both females and males (Fig. 10b, d and f). However, a trend toward a more rested phenotype of microglia is seen in sham-operated females fed the DHA-enriched diet, while in HI-females on the DHA-enriched diet a trend toward a more activated phenotype is detectable when compared to the corresponding animals that received the control diet (Fig. 10b, d and f).
Next, we analyzed the effects of DHA-enriched diet in the perilesional region. Here, no effects of HI were seen in both the contralateral and the ipsilateral hemisphere (Fig. 11). Furthermore, DHA-enriched diet did not effectively change microglia morphology in the perilesional area in both hemispheres (Fig. 11).
c
Sham HI Sham HI 200 300 400 500 IB A 1 + c e ll p e ri m e te r ✱ Control DHA Female Male 200 300 400 500 IB A 1 + c e ll p e ri m e te r ✱ Sham HI Sham HI 70 80 90 100 110 IB A 1 + F e re t' s d ia m e te r ✱ ✱ ✱ ✱ Control DHA Female Male 70 80 90 100 110 IB A 1 + F e re t' s d ia m e te r ✱ ✱ ✱ ✱ Female Male 0.05 0.10 0.15 0.20 0.25 IB A 1 + c ir c u la ri ty ✱ Sham HI Sham HI 0.05 0.10 0.15 0.20 0.25 IB A 1 + c ir c u la ri ty ✱ Control DHA DHA, HI DHA, Sham Control, HI Control, Shama
b
d
e
f
Figure 10. Effect of DHA-enriched nutrition on microglia morphology in the ipsilateral cortex 15 weeks after HI. On P9, mice were subjected to either sham operation or HI. Throughout the whole experiment, mice received either the control or DHA-enriched diet. Iba1 was used as a marker for microglia. The cell’s (a, b) perimeter, (c, d) Feret’s diameter and (e, f) circularity were assessed as measures for morphology analysis. (a, c, e) An HI-effect was seen in mice that received DHA-enriched nutrition and (c) in mice that received the control diet. (b, d, e) HI injury significantly changed the morphological profile of microglia in male mice that were fed the control diet. (d) The Feret’s diameter of iba1+ cells was significantly decreased in HI-females that received DHA-enriched nutrition. Two-way ANOVA showed no interaction effect of treatment and sex for all three parameters. (a, c, d) n = 19 – 20 animals per group. (b, d, f) n = 8 – 11 animals per group. Data represent means ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001.
Sham HI Sham HI 200 300 400 500 IB A 1 + c e ll p e ri m e te r Control DHA Female Male 300 350 400 450 500 IB A 1 + c e ll p e ri m e te r Sham HI Sham HI 80 90 100 110 IB A 1 + F e re t' s d ia m e te r Control DHA Sham HI Sham HI 0.05 0.10 0.15 0.20 0.25 IB A 1 + c ir c u la ri ty Control DHA Female Male 80 90 100 110 IB A 1 + F e re t' s d ia m e te r Female Male 0.05 0.10 0.15 0.20 0.25 IB A 1 + c ir c u la ri ty DHA, HI DHA, Sham Control, HI Control, Sham
a
b
c
d
e
f
Figure 11. Effect of DHA-enriched nutrition on microglia morphology in the perilesional region 15 weeks after HI. Mice pups were subjected to either sham surgery or HI at P9. For the whole duration of the experiment, mice were fed either the control or the DHA-enriched. Microglia were visualized by immunohistochemical staining for Iba1. The iba1 positive cell’s (a, b) perimeter, (c, d) Feret’s diameter and (e, f) circularity were assessed as measures for morphology analysis. (a, c, e) DHA-enriched nutrition did not affect microglia morphology (b, d, e) nor were any sex-dependent effects present. Two-way ANOVAs showed no interaction effect of treatment and sex. (a, c, d) n = 19 – 20 animals per group. (b, d, f) n = 8 – 11 animals per group. Data represent means ± SEM.
control, sham control, HI DHA, sham DHA, HI
c o rt e x p e ri le s io n
Figure 12. Microglia morphology in the ipsilateral hemisphere 15 weeks after HI induction. Representative fluorescent images of immunohistochemical staining for iba1 in cortical and perilesional regions. Scale bars: 50 µm.
The effects of DHA-enriched nutrition on synapse loss
Synaptophysin is a presynaptic vesicle protein and is often used as indication for the number of
functional synapses
48. Here, we used immunohistochemical analysis to determine the synaptophysin
positive (SYN+) area and intensity as markers for synaptic loss (Fig. 14). At 15 weeks after HI, we
investigated whether DHA-enriched nutrition affected the amount of synaptic loss in both cortex and
thalamus. One-way ANOVA analysis showed no effect of HI in both the cortex and thalamus in the
contralateral hemisphere. Furthermore, no effect of DHA-enriched nutrition on neither SYN+ area nor
intensity was detected (data not shown). In the ipsilateral hemisphere, HI had no effect on the SYN+
area and intensity for both the cortex and thalamus (Fig. 13). As seen in figure 13, DHA-enriched diet
showed no effects on SYN+ area and intensity. Two-way ANOVA analysis showed no sex-dependent
effects on treatment outcome (data not shown).
Sham HI Sham HI 0 1 2 3 4 S Y N + a re a ( % ) Control DHA Sham HI Sham HI 0 1 2 3 4 S Y N + a re a ( % ) Control DHA Sham HI Sham HI 30 35 40 45 50 S Y N + i n te n s it y Control DHA Sham HI Sham HI 30 35 40 45 50 S Y N + i n te n s it y Control DHA c o rt e x th a la m u s
a
b
c
d
Figure 13. Effect of DHA-enriched nutrition on synapse loss in the cortex and thalamus 15 weeks after HI injury. On P9, mice underwent either sham or HI surgery. Mice received either the control or the DHA-enriched diet for the whole duration of the experiment. Synaptophysin was used as a immunohistochemical marker for synapses. Quantification of synapse loss as measured by (a, c) synaptophysin positive area and (b, d) intensity in both (a, b) the cortex and (c, d) thalamus. Data represent mean ± SEM.
control, sham control, HI DHA, sham DHA, HI
c o rt e x p e ri le s io n
Figure 14. Immunohistochemical staining for synaptophysin 15 weeks after HI-insult. Representative images of synaptophysin positive area in the cortex and thalamus in the ipsilateral hemisphere. Scale bars: 50 μm.
4.
Discussion
In the present study, we investigated the efficacy of MSC treatment and DHA-enriched nutrition in dampening neuroinflammation in a neonatal HIE mouse model. First, we demonstrated that astrocyte reactivity is increased in the brain 4 and 15 weeks after HI-insult, a known hallmark for inflammation in the brain8,49. This was effectively
downregulated by MSC treatment and DHA-enriched nutrition in the perilesional and cortical region, respectively. Next, we showed that HI injury reduced the Ferret’s diameter of microglia in the cortex. However, MSC treatment nor DHA-enriched nutrition were effective in reversing this effect. Taken this together, it can be concluded that MSC treatment and DHA-enriched nutrition are promising strategies to dampen astrocyte reactivity in certain brain regions of neonatal HI-injured mice, while no strong evidence was found for the effects on microglia activation.
Previous research has demonstrated the effectiveness of both MSC and DHA therapy in reducing lesion size and improving functional outcome after HI injury26,29–31,40,41. However, the mechanisms of action of these therapies
are still not yet fully understood. Since neuroinflammation plays a critical role in mediating HI-induced brain damage8, we proposed this a process modulated by MSC treatment and DHA-enriched nutrition.
Neuroinflammation in the brain is characterized by a complex interplay of several cell types and processes, which all occur at different time points and for a variable period of time8,11. One of the initial steps of inflammatory
responses after HI-insult is the activation of microglia, which is characterized by a change in phenotype (i.e. rounder cell body and less ramified branches) and the secretion of both pro- and anti-inflammatory cytokines8,13,16.
Furthermore, shortly after the insult, astrocytes are activated which initially can be beneficial, but becomes detrimental at later stages after the insult8,18,19.
It is known from clinical and experimental studies that outcome and inflammatory responses in neonatal HIE are sex specific44–47, although the underlying mechanisms remain not fully understood. Based on trends, we detected
different patterns in the outcomes and responses between female and male mice. Treatment with MSCs seemed to be more effective reducing astrocyte activity in female mice, while in male mice the same effect was less apparent. In contrast, DHA-enriched nutrition appeared to be more effective in male mice and less in females when astrocyte activity was measured. This contrast can be explained by the different time points at which responses were measured in the two experiments. Whereas the effects of MSC treatment were determined at 4 weeks after HI-induction, effects of DHA-enriched nutrition were investigated at 15 weeks post-HI. At 4 weeks, levels of astrocytic reactivity were merely the same between vehicle treated HI-male and -female mice, while at 15 weeks, HI-females that received the control diet showed lower astrocytic reactivity compared to the corresponding male group. These observations seem to be in line with earlier findings concerning more persistent inflammatory responses in males than in females44–47. However, it must be noted that our study was initially not
set up to find any sex-related differences. As a consequence, this study lacks certain requirements, such as greater group sizes, to make hard conclusions on these observations. Nevertheless, since more studies have mentioned sexual dimorphism in neonatal HIE45,47, this might be an important implication for both further research and the
implementation of new therapies into the clinic.
Although several studies have reported enhanced microglia activation in both early and later stages of HIE15,25, we
did not detect a significant overall effect of HI on microglial activation at both 4 and 15 weeks after HI-induction. However, we could observe some promising trends in our results. Interestingly, results of MSC treatment showed trends of moving microglia toward a more rested morphological profile in both cortical and perilesional regions. Moreover, when we observed the morphology of microglia in the perilesional region for females and males separately, microglia seemed to be more activated in vehicle treated females than in vehicle treated males. In addition, in females MSC treatment seemed to move microglia morphology toward a more sham-like morphology with the effect on microglia’s perimeter being significant compared to vehicle treated females. This could imply a greater immune response of microglia in females and a greater effect of MSC treatment in females 4 weeks post-HI. However, we must emphasize these observations are based mostly on trends as seen in the graphs. Remarkably, at 15 weeks we observed an HI-effect in the cortex of animals that received that DHA-enriched diet, but not in animals that were fed the control diet, while this effect mostly disappeared when we investigated the results for both sexes separately. In contrast, an HI-effect was detected in males that received the control diet. Moreover, in the perilesional region microglia from animals that received the DHA-enriched diet seemed to have a more activated morphological profile than animals that received the control diet. When we investigated the outcomes for each sex separately, this trend seemed to be more apparent in males than in females. Together,
these findings strongly suggest sex is an important factor in both inflammatory responses and treatment outcomes.
On another note, here we analyzed microglia activation based on morphological features, while this analysis has limitations on its own. When microglia get activated, they initially take on an M1 phenotype while later alternative activation leads to an M2 phenotype15,16. Both phenotypes are characterized by enlarged cell bodies and fewer
branches, however their functional properties are different. While M1 is known to execute mostly pro-inflammatory properties, M2 leads to anti-pro-inflammatory signaling15. Moreover, microglia are known to actively
switch between the two phenotypes, depending on inflammatory stimuli8. Solely based on morphological changes,
we cannot determine whether microglia show pro-inflammatory characteristics (M1) or anti-inflammatory characteristics (M2). Therefore, additional strategies are desirable when microglia activation is studied. Detection of specific markers as CD86 and CD206 for M1 and M2, respectively, could be such strategy50.
In contrast to our hypothesis of synapse loss after HI, we did not find significant differences in the levels of synaptophysin between sham and HI-animals, nor was an effect of DHA-enriched nutrition detectable. Notably, previous studies show contradictory findings on the short and long term effects of ischemia on synaptophysin levels38,48,51–54. Yet, it seems that synaptophysin levels strongly depend on factors like type of brain injury, region
of interest and time after insult. Our data could therefore imply that significant synapse loss never happened since the cortex and thalamus might be less susceptible to synapse loss after HI. Or, synapses were already restored in these regions at 15 weeks post-HI. However, the exact reasons for our findings remain unclear which implicates the need of further research on the role of synaptophysin in neonatal HIE.
Looking at the future, there will be most certainly not one perfect therapy to treat neonatal HIE. Different patients ask for different strategies, especially when sex dimorphism is considered. Therefore, to broaden to scope of future treatments it would be interesting to investigate the efficacy of combined therapies. Therapies like MSC treatment and DHA-enriched nutrition can easily be combined and may enhance each other’s effectiveness. Nutrition in particular is a strategy that is easily implemented into the clinic, beside medical nutrition being relatively safe and free of side effects. We therefore suggest that future research on HIE should further elaborate on the role of sex dimorphism as well as the efficacy of combined therapies.
Altogether, both intranasal MSC treatment and DHA-enriched nutrition show to be promising therapeutic strategies to modulate inflammatory responses in the HI-injured brains of neonatal mice. Sexual dimorphism appears to play a role in therapeutic outcome and should therefore be taken into account when such therapies are implemented into the clinic. Since DHA-enriched diets are widely described and acknowledged as neuroprotective strategies, it is recommended that the path to optimizing nutrition is further explored. Also, optimization strategies may further enhance MSC treatment efficacy. Moreover, therapies like these can be easily combined which should definitely be taken in mind in the future.
5.
References
1. Bano, S., Chaudhary, V. & Garga, U. C. Neonatal hypoxic-ischemic encephalopathy: A radiological review. J. Pediatr. Neurosci. 12, 1–6 (2017).
2. Lee, A. C. C. et al. Intrapartum-related neonatal encephalopathy incidence and impairment at regional and global levels for 2010 with trends from 1990. Pediatr. Res. 74, 50–72 (2013).
3. Kurinczuk, J. J., White-Koning, M. & Badawi, N. Epidemiology of neonatal encephalopathy and hypoxic-ischaemic encephalopathy. Early Hum. Dev. 86, 329–338 (2010).
4. Eunson, P. The long-term health, social, and financial burden of hypoxic-ischaemic encephalopathy. Dev. Med. Child Neurol. 57, 48–50 (2015).
5. Arnaez, J. et al. Population-Based Study of the National Implementation of Therapeutic Hypothermia in Infants with Hypoxic-Ischemic Encephalopathy. Ther. Hypothermia Temp. Manag. 8, 24–29 (2018). 6. Douglas-Escobar, M. & Weiss, M. D. Hypoxic-Ischemic Encephalopathy A Review for the Clinician. JAMA
Pediatr. 169, 397–403 (2015).
7. Wassink, G., Gunn, E. R., Drury, P. P., Bennet, L. & Gunn, A. J. The mechanisms and treatment of asphyxial encephalopathy. Front. Neurosci. 8, 1–11 (2014).
8. Liu, F. & Mccullough, L. D. Inflammatory responses in hypoxic ischemic encephalopathy. Acta Pharmacol. Sin. 34, 1121–1130 (2013).
9. Nair, J. & Kumar, V. Current and Emerging Therapies in the Management of Hypoxic Ischemic Encephalopathy in Neonates. Children 5, 99 (2018).
10. Favié, L. M. A. et al. Nitric oxide synthase inhibition as a neuroprotective strategy following hypoxic-ischemic encephalopathy: Evidence from animal studies. Front. Neurol. 9, (2018).
11. Allen, K. A. & Brandon, D. H. Hypoxic Ischemic Encephalopathy: Pathophysiology and Experimental Treatments. Newborn Infant Nurs. Rev. 11, 125–133 (2011).
12. Fleiss, B. & Gressens, P. Tertiary mechanisms of brain damage: A new hope for treatment of cerebral palsy? Lancet Neurol. 11, 556–566 (2012).
13. Adeluyi, A. et al. Microglia morphology and proinflammatory signaling in the nucleus accumbens during nicotine withdrawal. Sci. Adv. 5, 1–12 (2019).
14. Jellema, R. K. et al. Cerebral inflammation and mobilization of the peripheral immune system following global hypoxia-ischemia in preterm sheep. J. Neuroinflammation 10, 1–19 (2013).
15. Ziemka-Nalecz, M., Jaworska, J. & Zalewska, T. Insights into the neuroinflammatory responses after neonatal hypoxia-ischemia. J. Neuropathol. Exp. Neurol. 76, 644–654 (2017).
16. Zhang, L., Zhang, J. & You, Z. Switching of the microglial activation phenotype is a possible treatment for depression disorder. Front. Cell. Neurosci. 12, 1–13 (2018).
17. Iadecola, C. & Anrather, J. The immunology of stroke: From mechanisms to translation. Nat. Med. 17, 796–808 (2011).
18. Siracusa, R., Fusco, R. & Cuzzocrea, S. Astrocytes: Role and functions in brain pathologies. Front. Pharmacol. 10, 1–10 (2019).
19. Pekny, M. & Pekna, M. Astrocyte reactivity and reactive astrogliosis: Costs and benefits. Physiol. Rev. 94, 1077–1098 (2014).
20. Nolan, J. P. et al. European Resuscitation Council and European Society of Intensive Care Medicine Guidelines for Post-resuscitation Care 2015. Section 5 of the European Resuscitation Council Guidelines for Resuscitation 2015. Resuscitation 95, 202–222 (2015).
21. Davidson, J. O., Wassink, G., van den Heuij, L. G., Bennet, L. & Gunn, A. J. Therapeutic hypothermia for neonatal hypoxic-ischemic encephalopathy - Where to from here? Front. Neurol. 6, (2015).
22. Markarian, G. Z., Lee, J. H., Stein, D. J. & Hong, S. C. Mild hypothermia: Therapeutic window after experimental cerebral ischemia. Neurosurgery 38, 542–551 (1996).
23. Azzopardi, D. et al. Effects of Hypothermia for Perinatal Asphyxia on Childhood Outcomes. N. Engl. J. Med. 371, 140–149 (2014).
24. Squillaro, T., Peluso, G. & Galderisi, U. Clinical trials with mesenchymal stem cells: An update. Cell Transplant. 25, 829–848 (2016).
25. Nair, S. et al. Neuroprotection offered by mesenchymal stem cells in perinatal brain injury: Role of mitochondria, inflammation and reactive oxygen species. J. Neurochem. 1–15 (2020).
doi:10.1111/jnc.15267
26. van Velthoven, C. T. J., Kavelaars, A., van Bel, F. & Heijnen, C. J. Mesenchymal stem cell treatment after neonatal hypoxic-ischemic brain injury improves behavioral outcome and induces neuronal and oligodendrocyte regeneration. Brain. Behav. Immun. 24, 387–393 (2010).
27. Sakai, T. et al. Functional recovery after the systemic administration of mesenchymal stem cells in a rat model of neonatal hypoxia-ischemia. J. Neurosurg. Pediatr. 22, 513–522 (2018).
28. Van Velthoven, C. T. J., Kavelaars, A., Van Bel, F. & Heijnen, C. J. 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 (2010).
29. Donega, V. et al. Assessment of long-term safety and efficacy of intranasal mesenchymal stem cell treatment for neonatal brain injury in the mouse. Pediatr. Res. 78, 520–526 (2015).
30. Donega, V. et al. Intranasal Mesenchymal Stem Cell Treatment for Neonatal Brain Damage: Long-Term Cognitive and Sensorimotor Improvement. PLoS One 8, 1–7 (2013).
31. Donega, V. et al. Intranasally administered mesenchymal stem cells promote a regenerative niche for repair of neonatal ischemic brain injury. Exp. Neurol. 261, 53–64 (2014).
32. Keunen, K., Van Elburg, R. M., Van Bel, F. & Benders, M. J. N. L. Impact of nutrition on brain development and its neuroprotective implications following preterm birth. Pediatr. Res. 77, 148–155 (2015).
33. Wang, D. et al. Enhanced neuroprotective effect of DHA and EPA-enriched phospholipids against 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) induced oxidative stress in mice brain. J. Funct. Foods 25, 385–396 (2016).
35. Mayurasakorn, K. et al. DHA but not EPA emulsions preserve neurological and mitochondrial function after brain hypoxia-ischemia in neonatal mice. PLoS One 11, 1–16 (2016).
36. Che, H. et al. Neuroprotective Effects of n-3 Polyunsaturated Fatty Acid-Enriched Phosphatidylserine Against Oxidative Damage in PC12 Cells. Cell. Mol. Neurobiol. 38, 657–668 (2018).
37. Liu, Z. H. et al. A single bolus of docosahexaenoic acid promotes neuroplastic changes in the innervation of spinal cord interneurons and motor neurons and improves functional recovery after spinal cord injury. J. Neurosci. 35, 12733–12752 (2015).
38. Tao, G. et al. Docosahexaenoic Acid Rescues Synaptogenesis Impairment and Long-Term Memory Deficits Caused by Postnatal Multiple Sevoflurane Exposures. Biomed Res. Int. 2016, (2016).
39. Belayev, L. et al. DHA modulates MANF and TREM2 abundance, enhances neurogenesis, reduces infarct size, and improves neurological function after experimental ischemic stroke. CNS Neurosci. Ther. 26, 1155–1167 (2020).
40. Berman, D. R., Mozurkewich, E., Liu, Y. Q. & Barks, J. Docosahexaenoic acid pretreatment confers neuroprotection in a rat model of perinatal cerebral hypoxia-ischemia. Am. J. Obstet. Gynecol. 200, 305.e1-305.e6 (2009).
41. Berman, D. R., Liu, Y., Barks, J. & Mozurkewich, E. Treatment with docosahexaenoic acid after hypoxia-ischemia improves forepaw placing in a rat model of perinatal hypoxia-hypoxia-ischemia. Am. J. Obstet. Gynecol. 203, 385.e1-385.e5 (2010).
42. Desai, A., Kevala, K. & Kim, H. Y. Depletion of brain docosahexaenoic acid impairs recovery from traumatic brain injury. PLoS One 9, (2014).
43. Rice, J. E., Vannucci, R. C. & Brierley, J. B. The influence of immaturity on hypoxic-ischemic brain damage in the rat. Ann. Neurol. 9, 131–141 (1981).
44. Al Mamun, A., Yu, H., Romana, S. & Liu, F. Inflammatory Responses are Sex Specific in Chronic Hypoxic– Ischemic Encephalopathy. Cell Transplant. 27, 1328–1339 (2018).
45. Hill, C. A. & Fitch, R. H. Sex differences in mechanisms and outcome of neonatal hypoxia-ischemia in rodent models: Implications for sex-specific neuroprotection in clinical neonatal practice. Neurol. Res. Int. 2012, 12–14 (2012).
46. Charriaut-Marlangue, C., Besson, V. C. & Baud, O. Sexually dimorphic outcomes after neonatal stroke and hypoxia-ischemia. Int. J. Mol. Sci. 19, (2018).
47. Mirza, M. A., Ritzel, R., Xu, Y., McCullough, L. D. & Liu, F. Sexually dimorphic outcomes and inflammatory responses in hypoxic-ischemic encephalopathy. J. Neuroinflammation 12, 1–10 (2015).
48. Tata, D. A., Dandi, E. & Spandou, E. Expression of synaptophysin and BDNF in the medial prefrontal cortex following early life stress and neonatal hypoxia-ischemia. Dev. Psychobiol. 1–10 (2020). doi:10.1002/dev.22011
49. Teo, J. D., Morris, M. J. & Jones, N. M. Hypoxic postconditioning reduces microglial activation, astrocyte and caspase activity, and inflammatory markers after hypoxia-ischemia in the neonatal rat brain. Pediatr. Res. 77, 757–764 (2015).
50. Zhou, T. et al. Microglia polarization with M1/M2 phenotype changes in rd1 mouse model of retinal degeneration. Front. Neuroanat. 11, 1–11 (2017).
51. Van De Berg, W. D. J. et al. Perinatal asphyxia results in changes in presynaptic bouton number in striatum and cerebral cortex - A stereological and behavioral analysis. J. Chem. Neuroanat. 20, 71–82 (2000).
52. Liu, N., Tong, X., Huang, W., Fu, J. & Xue, X. Synaptic Injury in the Thalamus Accompanies White Matter Injury in Hypoxia/Ischemia-Mediated Brain Injury in Neonatal Rats. Biomed Res. Int. 2019, (2019). 53. Griva, M. et al. Long-term effects of enriched environment following neonatal hypoxia-ischemia on
behavior, BDNF and synaptophysin levels in rat hippocampus: Effect of combined treatment with G-CSF. Brain Res. 1667, 55–67 (2017).
54. Korematsu, K., Goto, S., Nagahiro, S. & Ushio, Y. Changes of immunoreactivity for synaptophysin ('protein p38’) following a transient cerebral ischemia in the rat striatum. Brain Res. 616, 320–324 (1993).
