https://orcid.org/0000-0003-4629-5142
Abstract
Delirium (acute brain syndrome) is a common and serious neuropsychiatric disorder characterized by an acute decline in cognitive function. However, there is no effective treatment clinically. Here we investigated the potential effect of jujuboside A (JuA, a natural triterpenoid saponin) on cognitive impairment in delirium.
Delirium models of mice were established by injecting lipopolysaccharide (LPS) plus midazolam and implementing a jet lag protocol. Novel object recognition test and Y maze test were used to evaluate the effects of JuA on delirium-associated cognitive impairment. The mRNA and protein levels of relevant clock factors and inflammatory factors were measured by qPCR and Western blotting. Hippocampal Iba1+ intensity was determined by immunofluorescence staining.
JuA ameliorated delirium (particularly delirium-associated cognitive impairment) in mice, which was proved by the behavioural tests, including a preference for new objects, an increase of spontaneous alternation and improvement of locomotor activity. Furthermore, JuA inhibited the expression of ERK1/2, p-p65, TNFα and IL-1β in hippocampus, and repressed microglial activation in delirious mice. This was attributed to the increased expression of E4BP4 (a negative regulator of ERK1/2 cascade and microglial activation). Moreover, loss of E4bp4 in mice abrogated the effects of JuA on delirium as well as on ERK1/2 cascade and microglial activation in the hippocampus of delirious mice. Additionally, JuA treatment increased the expression of E4BP4 and decreased the expression of p-p65, TNFα and IL-1β in LPS-stimulated BV2 cells, supporting a protective effect of JuA on delirium.
JuA protects against delirium-associated cognitive impairment through promoting hippocampal E4BP4 in mice. Our findings are of great significance to the drug development of JuA against delirium and related disorders.
Introduction
Delirium is a common and serious neuropsychiatric disorder with an acute decline in cognitive function, which affects up to 50% of hospitalized elderly people and ~35% of patients in the intensive care unit.[1, 2] Due to poor prognosis including long-term cognitive impairment, prolonged hospital stays and increased mortality rates, delirium costs more than $164 billion per year in the USA and more than $182 billion per year in 18 European countries.[3] There is accumulating evidence that a number of factors can cause disruption of brain neuronal networks, thereby leading to acute cognitive dysfunction.[2] Some of these factors refer to neuroinflammation, physiological stress, hypoxia, metabolic derangements (impaired glucose oxidation) and genetic factors.[2] Currently, antipsychotic drugs (e.g. risperidone, olanzapine and quetiapine) are most commonly used for clinical treatment of delirium, but are concerned with inadequate effectiveness.[4] Delirium has been increasingly used as an indicator of healthcare quality for elderly and hospitalized people and has been considered as an important cause of the growing financial burden in society. Therefore, there is a considerable interest in finding new drugs and strategies for the prevention and treatment of delirium.
Jujuboside A (JuA) is a triterpenoid saponin extracted from the dried seeds of Semen Ziziphi Spinosae (SZS). SZS is a widely used herbal medicine in oriental countries, such as China, Korea, Japan and so on. It has been used as a sedative and hypnotic drug for more than 1000 years. JuA, a main active ingredient of SZS, has been demonstrated to possess many biological activities such as sedation, hypnosis, analgesia, anti-convulsion, neuroprotection and sleep improvement.[5–7] Of note, several studies reveal that JuA has protective effects on neuroinflammation and cognitive impairments in the mouse mode of dementia.[8, 9] The mechanisms for JuA’s effects on neurological disorders are becoming clear. For example, JuA could ameliorate the cognitive deficiency in Alzheimer’s patients with an Aβ clearance enhancement via activating the Axl/HSP90/PPARγ pathway.[10] JuA prevents sleep loss-induced disruption in the excitatory signalling pathway and memory impairment in mice with dementia by promoting GABAergic inhibition.[9] However, it is unknown whether JuA has a pharmacological effect on delirium and related disorders.
E4BP4 [E4 promoter-binding protein 4, also known as the nuclear factor interleukin-3-regulated protein (NFIL3)] is a nuclear transcriptional repressor of the basic leucine zipper (bZIP) family with an E4 promoter sequence containing an ATF consensus site.[10] E4BP4 was identified to be involved in the regulation of many circadian genes (e.g. Per2), immune processes (e.g. the development of natural killer T cells, TH17 cell differentiation and IL3-mediated pro-B cell survival), and lipid metabolism.[11–14] Moreover, E4BP4 is overexpressed in various cancers (e.g. lung cancer and colorectal cancer) and involved in cancer progression as it promotes cancer cell migration and metastasis, and enhances cancer cell lamellipodium formation and migration.[15–17] Notably, we recently show that E4BP4 functions as a cognition regulator in delirious mice, and can attenuate cognitive decline through inhibiting the ERK1/2 cascade and microglial activation.[4]
Current literature suggests that circadian clock genes can be targeted by small-molecule compounds to alleviate neurodegenerative diseases.[18–20] For instance, nobiletin, an agonist of the clock gene ROR, ameliorates memory impairment and amyloid β pathology in the transgenic mouse model with Alzheimer’s disease.[20] The PER2 (a circadian component) enhancer, nobiletin, attenuates delirium-like syndrome.[21] Induction of E4BP4 by the small molecule SR8278 improves cognitive function in delirious mice.[4] Timed illumination and melatonin administration partially attenuate sleep- and circadian rhythm-related symptoms (e.g. cognitive dysfunction and daytime sleepiness) that are associated with Alzheimer’s and Parkinson’s diseases.[21–23]
In this study, we aimed to assess the effects of JuA on delirium-associated cognitive impairment and investigate the underlying mechanisms. We first revealed the protective effect of JuA on delirium-associated cognitive impairment in mice. Furthermore, JuA induced hippocampal E4BP4 expression both in vivo and in vitro. The promotion of E4BP4 expression by JuA was associated with the inhibition of ERK1/2 cascade and neuroinflammation, which is one of the potential mechanisms of delirium-related cognitive impairment. In addition, loss of E4bp4 in mice abrogated the effect of JuA on cognitive decline. Therefore, we propose that JuA protects against delirium-associated cognitive impairment through promoting hippocampal E4BP4. Our findings are of great significance to the drug development of JuA against delirium and related disorders.
Materials and Methods
Materials
Lipopolysaccharide (LPS) and JuA were purchased from Aladdin Chemicals (Shanghai, China). Midazolam was obtained from the First Affiliated Hospital of Jinan University (Guangzhou, China). Dulbecco’s modified Eagle’s medium (DMEM, high glucose) was purchased from Sigma-Aldrich (St Louis, MO). Fetal bovine serum (FBS) was obtained from Hyclone (Logan, UT). Anti-Iba1 antibody was purchased from Abcam (Cambridge, MA). Anti-p-ERK1/2, anti-ERK1/2, anti-E4BP4, anti-REV-ERBα, anti-TNFα and anti-GAPDH antibodies were obtained from Proteintech (Chicago, IL). Anti-p-p65, anti-p65 and anti-IL-1β antibodies were purchased from Cell Signaling Technology (Boston, MA).
Mice
Male wild-type C57BL/6 mice (10 weeks of age) were purchased from HFK Biotech (Beijing, China). E4bp4-/- mice had been established in the author’s laboratory.[24] Mice were housed under a standard 12-h light/dark cycle environment with free access to water and food for 2 weeks to establish a stable circadian rhythm. All animal experiments were performed using protocols approved by the Institutional Animal Care and Use Committees of Guangzhou University of Chinese Medicine (Appr. date: 2021-04-13; IACUC Issue No: ZYD-2021-071).
Delirium model and drug treatment
Mice were randomly divided into six groups, which were named groups 1, 2, 3, 4, 5 and 6. There were six mice in each group. The mice in group 1 and group 2 were given an intrathecal injection of 10 mg/kg JuA and vehicle, respectively. The intrathecal injection was performed by a lumbar puncture at the intervertebral space of L5-6 using a 50 μl microinjector.[9] Six hours later, delirium was induced by an intraperitoneal injection of LPS (200 μg/kg), followed by a simultaneous injection of midazolam (10 mg/kg).[21] The two injections were defined as LM. The mice in group 3 and group 4 were given an intrathecal injection of 10 mg/kg JuA or vehicle once a day for consecutive 2 days, respectively. Similarly, 6 h after drug treatment on day 2, the two groups of mice were intraperitoneally injected with LM to induce delirium. The mice in group 5 and group 6 were given an intrathecal injection of 10 mg/kg JuA or vehicle once a day for consecutive 3 days, respectively. Six hours after drug treatment on day 3, mice were intraperitoneally injected with LM to induce delirium. Twenty-four hours after delirium induction, behavioural tests were conducted on mice in each group.
In another set of experiments, mice were given an intrathecal injection of 10 mg/kg JuA or vehicle once daily for 3 days. On day 3, mice were treated with LM, intraperitoneally. After another 24 h, mice were sacrificed, and hippocampi were collected.
Jet-lagged mice and drug treatment
Jet-lagged mice were established as previously described (8 h advances of the light/dark cycle every 2 days for 10 days).[25] On day 10, mice were given an intrathecal injection of 10 mg/kg JuA or vehicle once daily for 3 days. On day 13, mice underwent behavioural testing.
Wheel-running activity
Wheel-running experiments were carried out as previously described.[26] In brief, mice were housed individually in a cage (183 mm × 340 mm × 148 mm) equipped with a running wheel (Lafayette Instrument, Lafayette, IN), and the cage was placed in a light-tight ventilated box. After acclimation to the system, mice were treated with JuA (10 mg kg, i.t.) once daily for 3 days, and continuous recording was started. On day 3, mice were intraperitoneally injected with LM to induce delirium. The activity data were collected and analysed using Clcoklab software (Actimetrics, Wilmette, IL).
Novel object recognition (NOR) test
Novel object recognition (NOR) test was performed as previously described.[27] First, each mouse was placed into an open field (40 cm × 40 cm × 40 cm), and it was allowed to explore for 10 min. Second, two identically shaped objects were put in this open field, and the same mouse was allowed to explore for another 10 min. These two operations were called training process. Third, the mouse was taken out of the open field for 1 h. After that, one of the objects was changed to a new one with different shape. The mouse was re-exposed to the objects, a familiar one and a novel one, in the open field for 10 min to conduct the testing process. The time of each mouse spend on object exploration and the exploring numbers were recorded by a video tracking software (SMART Panlab, Harvard Apparatus, Holliston, MA). The object preference% was calculated as follows: object preference% = exploring numbers for (novel or old) object/total exploring numbers for (novel and old) objects × 100%.
Y maze test
Y maze test was performed using a Y maze device with three closed arms of 30 cm × 8 cm × 15 cm (length × width × height) as previously described.[28] For testing, each mouse was placed in the middle of the Y maze. The entries of each arm were recorded for the following 10 min. The consecutive entries into three different arms were regarded as one alternation. Spontaneous alternation rate% = the number of alternations/(the total number of arm entries-2) × 100%.[29]
Open field test (OFT)
Open field test (OFT) was performed to assess motor activity and anxiety-like behaviour. It was carried out by putting the mouse in the middle of an open field to record its exploration paths for 10 min. The total distance was analysed using a computerized image analysing system (Panlab Harvard Apparatus, Barcelona, Spain).
Immunofluorescent staining
The brain tissues were fixed overnight in 4% paraformaldehyde at 4℃. The fixed brain samples were embedded in paraffin and cut into 20 μm coronal sections. The obtained coronal sections were blocked using 5%BSA containing 0.2~0.3% Triton X-100 for 30 min. After that, the slides were sequentially incubated with anti-Iba1 antibody and fluorescent secondary antibody, which was followed by a 15-min incubation with 4’,6-diamidine-2-phenylindole dihydrochloride. Then sections were imaged using Nikon Optiphot fluorescent microscope (Tokyo, Japan), and the Iba1-positive cells were calculated by Image-pro pus 6.0 (Media Cybernetics, Rockville, MD).
Cell culture and treatment
BV2 cells were cultured in DMEM supplemented with 10% FBS at 37℃ in 5% CO2 humidified atmosphere. To start the experiment, BV2 cells were seeded into six-well plates. Once the confluence reached about ~90%, a group of cells were treated with different concentrations of JuA (2, 5 and 10 μM) or vehicle, and collected after 24 h treatment for qPCR and Western blotting. Another group of cells were stimulated by LPS for 8 h after being treated with JuA for 24 h and then collected for qPCR and Western blotting.
Serum shock experiment
Serum shock experiment was performed with BV2 cells as previously described.[30] In short, BV2 cells grown to about 70% of the confluence were pretreated with JuA (10 µM) or vehicle followed by LPS stimulation. After 8 h, the culture medium was changed to serum-free DMEM for an incubation of 12 h. Then 50% FBS was added into the wells, and cultured for 2 h. After that, the medium was discarded and changed back to serum-free DMEM. The cells were incubated for 6 or 10 h before collection for qPCR analysis.
qPCR assay
The total RNA of each sample was extracted according to the protocol using TRIzol reagent (Thermo Fisher Scientific, Waltham, MA). The qPCR reactions were performed as previously described.[9] The gene expression was normalized to Ppib and calculated using the 2-ΔΔCT method. All primer sequences are provided in Table 1.
Oligonucleotides used in this study
| Gene | Forward (5’-3’ sequence) | Reverse (5’-3’ sequence) |
|---|---|---|
| qPCR | ||
| Bmal1 | CTCCAGGAGGCAAGAAGATTC | ATAGTCCAGTGGAAGGAATG |
| E4bp4 | TATTGGGAGAAACGGCGGAAA | AGCCTTGGATGTCTGGTAGTC |
| Clock | ATGGTGTTTACCGTAAGCTGTAG | CTCGCGTTACCAGGAAGCAT |
| Per2 | CCACACTTGCCTCCGAAATA | ACTGCCTCTGGACTGGAAGA |
| Rev-erbα | TTTTTCGCCGGAGCATCCAA | ATCTCGGCAAGCATCCGTTG |
| Csf1r | TGTCATCGAGCCTAGTGGC | CGGGAGATTCAGGGTCCAAG |
| Il6ra | GCCACCGTTACCCTGATTTG | TCCTGTGGTAGTCCATTCTCTG |
| Tyrobp | CCCAAGATGCGACTGTTCTTC | GTCCCTTGACCTCGGGAGA |
| Trem2 | CTGGAACCGTCACCATCACTC | CGAAACTCGATGACTCCTCGG |
| Ccl2 | CCACAACCACCTCAAGCACT | AGGCATCACAGTCCGAGTCA |
| Ccl5 | GCTGCTTTGCCTACCTCTCC | TCGAGTGACAAACACGACTGC |
| Il-1β | GGCTGTATTCCCCTCCATCG | AAGGTGCTGATCTGGGTTGG |
| Il-6 | TAGTCCTTCCTACCCCAATTTCC | TTGGTCCTTAGCCACTCCTTC |
| Tnfα | AGGGTCTGGGCCATAGAACT | CCACCACGCTCTTCTGTCTAC |
| Mapk1 | TTGCTTTCTCTCCCGCACAAA | AGAGCCTGTTCAACTTCAATCC |
| Mapk3 | TCCGCCATGAGAATGTTATAGGC | GGTGGTGTTGATAAGCAGATTGG |
| Ppib | TCCACACCCTTTTCCGGTCC | CAAAAGGAAGACGACGGAGC |
| Gene | Forward (5’-3’ sequence) | Reverse (5’-3’ sequence) |
|---|---|---|
| qPCR | ||
| Bmal1 | CTCCAGGAGGCAAGAAGATTC | ATAGTCCAGTGGAAGGAATG |
| E4bp4 | TATTGGGAGAAACGGCGGAAA | AGCCTTGGATGTCTGGTAGTC |
| Clock | ATGGTGTTTACCGTAAGCTGTAG | CTCGCGTTACCAGGAAGCAT |
| Per2 | CCACACTTGCCTCCGAAATA | ACTGCCTCTGGACTGGAAGA |
| Rev-erbα | TTTTTCGCCGGAGCATCCAA | ATCTCGGCAAGCATCCGTTG |
| Csf1r | TGTCATCGAGCCTAGTGGC | CGGGAGATTCAGGGTCCAAG |
| Il6ra | GCCACCGTTACCCTGATTTG | TCCTGTGGTAGTCCATTCTCTG |
| Tyrobp | CCCAAGATGCGACTGTTCTTC | GTCCCTTGACCTCGGGAGA |
| Trem2 | CTGGAACCGTCACCATCACTC | CGAAACTCGATGACTCCTCGG |
| Ccl2 | CCACAACCACCTCAAGCACT | AGGCATCACAGTCCGAGTCA |
| Ccl5 | GCTGCTTTGCCTACCTCTCC | TCGAGTGACAAACACGACTGC |
| Il-1β | GGCTGTATTCCCCTCCATCG | AAGGTGCTGATCTGGGTTGG |
| Il-6 | TAGTCCTTCCTACCCCAATTTCC | TTGGTCCTTAGCCACTCCTTC |
| Tnfα | AGGGTCTGGGCCATAGAACT | CCACCACGCTCTTCTGTCTAC |
| Mapk1 | TTGCTTTCTCTCCCGCACAAA | AGAGCCTGTTCAACTTCAATCC |
| Mapk3 | TCCGCCATGAGAATGTTATAGGC | GGTGGTGTTGATAAGCAGATTGG |
| Ppib | TCCACACCCTTTTCCGGTCC | CAAAAGGAAGACGACGGAGC |
Oligonucleotides used in this study
| Gene | Forward (5’-3’ sequence) | Reverse (5’-3’ sequence) |
|---|---|---|
| qPCR | ||
| Bmal1 | CTCCAGGAGGCAAGAAGATTC | ATAGTCCAGTGGAAGGAATG |
| E4bp4 | TATTGGGAGAAACGGCGGAAA | AGCCTTGGATGTCTGGTAGTC |
| Clock | ATGGTGTTTACCGTAAGCTGTAG | CTCGCGTTACCAGGAAGCAT |
| Per2 | CCACACTTGCCTCCGAAATA | ACTGCCTCTGGACTGGAAGA |
| Rev-erbα | TTTTTCGCCGGAGCATCCAA | ATCTCGGCAAGCATCCGTTG |
| Csf1r | TGTCATCGAGCCTAGTGGC | CGGGAGATTCAGGGTCCAAG |
| Il6ra | GCCACCGTTACCCTGATTTG | TCCTGTGGTAGTCCATTCTCTG |
| Tyrobp | CCCAAGATGCGACTGTTCTTC | GTCCCTTGACCTCGGGAGA |
| Trem2 | CTGGAACCGTCACCATCACTC | CGAAACTCGATGACTCCTCGG |
| Ccl2 | CCACAACCACCTCAAGCACT | AGGCATCACAGTCCGAGTCA |
| Ccl5 | GCTGCTTTGCCTACCTCTCC | TCGAGTGACAAACACGACTGC |
| Il-1β | GGCTGTATTCCCCTCCATCG | AAGGTGCTGATCTGGGTTGG |
| Il-6 | TAGTCCTTCCTACCCCAATTTCC | TTGGTCCTTAGCCACTCCTTC |
| Tnfα | AGGGTCTGGGCCATAGAACT | CCACCACGCTCTTCTGTCTAC |
| Mapk1 | TTGCTTTCTCTCCCGCACAAA | AGAGCCTGTTCAACTTCAATCC |
| Mapk3 | TCCGCCATGAGAATGTTATAGGC | GGTGGTGTTGATAAGCAGATTGG |
| Ppib | TCCACACCCTTTTCCGGTCC | CAAAAGGAAGACGACGGAGC |
| Gene | Forward (5’-3’ sequence) | Reverse (5’-3’ sequence) |
|---|---|---|
| qPCR | ||
| Bmal1 | CTCCAGGAGGCAAGAAGATTC | ATAGTCCAGTGGAAGGAATG |
| E4bp4 | TATTGGGAGAAACGGCGGAAA | AGCCTTGGATGTCTGGTAGTC |
| Clock | ATGGTGTTTACCGTAAGCTGTAG | CTCGCGTTACCAGGAAGCAT |
| Per2 | CCACACTTGCCTCCGAAATA | ACTGCCTCTGGACTGGAAGA |
| Rev-erbα | TTTTTCGCCGGAGCATCCAA | ATCTCGGCAAGCATCCGTTG |
| Csf1r | TGTCATCGAGCCTAGTGGC | CGGGAGATTCAGGGTCCAAG |
| Il6ra | GCCACCGTTACCCTGATTTG | TCCTGTGGTAGTCCATTCTCTG |
| Tyrobp | CCCAAGATGCGACTGTTCTTC | GTCCCTTGACCTCGGGAGA |
| Trem2 | CTGGAACCGTCACCATCACTC | CGAAACTCGATGACTCCTCGG |
| Ccl2 | CCACAACCACCTCAAGCACT | AGGCATCACAGTCCGAGTCA |
| Ccl5 | GCTGCTTTGCCTACCTCTCC | TCGAGTGACAAACACGACTGC |
| Il-1β | GGCTGTATTCCCCTCCATCG | AAGGTGCTGATCTGGGTTGG |
| Il-6 | TAGTCCTTCCTACCCCAATTTCC | TTGGTCCTTAGCCACTCCTTC |
| Tnfα | AGGGTCTGGGCCATAGAACT | CCACCACGCTCTTCTGTCTAC |
| Mapk1 | TTGCTTTCTCTCCCGCACAAA | AGAGCCTGTTCAACTTCAATCC |
| Mapk3 | TCCGCCATGAGAATGTTATAGGC | GGTGGTGTTGATAAGCAGATTGG |
| Ppib | TCCACACCCTTTTCCGGTCC | CAAAAGGAAGACGACGGAGC |
Western blotting
The protein of each sample was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (a 10% gel) and transferred to polyvinylidene fluoride membranes. The proteins on polyvinylidene fluoride membranes were blocked with 5% skimmed milk for 1 h at room temperature, which was followed by an immediate incubation with the primary antibodies overnight at 4°C. After that, the membranes with proteins were washed with PBS twice, and the secondary antibody was added for 2 h incubation at room temperature. The membranes were visualized by an Omega LumG Imaging System (Aplegen, Pleasanton, CA). The protein levels were analysed using the Quantity One software and normalized to GAPDH.
Statistical analysis
All data were presented as mean ± standard errors of the mean (SEM). Student’s t-tests were used to analyse the statistical differences between two groups. One-way or two-way ANOVA followed by Bonferroni post hoc test was used for multiple group comparisons. All statistical analyses were performed using GraphPad Prism 8.0 (GraphPad Software Inc., San Diego, CA). The difference of P < 0.05 was considered statistically significant (*).
Results
JuA ameliorates delirium-associated cognitive decline in mice
A delirium model of mice was established by injecting LPS with a simultaneous injection of midazolam, which was defined as LM treatment as previously described.[4] Based on this model the pharmacological effects of JuA were determined (Figure 1A). Delirium-associated cognitive impairments were assessed by NOR experiment and Y maze tests. We found that JuA treatment (once daily for consecutive 2 or 3 days) protected the mice from cognitive decline, which was proved by the preference of novel object in the NOR test (Figure 1B and C). Also, the delirious mice treated with JuA displayed significantly increased spontaneous alternations in the Y maze test compared with the control group (Figure 1D). By contrast, JuA treatment (once only) failed to prevent mice from delirium-associated cognitive impairment due to an equal preference for old and novel objects as well as spontaneous alternations of < 60% (the alternation rates of 60–70% indicate normal cognitive function) (Figure 1B–D). Of note, delirium-associated behavioural disruption such as lethargy was ameliorated by JuA (for 3 days’ treatment), as evidenced by an increased locomotor activity in the OFT test (Figure 1E and F).[4] Circadian disturbance in wheel-running activity is another feature of delirium.[31, 32] We next tested whether JuA can improve delirium-associated disruption in wheel-running activity. As expected, JuA did not affect wheel-running activities in normal control mice (Figure 2A–C). However, JuA restored the dampened rhythms of the wheel running activities in delirious mice with a decreased activity in the subjective light and an increased activity in the subjective dark (Figure 2B and D). We additionally evaluated the protective effect of JuA on jet lag-induced delirium (another model of delirium[4]). Jet-lagged mice were established according to a published protocol (i.e. 8 h light advance every 2 days for 10 days, Figure 3A).[33] JuA-treated jet-lagged mice had an improved cognitive phenotype with an increased novel object preference in the NOR test and an increased spontaneous alternation in the Y maze test as well (Figure 3B–D). To sum up, all our findings indicated that JuA protects mice against delirium, particularly with respect to the cognitive decline.
JuA ameliorates LM-induced delirium in mice. (A) Schematic design for the animal studies, including delirium induction, JuA treatment and behavioural tests. (B) Representative exploratory paths of delirious mice treated with JuA and vehicle in NOR test. (C) Object preference (%) analysis for delirious mice treated with JuA and vehicle based on NOR test. (D) Spontaneous alterations (left) and arm entries (right) for JuA- and vehicle-treated delirious mice based on the Y maze test. (E) Representative activity trails for JuA- and vehicle-treated delirious mice in the OFT test. (F) Total distance for JuA- and vehicle-treated delirious mice based on OFT test. Data are mean ± SEM (n = 6 mice). *P < 0.05 (t-test).
JuA inhibits LM-induced locomotor disturbance in mice. (A) Experimental scheme for JuA treatment, delirium induction and wheel-running activity recording. (B) Representative wheel-running actogram for JuA- and vehicle-treated delirious mice. (C) Circadian wheel-running activities for JuA- and vehicle-treated mice before LM treatment (n = 4). (D) Circadian wheel-running activities for JuA- and vehicle-treated mice after LM treatment (n = 4). The wheel-running activities in left, middle and right panels of C and D were, respectively, recorded at 1-min, ~12-h and 60-min intervals. Data are mean ± SEM. *P < 0.05 (t-test).
JuA prevents delirium-associated cognitive decline in mice with jet lag. (A) Experimental design for establishment of jet-lagged mice, JuA treatment and behavioural tests. (B) Object preference (%) analysis for jet-lagged mice treated with JuA and vehicle based on NOR test. (C) Representative exploratory paths for JuA- and vehicle-treated jet-lagged mice in the NOR test. (D) Spontaneous alterations (left) and arm entries (right) for JuA- and vehicle-treated jet-lagged mice based on the Y maze test. Data are mean ± SEM (n = 6 mice). *P < 0.05 (t-test).
JuA induces E4BP4 expression in mice and in BV2 cells
Accumulating studies have demonstrated that circadian clock genes (such as E4bp4 and Per2) are involved in the regulation of delirium.[4, 21] To test whether the protective effects of JuA on the delirium-associated cognitive decline are due to the modulation of clock genes, we determined clock genes in mice and in BV2 cells with and without JuA treatment. We observed induced hippocampal expression of E4BP4 mRNA and protein in JuA-treated mice (Figure 4A and B). However, the expression of other clock genes such as Bmal1, Clock and Per2 was unaltered in JuA-treated mice (Figure 4A and B). Of note, JuA led to a moderate increase in Rev-erbα mRNA, however, REV-ERBα protein was unchanged (Figure 4A and B). In addition, JuA dose-dependently induced E4bp4 mRNA expression in BV2 cells (Figure 4C). E4BP4 protein level was also increased in a dose-dependent manner in JuA-treated cells, which was consistent with its mRNA changes (Figure 4D). It has been shown that inflammation could inhibit the clock gene expression.[4, 34] LPS is an NF-κB activator that can induce cellular inflammatory responses. Intriguingly, JuA attenuated the inhibitory effect of LPS on E4bp4 expression, as evidenced by an increased E4bp4 expression in LPS-stimulated BV2 cells (Figure 4E). In a word, these findings suggested that JuA can induce the expression of the clock gene E4bp4.
JuA induces E4BP4 expression. (A) Effects of JuA on the expression of clock genes in mouse hippocampus. (B) Effects of JuA on E4BP4 and REV-ERBα proteins in mouse hippocampus. (C) Effects of JuA on E4bp4 mRNA expression in BV2 cells. (D) Effects of JuA on E4BP4 protein in BV2 cells. (E) Effects of JuA on E4bp4 expression in synchronized BV2 cells with LPS stimulation. Data are mean ± SEM (n = 3). *P < 0.05 (t-test or one-way ANOVA with Bonferroni post hoc test). Rel, relative.
JuA represses ERK1/2 cascade and inflammation through promoting hippocampal E4BP4
Our previous study revealed that E4bp4 is a cognitive regulator in delirium via regulating the ERK1/2 cascade and microglial activation.[4] Since JuA can induce E4BP4 expression (Figure 4), we then determined whether JuA regulates the ERK1/2 cascade and microglial activation. We found that JuA upregulated E4bp4 mRNA expression and downregulated Mapk1 and Mapk3 mRNAs in the hippocampus of delirious mice at the same time (Figure 5A). Mapk1 and Mapk3 are two E4BP4 target genes encoding ERK2 and ERK1 proteins, respectively. Consistently, E4BP4 protein in the hippocampus was increased, whereas ERK1 and ERK2 proteins as well as their phosphorylated forms (i.e. p-ERK1 and p-ERK2) were decreased by JuA (Figure 5B). We also found that JuA inhibited the expression of microglial activation-related genes (such as Il6rα, Tyrobp, Csf1r and Trem2) in the hippocampus of delirious mice (Figure 5C). Consistent with these changes, the microglial activation marker Iba1 was down-regulated by JuA (Figure 5D). The decreased number of Iba1+ cells also supported the point that JuA inhibited the activation of microglia according to immunofluorescent staining (Figure 5E). Besides, JuA down-regulated the related inflammatory genes (such as Il-1β, Il-6, Ccl2 and Ccl5) and proteins TNFα, IL-1β and p-p65 (an NF-κB subunit and a downstream target of ERK1/2[35]). All these data clearly showed that JuA inhibited the ERK1/2 cascade and the activation of microglia in the hippocampus of mice, an effect attributing to the induction of hippocampal E4BP4.
JuA represses ERK1/2 cascade and microglial activation through an action on E4BP4 in the hippocampus of LM-induced delirious mice. (A) Effects of JuA on the expression of E4bp4 and Mapk1/3 mRNAs in LM-induced delirious mice. (B) Effects of JuA on the expression of ERK1/2 and E4BP4 proteins in LM-induced delirious mice. (C) Effects of JuA on microglial activation-related transcripts in LM-induced delirious mice. (D) Effects of JuA on the proteins of inflammatory factors in LM-induced delirious mice. (E) Immunofluorescent staining of hippocampus from JuA- and vehicle-treated delirious mice. *P < 0.05 (t-test). Rel, relative.
We further confirmed the downregulatory effects of JuA on the ERK1/2 cascade and inflammation in LPS-induced BV2 cells. JuA decreased Mapk1/3 mRNAs in a dose-dependent manner (Figure 6A). Similarly, the expression of ERK1/2 proteins was dose-dependently downregulated by JuA in the same manner as the mRNA regulation (Figure 6B). Additionally, JuA alleviated LPS-induced inflammatory responses as evidenced by decreases in the inflammatory factors (i.e. Il-1β, Il-6, Tnfα and Ccl2) and in related proteins (i.e. TNFα, IL-1β and p-p65), and the alleviating effects were dose-dependent (Figure 6C and D). Concurrently, JuA showed a positive regulatory effect on E4BP4 and an ensuing negative regulatory effect on p-ERK1/2 (Figure 6D). To sum up, these findings suggested that JuA repressed the ERK1/2 cascade and inflammation through promoting E4BP4 expression.
JuA represses ERK1/2 cascade and inflammation through an action on E4BP4 in BV2 cells. (A) JuA induces the expression of Mapk1/3 mRNA in BV2 cells in a dose-dependent manner. (B) JuA induces the expression of ERK1/2 proteins in BV2 cells in a dose-dependent manner. (C) JuA inhibits the expression of inflammatory genes in LPS-stimulated BV2 cells in a dose-dependent manner. (D) JuA inhibits inflammatory proteins in LPS-stimulated BV2 cells in a dose-dependent manner. Data are mean ± SEM (n = 3). *P < 0.05 (one-way or two-way ANOVA with Bonferroni post hoc test). Rel, relative.
Cognitive protection by JuA in delirium is dependent on E4bp4
Since JuA can alleviate delirium-associated cognitive decline via inducing E4BP4 expression, we thus investigated whether the cognitive protection by JuA is E4bp4-dependent. To this end, we generated mice with global deletion of E4bp4 (E4bp4-/- mice, verified by qPCR and Western blotting, Figure 8A) and tested the JuA effects in these animals. We found no evident preference for novel object in JuA-treated delirious E4bp4-/- mice as compared with controls in the NOR test (Figure 7A). Moreover, there are no differences in spontaneous alterations (the rates: 46.8% vs. 46.9%) in the Y maze test between JuA-treated and vehicle-treated delirious E4bp4-/- mice (Figure 7B). The knockouts displayed a similar preference for objects and a low rate of spontaneous alteration, suggesting a negligible effect of JuA on delirium-associated cognitive decline in E4bp4-/- mice.
Loss of E4bp4 in mice abrogates the efficacy of JuA against cognitive decline. (A) Object preference (%) analysis for delirious E4bp4-/- mice treated with JuA and vehicle based on NOR test. (B) Spontaneous alternations (left) and arm entries (right) for delirious E4bp4-/- mice treated with JuA and vehicle based on the Y maze test. Data are mean ± SEM (n = 6 mice).
E4bp4 deficiency abrogates the effects of JuA on inflammation in the hippocampus of mice. (A) E4BP4 mRNA and protein expression in hippocampus from E4bp4-/- and wild-type mice. (B) Effects of JuA on the expression of Mapk1/3 mRNA and the factors related to microglial activation in delirious E4bp4-/- mice. (C) Effects of JuA on the protein expression of ERK1/2 and inflammatory factors in delirious E4bp4-/- mice. (D) Immunofluorescent staining of hippocampus from E4bp4-/- and wild-type mice. Data are mean ± SEM (n = 3).
We further examined the expression of hippocampal Mapk1 and Mapk3 in JuA-treated E4bp4-/- mice after delirium induction. Unsurprisingly, JuA treatment did not change the mRNA levels of both Mapk1 and Mapk3 (Figure 8B). Consistently, the protein expression of p-ERK1/2 was unaltered by JuA in E4bp4-/- mice after delirium induction (Figure 8C). Immunofluorescent staining demonstrated an increase in Iba1+ cell number in the hippocampus of E4bp4-/- mice (Figure 8D), indicative of microglial activation as previously reported. However, microglial activation-related genes (such as Tyrobp, Il6rα, Trem2 and Csf1r, Figure 8B) and relevant inflammatory genes (such as Il-6 and Il-1β), as well as the related proteins (Iba1, IL-1β, TNFα and p-p65), remained unaffected in JuA-treated delirious E4bp4-/- mice (Figure 8C). In summary, our findings indicated that the protective effect of JuA on delirium-associated cognitive impairment in mice was dependent on E4bp4.
Discussion
We revealed the protective effect of JuA on delirium-associated cognitive impairment in mice. JuA prevented cognitive impairment both in LM-induced and jet lag-induced delirium in mice (Figures 1 and 3). Also, JuA improved delirium-associated behavioural disruptions in mice after LM induction (Figures 1 and 2). Mechanistic studies revealed that JuA promoted hippocampal expression of E4BP4 to repress the ERK1/2 cascade and microglial activation (a mechanism underlying delirium-associated cognitive impairment), thereby ameliorating cognitive decline in delirious mice (Figures 1–5). Microglial cell-based assays further confirmed that JuA downregulated ERK1/2 and inflammatory factors expression (such as IL-6 and IL-1β) through inducing E4BP4 expression (Figure 6). In addition, loss of E4bp4 in mice abrogated the effects of JuA on delirium-associated cognitive decline as well as on ERK1/2 cascades and microglial activation in the mouse hippocampus (Figures 7 and 8). Therefore, we proposed that the protective effects of JuA on cognitive decline were dependent on E4bp4. It is noteworthy that JuA is regarded as the main active ingredient of Ziziphi Spinosae which is already in medical use. Recent studies have revealed that Ziziphi Spinosae can improve the learning capacity and memory in dementia animal model.[8] It is thus likely that Ziziphi Spinosae (as a whole) also has a protective effect on delirium-associated cognitive decline and that this effect depends on JuA and E4BP4.
Neuroinflammation is a critical etiological factor for many brain disorders associated with cognitive dysfunction, including dementia and delirium.[36] It is commonly induced by peripheral inflammatory mediators (e.g. IL-6, TNFα and MCP1) that can cross the blood-brain barrier and enter the brain to activate microglia and astrocytes.[37] Microglial activation has been implicated in delirium development. Activated microglia release inflammatory factors, such as IL-6, TNFα and IL-1β, which impair memory and have long-term consequences on hippocampus and thus play a key role in the onset of delirium.[38] This is supported by a current study in which JuA prevents delirium-like syndrome in mice through inhibiting microglial activation and the production of inflammatory mediators. Thus, microglia may be a promising therapeutic target for delirium.
The current study and our previous one support that induction of E4BP4 expression is a novel strategy to repress microglial activation and manage delirium.[4] First, E4BP4, a clock component, is expressed in a circadian pattern with a peak value in the activity phase and a trough value in the rest phase in the hippocampus, which is anti-phase to the severity of delirium-like syndrome (i.e. more severe at the rest phase and less severe at the activity phase).[4] Second, deficiency of E4bp4 in mice exacerbated delirium-associated cognitive deficits in mice after delirium induction, however, the symptom was ameliorated by induction of E4bp4.[4] Third, activation of the clock component Per2, a target gene of E4bp4, improved cognitive function in delirious mice.[21] Lastly, our current study uncovered a protective role of JuA against delirium-like syndrome through increasing hippocampal E4BP4 expression (Figures 7 and 8).
In addition to cognitive and attentional deficits, the circadian locomotor disturbance is another core feature of delirium.[39] Clinical data indicate that patients with delirium are less active in the daytime and more active in the nighttime.[40] This circadian disturbance in locomotor activity has become an indicator to identify early clinical delirium. Consistently, the circadian locomotor disturbance was also observed in delirious mice.[4] In this study, we used wheel-running equipment to assess the effects of JuA on circadian changes in locomotor activity in mice. We found that LM-induced locomotor disturbance was improved by JuA, as evidenced by an increase in locomotor activity in the subjective dark phase and a decrease in the subjective light phase (Figure 3). Previous studies established that the locomotor disturbance in delirium is associated with a disruption in the sleep-wake cycle, characterized by daytime sleepiness and night-time activity.[7, 41] Since JuA has the potential to improve sleep quality,[9] it was reasonable to speculate that JuA can rescue delirium-associated sleep-wake cycle disturbance. However, whether this is true or not awaits further investigations.
It is noteworthy that JuA has low oral bioavailability and poor distribution in the brain.[42–44] This is why intrathecal injection was applied in this proof-of-concept study. E4BP4 modifiers with high oral bioavailability and good distribution in the brain should be explored for practical application and drug development. The dose (10 mg/kg) and treatment duration (i.e. 1, 2 and 3 days) of JuA for the management of cognitive deficits in delirious mice herein were formulated according to a previous study.[10] JuA ameliorated the cognitive deficits in a treating time-dependent manner with a superior efficacy for the 3 days’ treatment (Figure 1). Delirium-like syndrome was assessed using NOR, Y maze and OFT tests. NOR test is a benchmark test for recognition memory, which is a non-rewarded and fear-irrelevant test based on the innate tendency of rodents to seek novelty.[45] The Y maze test is widely used to assess spatial working memory and cognitive function in mice and to evaluate the efficacy of chemical entities against cognitive dysfunction.[21] OFT test is mainly used to assess the motor activity, exploratory activity and anxiety-like behaviour of animals.[46]
Conclusion
JuA protects against delirium-associated cognitive impairment through promoting hippocampal E4BP4. Our findings have implications for the drug development of JuA against delirium and related disorders.
Author Contributions
Participated in research design: JD, FZ, DD and BW. Conducted experiments: JD, FZ, MC, YX and LZ. Performed data analysis: JD, FZ, MC, DD and BW. Wrote or contributed to the writing of the manuscript: JD, MC and BW.
Funding
This work was supported by the Guangdong Basic and Applied Basic Research Foundation (No. 2022A1515010535).
Conflict of Interest
The authors have declared that no conflict of interest exists.
Data Availability
Data are available in the open-access article.
References
Author notes
Jianhao Du and Fugui Zhang. These authors contributed equally to this work.
