Received: 2 April 2023 | Revised: 21 May 2023 | Accepted: 30 May 2023

DOI: 10.1111/cns.14308

ORIGINAL ARTICLE

Cerebral glucagon-­like peptide-­1 receptor activation alleviates

traumatic brain injury by glymphatic system regulation in mice

Chuanxiang Lv1 | Shuai Han1 | Zhuang Sha2,3 | Mingqi Liu2,3 | Shiying Dong2,3 |
Chunyun Zhang1 | Zean Li1 | Kang Zhang1 | Shouyong Lu1 | Zhiyang Xu1 | Li Bie1 |
Rongcai Jiang2,3

1
Department of Neurosurgery, The First
Hospital of Jilin University, Changchun, Abstract
China
Aim: We aimed to assess the effects of cerebral glucagon-­like peptide-­1 receptor
2
Department of Neurosurgery, Tianjin
Medical University General Hospital,
(GLP-­1R) activation on the glymphatic system and whether this effect was therapeu-
Tianjin, China tic for traumatic brain injury (TBI).
3
Tianjin Neurological Institute, Key Methods: Immunofluorescence was employed to evaluate glymphatic system func-
Laboratory of Post-­Neuroinjury Neuro-­
repair and Regeneration in Central tion. The blood–­brain barrier (BBB) permeability, microvascular basement membrane,
Nervous System, Tianjin Medical and tight junction expression were assessed using Evans blue extravasation, immu-
University General Hospital, Ministry of
Education, Tianjin, China nofluorescence, and western blot. Immunohistochemistry was performed to assess
axonal damage. Neuronal apoptosis was evaluated using Nissl staining, terminal de-
Correspondence
Rongcai Jiang, Department of oxynucleotidyl transferase-­mediated dUTP nick end labeling (TUNEL) staining, and
Neurosurgery, Tianjin Medical University western blot. Cognitive function was assessed using behavioral tests.
General Hospital, Tianjin; Tianjin
Neurological Institute, Key Laboratory Results: Cerebral GLP-­1R activation restored glymphatic transport following TBI, al-
of Post-­Neuroinjury Neuro-­repair and leviating BBB disruption and neuronal apoptosis, thereby improving cognitive func-
Regeneration in Central Nervous System,
Tianjin Medical University General tion following TBI. Glymphatic function suppression by treatment using aquaporin
Hospital, Ministry of Education, 154 4 inhibitor TGN-­020 abolished the protective effect of the GLP-­1R agonist against
Anshan Road, Helping District, Tianjin
300052, China. cognitive impairment.
Email: [email protected] Conclusion: Cerebral GLP-­1R activation can effectively ameliorate neuropathologi-
Li Bie, Department of Neurosurgery, The cal changes and cognitive impairment following TBI; the underlying mechanism could
First Hospital of Jilin University, No. 1
Xinmin Street, Changchun, Jilin Province
involve the repair of the glymphatic system damaged by TBI.
130021, China.
Email: [email protected] KEYWORDS
cognitive impairment, glucagon-­like peptide-­1 receptor, glymphatic system, traumatic brain
Funding information injury
Science and Technology Development
Plan Project of Jilin Province, China,
Grant/Award Number: 20230204094YY
and YDZJ202201ZYTS001

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium,
provided the original work is properly cited.
© 2023 The Authors. CNS Neuroscience & Therapeutics published by John Wiley & Sons Ltd.

CNS Neurosci Ther. 2023;00:1–13.  wileyonlinelibrary.com/journal/cns | 1

 |

17555949, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cns.14308 by CochraneChina, Wiley Online Library on [02/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2 LV et al.

1 | I NTRO D U C TI O N male mice were purchased from HFK Bioscience Corporation, used
in all the experiments.
Recent studies have demonstrated a brain-­wide network of paravas-
cular pathways known as the glymphatic system, which utilizes peri-
vascular channels formed by astroglial cells surrounding penetrating 2.2 | Fluid percussion injury model
arteries and eliminates interstitial solutes and brain metabolic waste
from the brain by facilitating fluid exchange between the cerebrospinal The fluid percussion injury (FPI) model was prepared as described
fluid (CSF) and interstitial fluid (ISF).1–­3 Aquaporin 4 (AQP4), a water previously. 21 After isoflurane anesthesia, the mice were treated
channel mainly localized to the astrocytes in a polarized manner sur- using an FPI device. The sham group underwent the same process
rounding the perivascular space, plays a significant role in glymphatic except for the strike procedure.
system regulation.4 Glymphatic dysfunction occurs when normal po-
larization of AQP4 is disrupted.1,3,5,6 Amyloid beta and tau protein
clearance from the interstitial spaces is significantly impaired in mice 2.3 | Experiment design and drug administration
knocked out of AQP4 or treated by AQP4-­specific inhibitor TGN-­
020.3,7,8 Several neurological diseases, such as Alzheimer's disease, The animals were randomly assigned to four separate experi-
Parkinson's disease, multiple sclerosis, and traumatic brain injury (TBI) ments (Figure 1). Ex-­4 (HY-­13443, MedChemExpress), a GLP-­1R
are associated with glymphatic dysfunction attributed to perturbation agonist, was dissolved in physiological saline and delivered (10 μg/
9,10
of AQP4 expression or disruption of AQP4 polarization. kg, intravenously) at 1 h following TBI. The GLP-­1R competitive
Traumatic brain injury is caused by external attacks, usually ac- antagonist Exendin-­(9-­39) (Ex-­9, HY-­P 0264, MedChemExpress)
companied by serious neuropathological changes and cognitive im- was dissolved in physiological saline and was delivered (50 μg/
pairment. Studies have demonstrated that the glymphatic system kg, intravenously) 20 min before Ex-­4 administration. TGN-­020
7,11
is severely damaged following TBI. After TBI, glial fibrillary acidic (HY-­W008574, MedChemExpress), a specific inhibitor of AQP4,
protein (GFAP) positive astrocytes surround the ipsilateral hemisphere which was used as an effective means for pharmacological block-
overproliferated and activated to form a consolidated glial scar, which ing of the glymphatic system, was dissolved in a cyclodextrin de-
alters the polarization and localization of AQP4. These disturbances rivative, 20% sodium sulphobutylether-­β -­c yclodextrin (C871854,
following TBI lead to decreased fluid exchange and waste clearance in Macklin) in water for injection as described previously. 8 TGN-­020
the glymphatic system, resulting in the accumulation of neurotoxic sub- was delivered (250 mg/kg, intraperitoneally) 15 min before Ex-­4
stances, such as amyloid beta and tau protein aggregates, which fur- administration.
ther increases the risk of developing neurodegenerative diseases.7,11–­13
Glucagon-­like peptide-­1 (GLP-­1) is an important peptide hormone,
which possesses a variety of functions based on its receptor location, 2.4 | Intracisternal tracer infusions
from regulating insulin secretion to controlling satiety and modulating
autonomic nervous system activity.14–­16 GLP-­1R in the brain is widely A fluorescent CSF tracer (rhodamine B isothiocyanate-­dextran;
distributed in the cerebral cortex, hippocampus, hypothalamus, and R9379, Sigma Aldrich, 70 KDa, RITC-­dextran) was dissolved in
brain stem. This widespread cerebral distribution of GLP-­1R confers artificial CSF at a concentration of 0.5%. First, the mice were an-
GLP-­1 the potential to regulate multiple neurological and cognitive func- esthetized and fixed in a stereotactic frame, followed by surgical
tions.17 However, after entering the extracellular space, endogenous exposure of the cisterna magna. The CSF tracer was subsequently
GLP-­1 is quickly inactivated by dipeptidyl peptidase-­4 in a short time, infused into the subarachnoid CSF via cisterna magna puncture at
which restricts its pharmacological use.18 Thus, Exendin-­4 (Ex-­4), an an- a rate of 1 μL/min for 10 min (10 μL total volume) through a syringe
alog of GLP-­1 with structural modifications was synthesized. Emerging pump (KDS LEGATO 130, RWD Life Science). Thereafter, the anes-
evidence from several studies has revealed that the activation of GLP-­1R thetized mice were transcardially perfused and fixed with 4% para-
signaling protects the brain against neuroinflammation, oxidative stress, formaldehyde (PFA) 30 min following the start of the infusion. The
and neurotoxicity in multiple disease models.15,18–­20 Thus, we hypothe- brains were subsequently dissected and post-­f ixed in 4% PFA for
sized that cerebral GLP-­1R activation could ameliorate TBI-­induced neu- 24 h before being sliced into 100-­μm coronal sections. Tracer influx
rological impairment by restoration of the glymphatic system. into the brain was imaged ex vivo using whole-­slice conventional
fluorescence microscopy (Olympus BX61) and quantified using
ImageJ software (version 1.53, National Institutes of Health) as de-
2 | M ATE R I A L S A N D M E TH O DS scribed previously.7

2.1 | Animals
2.5 | Intraparenchymal injections
All the animal experimental procedures described in this study were
approved by the Animal Care and Use Committee of Tianjin Medical To evaluate the interstitial metabolite clearance pathway, a fluores-
University General Hospital, Tianjin, China; 8–­12-­week-­old C57BL/6 cent CSF tracer was stereotactically microinjected into the cerebral
 |

17555949, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cns.14308 by CochraneChina, Wiley Online Library on [02/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
LV et al. 3

F I G U R E 1 Basic characteristics of the study. (A) Experiment 1. Effects of cerebral GLP-­1R activation on the glymphatic system 3 days
following TBI. (B) Experiment 2. Effects of cerebral GLP-­1R activation on the BBB 3 days following TBI. (C) Experiment 3. Effects of cerebral
GLP-­1R activation on axonal injury and neuronal apoptosis 3 days following TBI. (D) Experiment 4–­1. Effects of TGN-­020 on glymphatic
transport function. (E) Experiment 4–­2. Effects of cerebral GLP-­1R activation on cognitive function following TBI. EB, Evans blue; Ex-­4,
Exendin-­4; Ex-­9, Exendin-­(9-­39); GLP-­1R, glucagon-­like peptide-­1 receptor; MWM test, Morris water maze test; NOR test, novel object
recognition test; TBI, traumatic brain injury; TUNEL staining, terminal deoxynucleotidyl transferase-­mediated dUTP nick end labeling staining.

cortex of the mice. A 33-­gauge stainless steel cannula (Hamilton) immunoreactivity than the perivascular endfeet (polarization = 100−
was inserted at the following stereotactic coordinates within the AQP4 percentage area).
cerebral cortex: 2.00 mm posterior, 1.50 mm lateral to the bregma,
and 2.00 mm below the brain surface. 30 min following the cannula
insertion, 500 nL of the CSF tracer was infused over 10 min. One 2.7 | Evans blue extravasation
hour later, the animal was rapidly perfusion fixed. Tracer efflux was
quantified as described above. Evans blue (EB) extravasation was used to assess BBB permeability
as described previously. 22 Briefly, EB solution (4 mL/kg, 2% in saline;
E2129, Sigma Aldrich) was administered via the tail vein and circu-
2.6 | AQP4 polarization evaluation lated for 2 h. After incubation with methylamide, the samples were
centrifuged. Supernatant absorbance was measured using a spectro-
AQP4 polarization was evaluated 3 days following TBI as described fluorophotometer to quantify EB concentration.
previously.5,8 Briefly, the median immunofluorescence intensity
of the perivascular regions was measured. Thereafter a threshold
analysis measured the percentage of the region exhibiting AQP4 2.8 | HE and Nissl staining
immunofluorescence greater than or equal to perivascular AQP4
immunofluorescence (AQP4 percentage area). Polarization was ex- Mouse brain tissue was fixed with 4% PFA, followed by dehydra-
pressed as the percentage of the region that exhibited lower AQP4 tion in series concentrations of ethanol. The paraffin-­embedded
 |

17555949, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cns.14308 by CochraneChina, Wiley Online Library on [02/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4 LV et al.

brain tissues were subsequently cut into 8-­μm slices for HE (G1120, staining. The percentage of TUNEL-­positive neurons was quantified
Solarbio) and Nissl (G1432, Solarbio) staining. HE and Nissl staining using a fluorescence microscope and ImageJ software.
were performed as described previously. 23 The sections were ob-
served under a light microscope (IX73, Olympus Corporation).
2.13 | Behavioral tests

2.9 | Immunohistochemical staining The Morris water maze (MWM) test was conducted based on a pre-
vious protocol to test the spatial learning and memory of the mice. 24
Immunohistochemistry (IHC) was used to detect amyloid precursor The novel object recognition (NOR) test was performed to assess
24
protein (APP) as previously described. After incubation in 0.3% short-­term memory 30 days following TBI, according to previous
H2O2 for 30 min, the antigen was retrieved by boiling the sections methods. 26,27
in citrate buffer. Thereafter, 1% bovine serum was added to the
sections for 30 min. The sections were incubated with primary rab-
bit anti-­APP antibody (1:500, ab32136, Abcam) overnight at 4°C. 2.14 | Statistical analysis
Secondary biotinylated goat anti-­rabbit antibody (GK500705, Gene
Tech) was subsequently applied for 1 h at room temperature. The All the statistical analyses were performed using SPSS 26.0 software
sections were observed under a light microscope. (IBM Corporation). Data are presented as mean ± standard deviation
(SD). The Shapiro–­Wilk test was used to assess the normality of the
data distribution. Statistical differences among the two groups were
2.10 | Immunofluorescence analyzed by unpaired two-­t ailed Student's t-­test. One-­way or two-­
way analysis of variance (ANOVA) with Tukey's post hoc test was
Immunofluorescence analysis was performed as described previ- performed for comparison among multiple groups. Statistical signifi-
ously.7 The tissue sections were incubated with the following pri- cance was set at p < 0.05.
mary antibodies: AQP4 (1:500, 59678, Cell Signaling Technology),
GFAP (1:500, 3670, Cell Signaling Technology), CD31 (1:500,
AF3628, R&D Systems), NeuN (1:500, ab177487, Abcam), and col- 3 | R E S U LT S
lagen IV (1:500, ab19808, Abcam). The slices were subsequently
incubated with species-­specific fluorescence-­conjugated secondary 3.1 | Cerebral GLP-­1R activation treatment
antibodies. Finally, the sections were observed under a fluorescence attenuated glymphatic system dysfunction 3 days
microscope and assessed using ImageJ software. following TBI

Glymphatic CSF-­ISF exchange has been demonstrated to play a

2.11 | Western blot central role in waste clearance. 28 However, this fluid exchange
function is severely compromised following TBI.7 To evaluate the
25
Western blot was performed as previously described. A protein therapeutic effect of cerebral GLP-­1R activation on the paravascu-
sample was collected from the TBI area of the brain tissue. The poly- lar CSF penetration following TBI, 10 μL of fluorescent CSF tracer
vinylidene difluoride membranes (PVDF, Millipore) were blocked was cautiously injected into the subarachnoid CSF of the cisterna
with 5% skim milk and incubated at 4°C overnight with the fol- magna. 30 min after injection, the brains were perfusion fixed. CSF
lowing primary antibodies: ZO-­1 (1:1000, 61-­7300, ThermoFisher tracer penetration into the brain parenchyma was evaluated using a
Scientific), claudin-­5 (1:1000, 35-­2500, ThermoFisher Scientific), oc- conventional fluorescence microscope. Compared with sham mice,
cludin (1:1000, 27260-­1-­AP, Proteintech), caspase-­3 (1:1000, A0214, CSF tracer penetration into the brain with TBI was significantly de-
ABclonal), Bcl2 (1:1000, A0208, ABclonal), and β-­actin (1:1000, TA-­ creased. Ex-­4 treatment significantly increased perivascular influx
09, ZSGB-­BIO) antibodies. The membranes were subsequently in- of the CSF tracer. However, when Ex-­9 was administered for 20 min
cubated with secondary antibodies for 1 h. The blots were obtained beforehand, improvement in CSF penetration attributed to Ex-­4 was
with a ChemiDoc Touch Imaging System and quantified using ImageJ not observed (Figure 2A–­C).
software. To evaluate the therapeutic effect of cerebral GLP-­1R activa-
tion on interstitial solute clearance post-­TBI, 0.5 μL of fluorescent
CSF tracer was stereotactically microinjected into the parenchymal
2.12 | TUNEL staining cortex around the lesion. One hour after injection, the brains were
perfusion fixed. The function of interstitial solute clearance was
Apoptotic cells were detected by TUNEL assay using the TUNEL assay measured by conventional fluorescence microscopy, comparing the
kit (G3250, Promega) according to the manufacturer's instructions. residual amount of fluorescent tracer in the brain tissue of mice in
Cortical neurons were labeled using NeuN immunofluorescence each group; the residual fluorescent tracer in the brain tissue with
 |

17555949, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cns.14308 by CochraneChina, Wiley Online Library on [02/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
LV et al. 5

F I G U R E 2 Cerebral GLP-­1R activation treatment attenuated glymphatic system dysfunction 3 days following TBI. (A) schematic shows that
fluorescent tracer was injected into the cisterna magna or brain parenchyma after cerebral GLP-­1R activation treatment 3 days following TBI. (B)
Representative brain sections stained for nuclei (DAPI; blue) and RITC-­dextran (red) influx into the brain parenchyma of mice from four groups.
Scale bar, 1500 μm. (C) Quantification of the percentage of RITC-­dextran covered area fraction in brain sections in (B). n = 6 mice per group.
(D) Representative brain sections stained for nuclei (DAPI; blue) and RITC-­dextran (red) clearance from the brain parenchyma of mice from
four groups. Scale bar, 1500 μm. (E) Coimmunofluorescence staining for AQP4 (green) and GFAP (red) around the lesion. Scale bar, 100 μm. (F)
Coimmunofluorescence staining for AQP4 (green) and CD31 (red) around the lesion. Scale bar, 100 μm. (G) Quantification of the percentage of
residual RITC-­dextran covered area fraction in brain sections in (D). n = 6 mice per group. (H) Quantification of the percentage of GFAP covered area
fraction in brain sections. n = 6 mice per group. (I) Quantification of the AQP4 polarization of mice from four groups. n = 6 mice per group. All data are
shown as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. AQP4, aquaporin 4; Ex-­4, Exendin-­4; Ex-­9, Exendin-­(9-­39); GFAP, glial fibrillary
acidic protein; GLP-­1R, glucagon-­like peptide-­1 receptor; RITC-­dextran, rhodamine B isothiocyanate-­dextran; TBI, traumatic brain injury.
 |

17555949, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cns.14308 by CochraneChina, Wiley Online Library on [02/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
6 LV et al.

TBI was significantly increased and Ex-­4 treatment effectively re- 3.3 | Cerebral GLP-­1R activation treatment
duced the fluorescent tracer residue. When Ex-­9 was administered attenuated axonal injury and neuronal apoptosis
previously, improvement in the interstitial solute clearance induced 3 days following TBI
by Ex-­4 disappeared (Figure 2A,D,G).
Previous studies have demonstrated that paravascular CSF re- Previous studies have demonstrated that axonal injury following TBI
circulation and ISF solute clearance are dependent upon the polar- increases metabolic waste products in the brain, resulting in an in-
ized astroglial AQP4 water channel, which facilitates fluid movement creased burden on the glymphatic system.7 Immunohistochemical
5,28
between the perivascular and interstitial spaces. Widespread staining with an anti-­APP antibody, which has previously been
reactive astrogliosis post-­TBI is closely associated with the loss of shown to accumulate in injured axons, was performed 3 days fol-
5,7
perivascular AQP4 polarization. First, we evaluated reactive as- lowing TBI. 24 Representative images demonstrated that TBI caused
trogliosis 3 days following TBI by immunofluorescence staining for a significant increase in the APP accumulation compared to that in
GFAP and AQP4. GFAP expression was significantly higher in the sham mice. Cerebral GLP-­1R activation effectively alleviated APP
TBI + vehicle group than that in the sham group. After Ex-­4 treat- deposition in the cortex around the lesion (Figure 4A,B). We sub-
ment, GFAP expression was reduced. However, when Ex-­9 was sequently determined the neuroprotective effect of GLP-­1R activa-
administered beforehand, GFAP expression remained high even tion on TBI. HE staining was performed to evaluate the histological
following Ex-­4 treatment (Figure 2E,H). We subsequently evaluated morphology of the cerebral cortex. As shown in the representative
the AQP4 polarization by immunofluorescence staining of CD31 and picture (Figure 4A), Ex-­4 administration significantly alleviated TBI-­
AQP4. Consistent with previous results, AQP4 expression in the induced cerebral cortex damage and loss as well as traumatic cer-
perivascular extremity was significantly reduced 3 days following ebral parenchymal hemorrhage. However, this protective effect was
TBI. AQP4 polarization was effectively restored by Ex-­4 treatment. not observed in the TBI + Ex-­9 + Ex-­4 group. To assess the influence
However, when Ex-­9 was administered 20 min beforehand, Ex-­4 of GLP-­1R activation on neuronal death and degeneration follow-
treatment did not exert a protective effect against AQP4 polariza- ing TBI, we performed TUNEL and Nissl staining of the damaged
tion (Figure 2F,I). brain tissue of TBI mice 3 days following injury. As indicated by the
representative images (Figure 4A,C,D,E), cerebral GLP-­1R activation
effectively controlled TBI-­induced neuronal degeneration and ap-
3.2 | Cerebral GLP-­1R activation treatment optosis, thereby increasing the survival of neurons 3 days following
attenuated BBB disruption 3 days following TBI TBI. Consistent with the staining assays described above, cerebral
GLP-­1R activation following TBI upregulated Bcl-­2 expression and
Similar to the glymphatic system, the BBB plays an important role in downregulated cleaved caspase-­3 expression in the cortex of the
brain fluid exchange and substance clearance, which is regulated by TBI group, as shown in the western blot analysis (Figure 4F–­H).
AQP4. Thus, we examined the effect of cerebral GLP-­1R activation The distributed dispersion of GLP-­1R increases the complexity
on the BBB function. EB extravasation was employed to assess BBB of its function. Weight loss and hypoglycemia associated with Ex-­4
permeability. Three days following TBI, we observed that errhysis should be considered during TBI treatment. Body weight loss in all the
and EB dye extravasation in the ipsilateral hemisphere in the TBI groups was calculated using the body weight before surgery and 3 days
group were significantly increased compared to those in the sham following TBI. The body weight of each group decreased to varying de-
group, indicating severe BBB breakdown. However, the increased EB grees, with no significant difference in the weight loss between the TBI
levels and errhysis were markedly attenuated by EX-­4 treatment. No groups (Figure 4I). Changes in the blood glucose levels were monitored
obvious discrepancies in the EB levels or errhysis were observed be- until 24 h following Ex-­4 administration. The blood glucose levels in
tween the TBI + vehicle and TBI + Ex-­9 + Ex-­4 groups (Figure 3A,B). each group significantly increased owing to the stress response to sur-
Collagen IV and CD31 immunofluorescence staining was performed gery, with no significant differences among the TBI groups (Figure 4J).
to detect the integrity of the basement membrane of the microves-
sels as described previously. 29 Collagen IV basement membrane cov-
erage of CD31 microvessels decreased significantly 3 days following 3.4 | Cerebral GLP-­1R activation treatment can
TBI. Cerebral GLP-­1R activation effectively alleviated damage to the effectively alleviate cognitive impairment following
microvessel basement membrane (Figure 3C,D). Tight junction pro- TBI, which can be reversed by pharmacological
teins are essential for maintaining the BBB. Therefore, tight junction blocking of the glymphatic system
protein (ZO-­1, occludin, and claudin-­5) expression was quantified
using Western blot 3 days following TBI. After TBI, GLP-­1R acti- Previously, it was reported that astrocytic AQP4 polarization could
vation effectively reversed the reduction and largely maintained be interrupted by TGN-­020. Therefore, TGN-­020 can be used for
the protein levels of ZO-­1, occludin, and claudin-­5 (Figure 3E–­H). effective pharmacological blocking of the glymphatic system.8 Prior
Overall, these results indicate that TBI-­induced BBB disruption can to its blocking effect, we examined its effect on the glymphatic
be salvaged by cerebral GLP-­1R activation. transport of RITC-­dextran. As shown in the representative images
 |

17555949, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cns.14308 by CochraneChina, Wiley Online Library on [02/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
LV et al. 7

F I G U R E 3 Cerebral GLP-­1R activation treatment attenuated BBB disruption 3 days following TBI. (A) Representative images of the EB
extravasation assay. Scale bar, 0.5 cm (B) Quantification of EB concentration for each group. n = 6 mice per group. (C) Coimmunofluorescence
staining for Collagen IV (green) and CD31 (red) around the lesion. Scale bar, 100 μm. (D) Quantification of Collagen IV + basement membrane
coverage on CD31 microvessels. n = 6 mice per group. (E) Representative Western blots of ZO-­1, occludin and claudin-­5. (F–­H) Quantification
of relative protein expression normalized to the optical density of β-­actin. n = 6 mice per group. All data are shown as mean ± SD. *p < 0.05,
**p < 0.01, ***p < 0.001, ****p < 0.0001. BBB, blood–­brain barrier; Col IV, Collagen IV; EB, Evans blue; Ex-­4, Exendin-­4; Ex-­9, Exendin-­(9-­39);
GLP-­1R, glucagon-­like peptide-­1 receptor; TBI, traumatic brain injury.

(Figure 5A–­E), the glymphatic transport of RITC-­dextran was effec- NOR tests. In the MWM test, the escape latency of the TBI + ve-
tively blocked by TGN-­020. hicle group was greater than that of the sham group, and Ex-­4
To study the effect of cerebral GLP-­1R activation on impaired administration decreased escape latency compared to the TBI + ve-
cognitive function following TBI, we performed the MWM and hicle group on day 18 and day 19 following TBI. When TGN-­020
 |

17555949, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cns.14308 by CochraneChina, Wiley Online Library on [02/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
8 LV et al.

was administered 15 min beforehand, Ex-­4 administration failed TBI + Vehicle group on day 19 (Figure 6D). No significant differ-
to decrease escape latency (Figure 6A–­C). The distance traveled ence was observed in the swimming speed between all the groups
to the platform was shorter in the TBI + Ex-­4 group than in the (Figure 6E). Thereafter, we removed the hidden platform on day
 |

17555949, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cns.14308 by CochraneChina, Wiley Online Library on [02/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
LV et al. 9

F I G U R E 4 Cerebral GLP-­1R activation treatment attenuated axonal injury and neuronal apoptosis 3 days following TBI. (A) The
representative images of APP immunohistochemistry, H&E and Nissl staining of ipsilateral cerebral cortex 3 days following TBI from four
groups. Scale bar, 100 μm. (B) Quantitative analysis of APP accumulation. n = 6 mice per group. (C) Quantification of Nissl staining for
neuronal loss analysis at 3 days following TBI. n = 6 mice per group. (D) TUNEL assay (green) and costaining of NeuN (red) in injured cortex.
TUNEL+/NeuN+ cells are apoptotic neurons. Scale bar, 50 μm. (E) Count of apoptotic neurons in the cortex for each group. n = 6 mice per
group. (F) Representative Western blots of cleaved caspase-­3, and Bcl-­2. (G, H) Quantification of relative protein expression normalized
to the optical density of β-­actin. n = 6 mice per group. (I) The effects of cerebral GLP-­1R activation on body weight loss at 3 days following
TBI. n = 10 mice per group (J) The effects of cerebral GLP-­1R activation on blood glucose change at 24 h following Ex-­4 administration.
n = 10 mice per group. All data are shown as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. APP, amyloid precursor protein;
Ex-­4, Exendin-­4; Ex-­9, Exendin-­(9-­39); GLP-­1R, glucagon-­like peptide-­1 receptor; HE staining, Hematoxylin and Eosin staining; IHC staining,
Immunohistochemistry staining; TBI, traumatic brain injury; TUNEL (staining), terminal deoxynucleotidyl transferase-­mediated dUTP nick
end labeling (staining).

20 following TBI to evaluate the number of crossings at the plat- destruction following TBI.7 In previous studies, Ex-­4 was able to
form site, and a significant increase in the crossing number was lower amyloid beta (1–­42) induced oxidative stress and inflamma-
observed in the TBI + Ex-­4 group compared to that in the TBI + ve- tion in astrocytes, demonstrating excellent neuroprotective ef-
hicle group. When TGN-­020 was administered beforehand, no im- fects.18 This could promote a reasonable hypothesis for cerebral
provement was observed in the number of crossings treated with GLP-­1R activation to improve the function of glymphatic system.
Ex-­4 (Figure 6F). Furthermore, the TBI + vehicle group spent less Our results demonstrate that cerebral GLP-­1R activation can effec-
time in the target region than the sham group, and Ex-­4 treatment tively ameliorate the reactive astrogliosis and loss of perivascular
ameliorated this phenomenon. No significant differences were AQP4 polarization following TBI, thereby improving the transport
observed between the TBI + vehicle and TBI + TGN-­020 + Ex-­4 function of glymphatic system.
groups (Figure 6G). In the NOR test, the TBI + vehicle group spent In addition to the glymphatic pathway, the BBB also plays an
a lower percentage of time around the novel object than the sham important role in facilitating the transmission of neuroactive agents
group. Ex-­4 treatment significantly increased the percentage of to maintain homeostasis of the brain microenvironment.31 BBB
time around the novel object. TGN-­020 administration before Ex-­4 destruction, which occurs within hours or days following injury, is
slightly counterbalanced this improvement, though not statistically closely related to edema, neuroinflammation, and cell death.32 We
significantly (Figure 6H–­J). Collectively, these data demonstrated demonstrated that cerebral GLP-­1R activation ameliorated TBI-­
that Ex-­4 treatment has a neuroprotective effect on the neurolog- induced decrease of tight junction protein and microvessel cover-
ical function following TBI, which can be reversed by TGN-­020 ad- age of collagen IV basement membrane, protecting the integrity and
ministered beforehand. function of the BBB.
Traumatic brain injury is an established risk factor for progres-
sive neurodegenerative diseases. Following TBI-­induced axonal in-
4 | DISCUSSION jury, high levels of tau are released into the interstitial fluid. Axonal
injury following TBI is considered to play a central role in the gen-
Previous studies have indicated that the glymphatic system is eration of proteinopathies of hyperphosphorylated tau and amyloid
closely related to a variety of central nervous system diseases.9,28 beta. A considerable portion of interstitial monomeric tau is cleared
The glymphatic influx and solute clearance function were seri- from the cortex by the glymphatic system.7,33 APP immunohisto-
ously impaired following TBI, resulting in the accumulation of neu- chemistry demonstrated that cerebral GLP-­1R activation effectively
rotoxic substances, such as amyloid beta and tau protein in the alleviated the axonal injury caused by TBI, thereby reducing glym-
brain.7 phatic system burden. Apoptosis is considered a key mechanism of
GLP-­1 was originally identified as a 31-­amino acid polypeptide cell death following TBI.34 Our results demonstrated that cerebral
secreted by L cells in the small intestine. It is also produced in the GLP-­1R activation effectively reduced TBI-­induced axonal injury and
brain and plays neuroprotective roles by activating the GLP-­1 recep- neuronal apoptosis.
tor.18 Studies have demonstrated that the number of amyloid beta Cognitive dysfunction is a common adverse consequence of
and dense-­core plaques in the cortex of Alzheimer's mice model TBI, which can reduce the patients' quality of life and affect their
APP/PS1 mice treated with liraglutide (GLP-­1R agonist) decreased social reintegration.35 GLP-­1R activation in the brain improves cog-
by 40%–­50%, while the level of soluble amyloid oligomers decreased nitive performance in mice following TBI. However, when the glym-
by 25%.30 Our results demonstrated that cerebral GLP-­1R activation phatic system was blocked beforehand, this improvement was not
significantly increased paravascular fluid exchange and attenuated observed, suggesting that the glymphatic system could be the key
glymphatic system destruction post-­TBI. pathway through which GLP-­1R exerts its neuroprotective effect.
Studies have demonstrated that mild-­to-­moderate reactive Owing to the complexity of the GLP-­1R pathway function, we fo-
astrogliosis and the accompanying loss of perivascular AQP4 po- cused on its possible side effects other than neuroprotection, such
larization are important factors attributed to glymphatic pathway as hypoglycemia and weight loss, to ensure medication safety.
 |

17555949, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cns.14308 by CochraneChina, Wiley Online Library on [02/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
10 LV et al.

F I G U R E 5 The glymphatic transport is pharmacological blocked by TGN-­020 (A) schematic shows that fluorescent tracer was injected
into the cisterna magna or brain parenchyma after TGN-­020 or vehicle (cyclodextrin derivative) treatment. (B) Representative brain sections
stained for nuclei (DAPI; blue) and RITC-­dextran (red) influx into the brain parenchyma of mice from two groups. Scale bar, 1500 μm. (C)
Representative brain sections stained for nuclei (DAPI; blue) and RITC-­dextran (red) clearance from the brain parenchyma of mice from two
groups. Scale bar, 1500 μm. (D) Quantification of the percentage of RITC-­dextran covered area fraction in brain sections in (B). n = 6 mice
per group. (E) Quantification of the percentage of residual RITC-­dextran covered area fraction in brain sections in (C). n = 6 mice per group.
All data are shown as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. CSF, cerebrospinal fluid; RITC-­dextran, rhodamine B
isothiocyanate-­dextran.

GLP-­1R agonists stimulated insulin secretion in a glucose-­dependent (TRPV4) channel, calmodulin (CaM), cyclic AMP (cAMP), and protein
manner. Thus, GLP-­1R agonists stimulate insulin secretion only in kinase A (PKA) were involved in the phosphorylation and subcellular
the cases of elevated blood glucose.36,37 Our results demonstrate relocation of AQP4.39 Therefore, CaM and PKA could be possible
that Ex-­4 (10 μg/kg, administered intravenously), while exerting neu- targets of cerebral GLP-­1R activation for glymphatic system regula-
roprotective effects, does not cause significant disturbance to body tion following TBI.
weight and blood glucose. Our study has certain limitations. First, we administered a single
The loss of perivascular AQP4 polarization following TBI con- intravenous injection to explore the neuroprotective effects against
tributed to the impairment of glymphatic pathway function.5,7 AQP4 TBI. In future studies, subcutaneously embedded microosmotic
phosphorylation has been shown to play an important role in AQP4 pumps should be used for continuous administration to better sim-
trafficking and subcellular localization.38 Previous research shows ulate endogenous GLP-­1 secretion. Second, the molecular mecha-
that acute hypoxia leads to subcellular relocalization of AQP4 in pri- nism of GLP-­1R in the glymphatic system should be clarified in future
mary cortical astrocytes. The transient receptor potential vanilloid 4 studies.
 |

17555949, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cns.14308 by CochraneChina, Wiley Online Library on [02/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
LV et al. 11

F I G U R E 6 Cerebral GLP-­1R activation treatment can effectively alleviate cognitive impairment following TBI, which can be reversed
by pharmacological blocking of glymphatic system. (A) Schematic representation of the method and process for the MWM test. (B)
Representative thermal imaging of the probe trial and swimming traces for the MWM test. (C–­G) Data from the MWM test was analyzed to
evaluate the spatial learning and memory ability of mice following TBI. n = 10 mice per group (H) Schematic representation of the method
and process for the NOR test. (I) Representative thermal imaging of the probe trial for the NOR test. (J) Time spent with the novel object was
analyzed to evaluate the memory ability of mice following TBI. n = 10 mice per group. All data are shown as mean ± SD. *p < 0.05, **p < 0.01,
***p < 0.001, ****p < 0.0001. Ex-­4, Exendin-­4; Ex-­9, Exendin-­(9-­39); GLP-­1R, glucagon-­like peptide-­1 receptor; MWM test, Morris water maze
test; NOR test, novel object recognition test; TBI, traumatic brain injury.

5 | CO N C LU S I O N S AU T H O R C O N T R I B U T I O N S
Chuanxiang Lv, Li Bie, and Rongcai Jiang conceived and designed the
This study suggests that cerebral GLP-­1R activation exerts a neu- study. Chuanxiang Lv and Shuai Han wrote the manuscript, which
roprotective effect by ameliorating glymphatic system damage was revised by Li Bie and Rongcai Jiang, approved by all the authors.
following TBI, thereby promoting recovery from TBI. Therefore, Zhuang Sha, Mingqi Liu and Shiying Dong performed model prepara-
GLP-­1R could be a promising pharmacotherapeutic target for TBI. tion and CSF tracer infusions. Chunyun Zhang and Zean Li performed
 |

17555949, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cns.14308 by CochraneChina, Wiley Online Library on [02/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
12 LV et al.

immunostaining. Kang Zhang and Shouyong Lu performed behavio- 14. Hölscher C. Protective properties of GLP-­1 and associated pep-
tide hormones in neurodegenerative disorders. Br J Pharmacol.
ral tests. Zhiyang Xu performed statistical analyses of the data.
2022;179(4):695-­714.
15. Kopp KO, Glotfelty EJ, Li Y, Greig NH. Glucagon-­like peptide-­1
AC K N OW L E D G M E N T S (GLP-­1) receptor agonists and neuroinflammation: implica-
We are grateful to the staff of Tianjin Neurological Institute for their tions for neurodegenerative disease treatment. Pharmacol Res.
2022;186:106550.
service.
16. Laurindo LF, Barbalho SM, Guiguer EL, et al. GLP-­1a: going beyond
traditional use. Int J Mol Sci. 2022;23(2):739.
F U N D I N G I N FO R M AT I O N 17. Muscogiuri G, DeFronzo RA, Gastaldelli A, Holst JJ. Glucagon-­like
Science and Technology Development Plan Project of Jilin Province, Peptide-­1 and the central/peripheral nervous system: crosstalk in
diabetes. Trends Endocrinol Metab. 2017;28(2):88-­103.
China under Grant (20230204094YY) and (YDZJ202201ZYTS001).
18. Diz-­Chaves Y, Mastoor Z, Spuch C, González-­Matías LC, Mallo F.
Anti-­inflammatory effects of GLP-­1 receptor activation in the brain
C O N F L I C T O F I N T E R E S T S TAT E M E N T in neurodegenerative diseases. Int J Mol Sci. 2022;23(17):9583.
The authors declare that they have no competing interests. 19. Yassine HN, Solomon V, Thakral A, et al. Brain energy failure in de-
mentia syndromes: opportunities and challenges for glucagon-­like
peptide-­1 receptor agonists. Alzheimers Dement. 2021;18:478-­497.
DATA AVA I L A B I L I T Y S TAT E M E N T
20. Kabahizi A, Wallace B, Lieu L, et al. Glucagon-­like peptide-­1 (GLP-­1)
The data that support the findings of this study are available from signalling in the brain: from neural circuits and metabolism to ther-
the corresponding author upon reasonable request. apeutics. Br J Pharmacol. 2022;179(4):600-­624.
21. Zhou Y, Cai W, Zhao Z, et al. Lactadherin promotes microvesicle
clearance to prevent coagulopathy and improves survival of severe
ORCID
TBI mice. Blood. 2018;131(5):563-­572.
Chuanxiang Lv https://orcid.org/0009-0004-4454-3776 22. Hu J, Wang X, Chen X, et al. Hydroxychloroquine attenuates neu-
Zhuang Sha https://orcid.org/0000-0002-8110-1481 roinflammation following traumatic brain injury by regulating the
Chunyun Zhang https://orcid.org/0000-0002-4284-8064 TLR4/NF-­κB signaling pathway. J Neuroinflammation. 2022;19(1):71.
23. Yang L, Tao Y, Luo L, Zhang Y, Wang X, Meng X. Dengzhan Xixin in-
jection derived from a traditional Chinese herb Erigeron breviscapus
REFERENCES ameliorates cerebral ischemia/reperfusion injury in rats via modu-
1. Iliff JJ, Nedergaard M. Is there a cerebral lymphatic system? Stroke. lation of mitophagy and mitochondrial apoptosis. J Ethnopharmacol.
2013;44(6 Suppl 1):S93-­S95. 2022;288:114988.
2. Jessen NA, Munk AS, Lundgaard I, Nedergaard M. The glymphatic 24. Gao C, Qian Y, Huang J, et al. A three-­day consecutive fingoli-
system: a Beginner's guide. Neurochem Res. 2015;40(12):2583-­2599. mod administration improves neurological functions and mod-
3. Iliff JJ, Wang M, Liao Y, et al. A paravascular pathway facili- ulates multiple immune responses of CCI mice. Mol Neurobiol.
tates CSF flow through the brain parenchyma and the clear- 2017;54(10):8348-­8360.
ance of interstitial solutes, including amyloid β. Sci Transl Med. 25. Liu XL, Sun DD, Zheng MT, et al. Maraviroc promotes recovery from
2012;4(147):147ra111. traumatic brain injury in mice by suppression of neuroinflammation
4. Peng S, Liu J, Liang C, Yang L, Wang G. Aquaporin-­4 in glymphatic and activation of neurotoxic reactive astrocytes. Neural Regen Res.
system, and its implication for central nervous system disorders. 2023;18(1):141-­149.
Neurobiol Dis. 2023;179:106035. 26. Barrett RM, Malvaez M, Kramar E, et al. Hippocampal focal knock-
5. Kress BT, Iliff JJ, Xia M, et al. Impairment of paravascular clearance out of CBP affects specific histone modifications, long-­term po-
pathways in the aging brain. Ann Neurol. 2014;76(6):845-­861. tentiation, and long-­term memory. Neuropsychopharmacology.
6. Munk AS, Wang W, Bèchet NB, et al. PDGF-­B is required for devel- 2011;36(8):1545-­1556.
opment of the glymphatic system. Cell Rep. 2019;26(11):2955-­2969. 27. Friedman-­Levi Y, Liraz-­Zaltsman S, Shemesh C, et al. Pharmacological
e2953. blockers of CCR5 and CXCR4 improve recovery after traumatic
7. Iliff JJ, Chen MJ, Plog BA, et al. Impairment of glymphatic path- brain injury. Exp Neurol. 2021;338:113604.
way function promotes tau pathology after traumatic brain injury. J 28. Plog BA, Nedergaard M. The glymphatic system in central nervous
Neurosci. 2014;34(49):16180-­16193. system health and disease: past, present, and future. Annu Rev
8. Harrison IF, Ismail O, Machhada A, et al. Impaired glymphatic func- Pathol. 2018;13:379-­394.
tion and clearance of tau in an Alzheimer's disease model. Brain. 29. Xu H, Cao Y, Yang X, et al. ADAMTS13 controls vascular remod-
2020;143(8):2576-­2593. eling by modifying VWF reactivity during stroke recovery. Blood.
9. Rasmussen MK, Mestre H, Nedergaard M. The glymphatic pathway 2017;130(1):11-­22.
in neurological disorders. Lancet Neurol. 2018;17(11):1016-­1024. 30. McClean PL, Parthsarathy V, Faivre E, Hölscher C. The diabetes
10. Hablitz LM, Nedergaard M. The glymphatic system. Curr Biol. drug liraglutide prevents degenerative processes in a mouse model
2021;31(20):R1371-­R1375. of Alzheimer's disease. J Neurosci. 2011;31(17):6587-­6594.
11. Sullan MJ, Asken BM, Jaffee MS, DeKosky ST, Bauer RM. Glymphatic 31. Chen J, Wang X, Hu J, et al. FGF20 protected against BBB dis-
system disruption as a mediator of brain trauma and chronic trau- ruption after traumatic brain injury by upregulating junction pro-
matic encephalopathy. Neurosci Biobehav Rev. 2018;84:316-­324. tein expression and inhibiting the inflammatory response. Front
12. Xuan X, Zhou G, Chen C, et al. Glymphatic system: emerging ther- Pharmacol. 2020;11:590669.
apeutic target for neurological diseases. Oxid Med Cell Longev. 32. Shlosberg D, Benifla M, Kaufer D, Friedman A. Blood-­brain barrier
2022;2022:6189170. breakdown as a therapeutic target in traumatic brain injury. Nat Rev
13. Magnoni S, Esparza TJ, Conte V, et al. Tau elevations in the brain Neurol. 2010;6(7):393-­4 03.
extracellular space correlate with reduced amyloid-­β levels and pre- 33. Graham NSN, Cole JH, Bourke N, Schott JM, Sharp DJ. Distinct pat-
dict adverse clinical outcomes after severe traumatic brain injury. terns of neurodegeneration after TBI and in Alzheimer’s disease.
Brain. 2012;135(Pt 4):1268-­1280. Alzheimer’s Dement. 2023:1-­13. https://doi.org/10.1002/alz.12934
 |

17555949, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/cns.14308 by CochraneChina, Wiley Online Library on [02/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
LV et al. 13

34. Qu W, Liu NK, Wu X, et al. Disrupting nNOS-­PSD95 interaction im- S U P P O R T I N G I N FO R M AT I O N

proves neurological and cognitive recoveries after traumatic brain
Additional supporting information can be found online in the
injury. Cereb Cortex. 2020;30(7):3859-­3871.
35. Lai JQ, Shi YC, Lin S, Chen XR. Metabolic disorders on cognitive Supporting Information section at the end of this article.
dysfunction after traumatic brain injury. Trends Endocrinol Metab.
2022;33(7):451-­462.
36. Rask E, Olsson T, Söderberg S, et al. Insulin secretion and incretin
hormones after oral glucose in non-­obese subjects with impaired How to cite this article: Lv C, Han S, Sha Z, et al. Cerebral
glucose tolerance. Metabolism. 2004;53(5):624-­631.
glucagon-­like peptide-­1 receptor activation alleviates
37. Baggio LL, Drucker DJ. Biology of incretins: GLP-­1 and GIP.
Gastroenterology. 2007;132(6):2131-­2157. traumatic brain injury by glymphatic system regulation in
38. Vandebroek A, Yasui M. Regulation of AQP4 in the central nervous mice. CNS Neurosci Ther. 2023;00:1-13. doi:10.1111/cns.14308
system. Int J Mol Sci. 2020;21(5):1603.
39. Kitchen P, Salman MM, Halsey AM, et al. Targeting Aquaporin-­4
subcellular localization to treat central nervous system edema. Cell.
2020;181(4):784-­799 e719.