USP43 promotes cerebral ischemia–reperfusion injury via activation of TAK1

Abstract

Background

Neuroinflammation and apoptosis are important processes of cerebral ischemia–reperfusion injury. USP43 has been shown to play an important role in a variety of cancers, but its function in the field of neuroscience is unclear.

Results

We used an ischemia–reperfusion model of transient embolization of the middle cerebral artery in mice and found that USP43 protein expression was elevated in mice after ischemia–reperfusion injury. Usp43 knockout reduced cerebral infarct volume, inflammatory response, and cell apoptosis, compared to control mice. In vitro, we used an oxygen-glucose-deprived primary neuronal model of rat cortex. Compared with the control group, the Usp43 knockdown group had better cell activity, and the inflammatory response and apoptosis were reduced. Cells in the Usp43 overexpression group exhibited the opposite performance. Mechanistically, we found that USP43 directly interacts with TAK1 by exerting its function as a deubiquitinating enzyme, removing the K48 chain ubiquitination of TAK1 and activating the TAK1-JNK/p38 signaling pathway. Inhibition of USP43 enzyme activity or the use of TAK1 inhibitors can reverse the inflammatory response and neuronal apoptosis induced by USP43 overexpression.

Conclusions

These results suggest that USP43 promotes I–R damage by activating TAK1 and its downstream signaling pathways. Inhibition of USP43 may be a potential treatment modality for acute stroke.

Background

Among the non-communicable diseases, stroke continues to be the leading cause of disability and death for the global population [1, 2]. Ischemic strokes account for approximately 65% of new strokes each year [3]. Stroke causes a series of neuronal injuries, which subsequently lead to irreversible neurological deficits, including paralysis, sensory deficits, and cognitive dysfunction [4]. Revascularization techniques such as pharmacological thrombolysis and acute endovascular thrombolysis are currently the mainstay of clinical treatment of ischemic stroke [5, 6]. However, due to the narrow time window for treatment and the potential risk of bleeding, the prognosis for stroke is unsatisfactory [7]. In addition, revascularization inevitably produces ischemia–reperfusion injury [8, 9]. Cerebral ischemia–reperfusion (I–R) injury involves an array of complex pathophysiological processes, including inflammation, apoptosis, oxidative stress and neuroexcitotoxicity [10–12]. Drug discovery against these targets is on the horizon, but effects have mostly been limited to cell and animal studies. Therefore, we need to better understand the pathological process of cerebral I–R injury and to look for more effective treatments.

Ubiquitin specific peptidase 43 (USP43) is a linear protein with 1123 amino acids, belonging to the deubiquitinating enzyme (DUB) family, with deubiquitinating enzyme activity. It is located on human chromosome 7 p13.1. In humans, USP43 is mainly expressed in the brain, aorta, and lungs [13], and is mainly expressed in neurons in brain tissue. USP43 current research has focused on oncology, with a few studies related to immune regulation [14–20]. USP43 has been found to be upregulated in breast cancer tissue and has been shown experimentally to regulate the cell cycle and promote proliferation and metastasis of breast cancer cells. Ye et al. found that USP43 was highly expressed in colorectal cancer tissues. Meanwhile, USP43 can regulate the stability of ZEB1 protein and promote the proliferation and metastasis of colorectal cancer. Lavaud et al. found that the expression of USP43 was increased in osteosarcoma cells. USP43 predicts survival in patients with non-small cell lung cancer, according to Peinado-Serrano et al. Zhao et al. found that increased expression of USP43 in pancreatic ductal adenocarcinoma tissue promotes PANC-1 cell proliferation. In bladder cancer, USP43 stabilizes c-myc to promote glycolysis and metastasis in bladder cancer. In ovarian cancer, USP43 expression was upregulated and increased the migration and invasion ability of ovarian cancer cells and decreased their sensitivity to cisplatin. Zhao et al. found that USP43 modulates IFN expression and regulates antiviral immunity. In addition, USP43 has been found to be associated with gestational diabetes and cervical spondylosis [21, 22]. USP43 may be a potential predictor of cancer prognosis and a favorable target for future monitoring of cancer treatments. However, the role of USP43 in ischemic stroke remains unclear.

This study aimed to investigate the role and potential mechanisms of USP43 in I–R neuronal injury. In this study, increased expression of USP43 was observed during brain I–R injury. In addition, we demonstrated that USP43 exacerbates brain injury by exacerbating post-ischemic inflammation and apoptosis. In contrary, USP43 deficiency ameliorates neuronal apoptosis and inflammatory response. Mechanistically, we found that USP43 interacts with TGF-β-activated kinase 1 (TAK1) and inhibits its degradation by removing K48-linked ubiquitin on TAK1, thereby activating the TAK1-JNK/p38 signaling pathway. In conclusion, we revealed the function of USP43-TAK1 axis in cerebral ischemia–reperfusion injury and identified USP43 as a potential target for the treatment of cerebral I–R injury.

Methods

Animals

The animal experiment protocol was approved by the Experimental Animal Welfare Ethics Committee,Zhongnan Hospital of Wuhan University. (APPROVAL#ZN2024275). According to the National Institutes of Health's guidelines for the care and use of laboratory animals, animals receive good human care. Male, 10–12 weeks old, 26–28 g, Usp43 knockout (KO) mice and their littermate negative mice (C57BL/6 background) were used for the experiments. All experimental mice were housed in the SPF system under the following conditions: room temperature between 22 and 24 °C, humidity between 40 and 70%, alternating light and dark illumination for 12 h, and free access to food and water. Usp43 knockout (KO) mice were purchased from Ganan Institute of Innovation and Translational Medicine.

Transient middle cerebral artery occlusion (t/MCAO) model

A cerebral ischemia–reperfusion injury model was established using transient middle cerebral artery occlusion surgery (t/MCAO) [23]. Anesthesia was induced and maintained in mice using 2.0% isoflurane and an oxygen/nitrous oxide gas mixture, and skin preparation was carried out on the neck and left part of the skull. A longitudinal incision was made in the skin of the left parietal of the skull to expose the skull and peel off the connective tissue on the surface of the skull. The fiber-optic probe of the laser Doppler flowmeters was fixed with bio-adhesive 1.5 mm behind the fontanel and 3–4 mm from the middle suture of the skull in order to detect the cerebral blood flow during the surgical procedure. The rectal temperature probe was inserted, and the body temperature was maintained at 37 ± 0.5 °C throughout the operation. A median neck incision was made to expose the left common carotid artery, internal carotid artery and external carotid artery. A 6–0 silicone-coated monofilament wire (Cat#602156PK10Re, Doccol) was inserted into the left external carotid artery, advanced toward the internal carotid artery, and wedged into the cerebral artery ring to block the beginning of the middle cerebral artery (MCA). After 45 min of ischemia, the wire bolus was slowly withdrawn to restore the middle cerebral artery blood flow, the wound was sutured, and the animals were kept routinely after they were awake. A decrease in blood flow of more than 75% during ischemia and a recovery of more than 70% after reperfusion were considered successful modeling. Samples were taken 24 h or 48 h after reperfusion. The experimental mice in the sham-operated group (Sham group) did not receive ischemic treatment, but the other operative steps were the same as in the t/MCAO group.

Neurological score

The neurological impairment of the mice was assessed using a modified 9- point scale based on Berderson score after 24 h reperfusion [24]. The scale is described as follows: 0 = no neurological deficits; 1 = the contralateral forelimb curled up, or unable to fully reach the affected forelimb when the tail is suspended; 2 = the contralateral shoulder is adducted when the tail is suspended; 3 = reduced resistance to a push to the contralateral side; 4 = moving spontaneously but turning to the contralateral when dragged by the tail; 5 = only spontaneously hovering or walking contralateral; 6 = moving only in response to stimulus; 7 = no response to stimuli; and 8 = stroke- related death.

Brain samples preparation

After neurological scoring in mice, mice were anaesthetized and sacrificed with pentobarbital sodium. Subsequently, the heart was fully exposed along the median line, and cardiac perfusion was performed with 4 °C PBS buffer, and brain tissue was taken for use in molecular biology or pathology testing after no obvious blood coloration was observed in the liver of the mice.

2,3,5-Triphenyltetrazolium chlorid (TTC) staining and infarct volume calculation

TTC staining was used to calculate the infarct volume after 24 h of reperfusion. Brain tissues were frozen in the refrigerator at – 20 ℃ for 30 min and cut into 7 consecutive slices of 1 mm thickness. The slices were immediately placed in 2% TTC staining solution and incubated at a constant temperature of 37 °C for 10 min, and the slices were turned from time to time to make the tissues uniformly stained. After the t/MCAO surgery, normal brain tissue was stained red, while infarct tissue and penumbra were stained pale grey. After fixation in 4% neutral paraformaldehyde solution, photographs were taken and the infarct volume was counted using Image Pro Plus (version 6.0) software. Infarct volume (%) was calculated using the following formula: (Volume of the contralateral hemisphere minus the volume of the non- lesioned ipsilateral hemisphere)/(contralateral volume times 2) × 100%.

Immunofluorescence staining

For F4/80 staining, paraffin sections were repaired with EDTA at high temperature for 20 min, then mouse brain tissue sections were closed with 10% goat serum and incubated with F4/80 antibody (Cat#GB11027, Servicebio) at 4 °C overnight. After completion of the incubation, the sections were washed with PBS. Then, the secondary antibody (Alexa Fluor^® 568 goat anti-Rabbit IgG(H + L) Cat#A11036, Invitrogen, Carlsbad, CA, USA) was added and incubated for 1 h at room temperature, and the nuclei were stained with DAPI. Sections were observed under a fluorescence microscope (Cat#BX51, OLYMPUS), photographed, and pictures were analyzed using Image Pro Plus (version 6.0) software.

For TUNEL staining, paraffin sections were stained using a TUNEL staining kit (Cat#11684817910, Roche) following the steps in the instructions. The nuclei were stained with DAPI. Sections were observed under a fluorescence microscope (Cat#BX51, OLYMPUS), photographed, and pictures were analyzed using Image Pro Plus (version 6.0) software.

Serum inflammatory factor test

The contents of serum TNF-α (Cat#ELK1387-1, ELK), IL-6 (Cat#ELK1157-1, ELK), and IL-1β (Cat#ELK1271-1, ELK) were detected using an Elisa kit according to the steps in the instructions.

Isolation and culture of primary neurons

The cerebral cortex was collected from Sprague–Dawley rats at 1–2 days postnatal age, clipped and placed in 2 ml of 0.125% trypsin (Cat#27250018, GIBCO) and incubated for 15 min in a 37 °C to obtain cortical cells. Then, the digestion reaction was terminated using DMEM-F12 medium (Cat#PB150315-500, Pernosay) containing 10% fetal bovine serum (Cat#FBS-S500, Newzerum) and DNAase (Cat#10104159001, Roche). Clumped cells or incompletely digested tissue mass were removed with a 40um cell sieve (Cat#352340, Corning) and cells were collected by centrifugation at 1500 rmp for 5 min at 4 °C, after which the cells were resuspended in DMEM-F12 medium containing 10% fetal bovine serum and 1% dual antibiotic (Cat#PB180120, Pernoside). After counting, the cells were inoculated on poly-lysine (10 mg/ml, Sigma) coated plates and incubated at 37 ℃ and 5% CO₂ for 3 h. The cell status was observed under the microscope, if the cells started to grow tadpole-like synapses, the old medium was discarded at this time. Neurobasal (Cat#10888022, GIBCO) medium containing 1% l-Glutamine (Cat#ST083, Beyotine), 1% dual antibiotic and 2% B27 (Cat#17504044, GIBCO) was added to the cells under the light after that the culture was continued. The medium was changed every 48 h and the cells were cultured for 7 days for subsequent experiments.

Oxygen and glucose deprivation/reoxygenation (OGD/R) stimulation and TAK1 inhibitor treatment

For OGD/R stimulation, the original medium was replaced with serum-free, glucose-free and sodium pyruvate-free DMEM-F12 medium. After that, the cells were incubated in an incubator containing 95% N₂ and 5% CO₂ for 3 h for hypoxia treatment. Subsequently, the cells were cultured in an incubator containing 95% air and 5% CO₂ using normal medium for 6 h. Control neurons did not undergo OGD/R.

To inhibit the activity of TAK1, the TAK1 inhibitor 5Z-7-oxozeaenol (Cat#O9890-1 MG, Sigma, 600 nM) was added to the medium 1 h before hypoxia.

Construction and infection of Usp43 knockdown and overexpression adenovirus

The Usp43 gene has a high homology in mice and rats. In order to overexpress Usp43, the mouse cDNA sequence was used as a template to amplify the CDS sequence of Usp43 and cloned into the adenoviral entry vector (pENTR-CMV-ATG-flag-T2A-EGFP) to obtain the Usp43 overexpression entry plasmid. To knockdown Usp43, three short hairpin RNA sequences targeting to the Usp43 gene were designed and synthesized, which were cloned into the adenoviral entry vector (pENTR-U6-CMV-ATG-flag-T2A) to obtain the Usp43 knockdown entry plasmids. The above entry plasmids were recombined with adenoviral recombinant plasmid pAd⁄PL-DEST™ (Cat#V49420, ThermoFisher) for Gateway site-specific recombination (GateWay@LR Clonase TM II Enzyme Mix, Cat#2484478, ThermoFisher). The plasmids were subsequently transfected into HEK293A cells using PEI transfection reagent (Cat#24765-100, Polysciences) after enzymatic linearisation. Adeasy Adenovirus Packaging System (Cat#240009, Agilent Technologies) were used to obtain the adenovirus. Recombinant adenovirus titer was determined to be 1 × 10¹⁰ pfu/ml. Similar adenoviral vectors encoding the GFP gene (AdGfp) and scrambled short hairpin RNA (AdshRNA) were used as controls (Table 1).

Table 1.

Primer of Usp43 plasmid

Primer		Sequence (5′–3′)
AdUsp43	F	GCTAGCGATATCGGATCCGCCACCATGGATCCGGGCGTGG
	R	ACTAGTGGTACCAAGCTTGAAGCTGGACTCAGGTAAGGATTTCT
AdshUsp43-1	F	CCGGGCAACACCTGTTTCATGAACGCTCGAGCGTTCATGAAACAGGTGTTGCTTTTTG
	R	AATTCAAAAAGCAACACCTGTTTCATGAACGCTCGAGCGTTCATGAAACAGGTGTTGC
AdshUsp43-2	F	CCGGCAATGTGTATGCTTTCCAAGTCTCGAGACTTGGAAAGCATACACATTGTTTTTG
	R	AATTCAAAAACAATGTGTATGCTTTCCAAGTCTCGAGACTTGGAAAGCATACACATTG
AdshUsp43-3	F	CCGGGCTCCCAGTTCCAAGGCAATTCTCGAGAATTGCCTTGGAACTGGGAGCTTTTTG
	R	AATTCAAAAAGCTCCCAGTTCCAAGGCAATTCTCGAGAATTGCCTTGGAACTGGGAGC

Cell viability and lactate dehydrogenase (LDH) content assay

After viral infection and OGD/R, cell viability was assayed using the Cell counting kit-8 (CCK-8) assay kit (Cat#C0039, Bicentennial) following the steps in the instructions. LDH content was assayed using the LDH assay kit (Cat#C0016, Bicentennial) following the steps in the instructions.

Quantitative real-time PCR

Tissues or cells were lysed with Trizol, RNA was extracted with chloroform. The obtained RNA was reverse transcribed by Reverse Transcription Kit (Cat#R323-01, Novozymes) to obtain cDNA. Primers for the target genes were designed, and the cDNA was used as a template for qPCR (Cat#Q311-02, Novozymes). The β-actin was used as an internal reference to analyze the relative expression of the mRNA of target genes (Table 2).

Table 2.

Primer sequences for PCR

Gene	Species	Sequence (5′–3′)
Usp43	Mouse	F	CTCTGCCCGACATCCTGATC
		R	CATGGGGAGCCATGTTGAGT
Usp43	Rat	F	CATGGGAGTCTGCAAGCCTA
		R	CCTTGTGCTGTTGTTGAGGC
Tnf	Mouse	F	CATCTTCTCAAAATTCGAGTGACAA
		R	TGGGAGTAGACAAGGTACAACCC
Tnf	Rat	F	ATGGGCTCCCTCTCATCAGT
		R	GCTTGGTGGTTTGCTACGAC
Il6	Mouse	F	TAGTCCTTCCTACCCCAATTTCC
		R	TTGGTCCTTAGCCACTCCTTC
Il6	Rat	F	CCCAACTTCCAATGCTCTCCT
		R	TAGCACACTAGGTTTGCCGA
Il1b	Mouse	F	CCGTGGACCTTCCAGGATGA
		R	GGGAACGTCACACACCAGCA
Il1b	Rat	F	GACTTCACCATGGAACCCGT
		R	CAGGGAGGGAAACACACGTT
Cxcl2	Mouse	F	GCGCCCAGACAGAAGTCATA
		R	CAGTTAGCCTTGCCTTTGTTCA
Ccl2	Mouse	F	TACAAGAGGATCACCAGCAGC
		R	ACCTTAGGGCAGATGCAGTT
Bax	Mouse	F	TGAGCGAGTGTCTCCGGCGAAT
		R	GCACTTTAGTGCACAGGGCCTTG
Bax	Rat	F	AGGACGCATCCACCAAGAAG
		R	CAGTTGAAGTTGCCGTCTGC
Bcl2	Mouse	F	CAACAGGGAGATGTCACCCC
		R	TCAAACAGAGGTCGCATGCT
Bcl2	Rat	F	CTGGTGGACAACATCGCTCT
		R	GCATGCTGGGGCCATATAGT
β-actin	Mouse	F	GTGACGTTGACATCCGTAAAGA
		R	GCCGGACTCATCGTACTCC
β-actin	Rat	F	CCGCGAGTACAACCTTCTTG
		R	TGACCCATACCCACCATCAC

Western blotting

Tissues were lysed by adding appropriate amounts of RIPA lysate (65 mMTris-HCl pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% Nonidet P-40, 0.5% sodium deoxycholate and 0.1% SDS) containing protease inhibitor (Cat#0469312001, Roche) and phosphatase inhibitor (Cat#4906837001, Roche). Then, broken by sonication, centrifuged, and total protein was obtained after supernatant removal. Cells were lysed with SDS lysis buffer (50 mM Tris–HCl pH 6.8, 2% SDS, 10% glycerol) and then incubated at 95 °C for 15 min, after which they were centrifuged and the supernatant was taken to obtain total protein. The protein concentration was determined using BCA protein quantification kit (Cat#23225, Thermo). A protein sample of the same quality was added to the loading buffer and separated by 10% SDS-PAGE electrophoresis. Then, the protein was subsequently transferred to a 0.45 μm polyvinylidene difluoride membrane (Cat#IPVH00010, Millipore). The membranes were blocked at room temperature in a 5% skimmed milk solution for 1 h, and then the membrane was incubated with the appropriate primary antibody at 4 °C overnight. After washing to remove the primary antibody, the membrane was then incubated at room temperature with the corresponding secondary antibody (Jackson ImmunoResearch) for 1 h. Color was developed with ECL luminescent substrate (Cat#1705062, Bio-Rad) and signals were collected by a Burroughs gel imaging system (ChemiDoc XRS +). β-actin was used as the internal control. Images were analyzed by Image Pro Plus (version 6.0) software (Table 3).

Table 3.

Antibody information for western blotting

Antibody	Cat no.	Manufacturer
USP43	ER65281	Hua’an Biological
p-Ikkβ	2694	CST
Ikkβ	A0714	Abclonal
p-p65	3033	CST
p65	8242	CST
IκBα	4814	CST
Bcl2	3498	CST
Bax	2772	CST
Cleaved- Caspase3	9664	CST
p-ERK	4370	CST
ERK	4695	CST
p-JNK	4668	CST
JNK	9252	CST
p-p38	4511	CST
p38	8690	CST
p-TAK1	4508	CST
TAK1	5206	CST
p-ASK1	28846-1-AP	Proteintech
ASK1	28201-1-AP	Proteintech
Flag	M185-3L	MBL
HA	AE105	Abclonal
Myc	M047-3	MBL
β-actin	AC026	Abclonal

Plasmid construction

The full-length CDS sequences of USP43 and TAK1 were amplified from human cDNA library and ligated into pcDNA5-flag, pcDNA5-HA or GST-HA vectors to obtain the full-length overexpression plasmids (Flag-tagged, HA-tagged, GST-HA-tagged) of USP43 and TAK1. At the same time, we designed primers to construct a truncated plasmid of USP43 and TAK1 (Table 4).

Table 4.

Primer sequences for plasmid construct

Primer		Sequence (5′–3′)
Flag-USP43	F	TCGGGTTTAAACGGATCCATGGACCTGGGCCCCGGGG
	R	GGGCCCTCTAGACTCGAGTCAAAAGCTGGACTCCGG
HA-TAK1	F	TCGGGTTTAAACGGATCCATGTCTACAGCCTCTGCCGCCT
	R	GGGCCCTCTAGACTCGAGTCATGAAGTGCCTTGTCGTTTCT
GST-HA-TAK1	F	TCGGGTTTAAACGGATCCATGTCTACAGCCTCTGCCGCCT
	R	GGGCCCTCTAGACTCGAGTCATGAAGTGCCTTGTCGTTTCT
GST-HA-USP43	F	TCGGGTTTAAACGGATCCATGGACCTGGGCCCCGGGG
	R	GGGCCCTCTAGACTCGAGTCAAAAGCTGGACTCCGG
Flag-USP43 (1–712)	F	TCGGGTTTAAACGGATCCATGGACCTGGGCCCCG
	R	GGGCCCTCTAGACTCGAGCCGCTTCTGATAGAACAGGATATAAGCC
Flag-USP43 (713–1123)	F	TCGGGTTTAAACGGATCCAACAGCATCCCTCCCTGGT
	R	GGGCCCTCTAGACTCGAGAAAGCTGGACTCCGGTAAGGT
Flag-TAK1	F	TCGGGTTTAAACGGATCCATGTCTACAGCCTCTGCCGCCT
	R	GGGCCCTCTAGACTCGAGTCATGAAGTGCCTTGTCGTTTCT
Flag-TAK1 (1–300)	F	AGAGCCCGGGCGGATCCATGTCTACAGCCTCTGCCGC
	R	TCGACGAATTGCTCGAGTCAATACTGTAATGGCTCATCTGCTCC
Flag-TAK1 (301–579)	F	AGAGCCCGGGCGGATCCCCTTGTCAGTATTCAGATGAAGGAC
	R	TCGACGAATTGCTCGAGTCATGAAGTGCCTTGTCGTTTCT

Immunoprecipitation

293 T cells were transfected with the desired plasmids or primary neurons were infected with Usp43 overexpressing adenovirus. The cells were lysed with IP lysis buffer (20 mM Tris–HCl, pH 7.4; 150 mM NaCl; 1 mM EDTA and 1% NP-40) 24 h after transfection. After high-speed centrifugation at 4 °C, the protein-containing supernatant was incubated with protein G agarose beads and the indicated anti-tag antibodies or IgG antibody in IP overnight at 4 °C. Centrifugation was performed at 3,000 rpm at 4 degrees Celsius, the agarose beads were washed about three times with buffer containing 300 mM and 150 mM NaCl, respectively. The agarose beads were resuspended with 2 × SDS uploading buffer and boiled at 95 °C for 5–10 min, and then analyzed by Western blot assay.

Glutathione-S-transferase pull-down assay

GST-HA-TAK1 (or USP43) and Flag-USP43 (or TAK1) proteins were overexpressed in 293 T cells, then the cells were lysed using lysis buffer, after that GST bead was added for immunoprecipitation. The enriched proteins and the protein lysates without co-precipitation were subjected to SDS-PAGE electrophoresis after washing the magnetic beads 3 times with IP buffer.

Ubiquitination experiments

Flag-USP43 and HA-TAK1 plasmids, as well as Myc-ubiquitin molecules with different ubiquitination sites were co-transfected into 293 T cells. Following the collection of the cells, they were then heated at 95 °C for 5–10 min under 1% SDS. The cells lysate was prepared by ultrasonication, utilizing IP buffer. Immunoprecipitation was performed by adding the HA antibody. The enriched proteins were subjected to SDS-PAGE electrophoresis.

Statistical analysis

All data were analyzed using SPSS 21.0 statistical software (IBM Corp.). Data were expressed as the mean ± SD. Shapiro–Wilk analysis was used to test data normality. If the data were normally distributed, comparisons among multiple groups were performed via One way-ANOVA tests followed by Bonferroni post hoc analysis or Tamhane's T2 analysis. Differences between the two groups were determined with an unpaired Student's t- test. Nonnormally distributed data were compared using Mann–Whitney U tests. p < 0.05 was considered significant.

Results

Increased expression of USP43 during cerebral I–R injury

It has been shown that neuroinflammatory mechanisms significantly lead to neuronal injury in the acute phase of cerebral ischemia, ultimately inducing brain injury and neurological deficits through ischemia/reperfusion (I–R) injury [25, 26]. To investigate the relationship between USP43 and cerebral ischemia–reperfusion injury, a mouse tMCAO/R model which simulated the condition in humans was established. The brain tissue of the modelled mice was then collected for molecular biology and pathological experiments (Fig. 1a). After 24 h and 48 h reperfusion of t/MCAO, F4/80 positive cell significantly increased (Fig. 1b). Then the levels of mRNA of tumor necrosis factor (Tnf), interleukin 6 (Il6) and C–C motif chemokine ligand 2 (Ccl2) were examined, it was found that after 24 h and 48 h reperfusion of t/MCAO, the expression levels of Tnf, Il6 and Ccl2 were elevated (Fig. 1c). The above results validated the success of our animal model construction. To explore the expression of Usp43 in the injury process, mRNA and protein levels in mouse brain tissue were examined after 24 and 48 h of t/MCAO reperfusion. It was found that at these two time points, RNA levels did not change, while protein levels increased (Fig. 1d, e). In vitro, primary neuron was used to build OGD/R models. It was found that after OGD/R stimulation, the mRNA levels of Tnf and Il6 were elevated (Fig. 1f). This suggests that our cellular model construction was successful. In primary neuronal cells, it was similarly found that RNA levels did not change, while protein levels increased after OGD/R treatment (Fig. 1g, h). These results support a possible association between Usp43 and I–R injury of brain.

Fig. 1.

Increased expression of ubiquitin specific peptidase 43 (USP43) during cerebral ischemia–reperfusion (I–R) injury. a A median neck incision was made to expose the left common carotid artery, internal carotid artery and external carotid artery. A 6–0 silicone-coated monofilament wire was inserted into the left external carotid artery, advanced toward the internal carotid artery, and wedged into the cerebral artery ring to block the beginning of the middle cerebral artery (MCA). After 45 min of ischemia, the wire bolus was slowly withdrawn to restore the middle cerebral artery blood flow. Samples were taken 24 h or 48 h after reperfusion. The left hemisphere of the mouse brain was taken as material for qPCR and western blot while both hemispheres were taken for pathological examination after the olfactory bulb and cerebellum were removed. Created in BioRender. dong, y. (2025) https://BioRender.com/61efs9j. b F4/80 staining showed macrophages/microglia in cerebral tissues at 24 and 48 h after reperfusion of temporary middle cerebral artery occlusion (t/MCAO). F4/80 (red), DAPI (blue). Right, F4/80- positive cells were quantified. n = 4 mice per group, Scale bar, 50 μm. c RT-qPCR showed pro-inflammatory factor mRNA levels of t/MCAO mice at 24 and 48 h after reperfusion, n = 4 mice per group. d RT-qPCR showed mRNA levels of Usp43 at 24 and 48 h after reperfusion, n = 4 mice per group. e Western blot analysis and quantification showed that USP43 protein levels at 24 and 48 h after I–R injury compared to the sham group, n = 4 mice per group. f RT-qPCR showed the mRNA levels of pro-inflammatory factors in rat primary neurons with hypoxia and glucose deficiency (OGD) for 3 h and reoxygenation (OGD/R) for 6 h, n = 3 independent cell cultures. g RT-qPCR showed mRNA levels of Usp43 in rat primary neurons after 6 h of OGD/R, n = 3 independent cell cultures. h Western blot analysis and quantification showed that USP43 expression in primary neurons at 6 h OGD/R, n = 3 independent cell cultures. n.s. p > 0.05 vs. Sham or Control, *p < 0.05 vs. Sham or Control, **p < 0.01 vs. Sham or Control

Usp43 promotes brain injury, inflammatory response and apoptosis during cerebral I–R injury

Given the increased expression of Usp43 in cerebral I–R injury, to further determine whether Usp43 plays a role in regulating cerebral I–R injury, the model was established by performing t/MCAO surgery using USP43 knockout mice and littermate negative mice. Usp43 knockout mice were validated by western blot (Fig. 2a). Under normal conditions, knockout of USP43 does not result in alterations to neurological scores or the ischemic area in mice. Compared with the sham group, the mice had higher neurological scores and a larger ischemic area after 24 h reperfusion of t/MCAO. By contrast, USP43-KO mice had lower neurological scores and a smaller ischemic area than WT mice (Fig. 2b–d). To uncover the role of Usp43 in neuroinflammation, F4/80 immunofluorescence staining was performed on mouse brain tissue. F4/80-positive cells were significantly elevated in the brain tissue of mice after 24 h reperfusion of t/MCAO compared to the Sham group. In contrast, the number of positive cells was found to be diminished in the brain tissue of USP43-KO mice in comparison to WT mice (Fig. 2e). Meanwhile, the levels of Tnf, Il6, interleukin 1β (Il1b), Ccl2 and chemokine (C-X-C motif) ligand 2 (Cxcl2) increased after 24 h reperfusion of t/MCAO. Compared with the WT group, the expression of the above inflammatory factors decreased in the brain tissue of USP43-KO mice (Fig. 2f). Furthermore, the same results were observed in serum cytokine assays (Fig. 2g). Protein levels of molecules related to the nuclear factor-kappa-B (NF-κB) signaling pathway were also examined. After t/MCAO treatment, the NF-κB signaling pathway was activated as evidenced by a significant increase in the phosphorylation of inhibitor of nuclear factor kappa-B kinase beta (Ikkβ) and nuclear factor NF-kappa-B p65 (p65), and a significant decrease in inhibitor of nuclear factor kappa- Bα (IκBα). In contrast, this activation was significantly reduced in the brain tissue of mice in the USP43-KO group (Fig. 2h). Furthermore, the absence of USP43 does not affect brain tissue inflammation at the baseline level. Therefore, Usp43 deficiency was proved that can inhibit the activation of the NF-κB signaling pathway and alleviate the inflammatory response during cerebral I–R injury. After ischemia, hypoperfusion of brain tissue leads to a decrease in oxygen, ATP, and glucose, which leads to cell death over time. Cell death is exacerbated by increased oxygen free radicals and hyperactivation of the inflammatory response after reperfusion [10, 27]. Therefore, the possible role of USP43 in cell death was investigated. Immunofluorescence results demonstrated an increase in TUNEL-positive cells after 24 h reperfusion of t/MCAO in comparison with the Sham group. However, a significant decrease in the number of positive cells was observed in the KO group compared to the WT group (Fig. 2i). Through RT- PCR and western blotting, levels of pro-apoptotic molecules, such as BCL2-associated X protein (Bax) and cleaved Caspase3 (C-Caspase3), were elevated in the t/MCAO group of mice in comparison to the Sham group. However, these levels were lower in USP43-KO mice than in WT mice. In contrast, the levels of anti-apoptotic molecules such as B-cell lymphoma-2 (Bcl2) were reduced in brain tissues after t/MCAO compared to the Sham group, and were higher in the USP43-KO group than in the WT group (Fig. 2j, k). Moreover, the deletion of USP43 does not affect apoptosis in brain tissue at baseline. In summary, Usp43 ablation reduces cells apoptosis in cerebral I–R injury.

Fig. 2.

USP43 promotes brain injury, inflammatory response and apoptosis during cerebral I–R injury. a Usp43-KO (knockout) mice were verified by Western blotting, n = 3 mice per group. b Neurological score was performed of the wild type (WT) mice and Usp43-KO mice, n = 12 mice per group. c TTC staining was performed on WT and Usp43-KO mice in the sham and I–R groups, n = 6 mice per group. d Quantitative analysis of infarct volume of WT and Usp43-KO mice, n = 6 mice per group. e F4/80 staining showed macrophages/microglia in brain sections of WT and Usp43-KO mice. F4/80 (red), DAPI (blue). Right, F4/80-positive cells were quantified. n = 4 mice per group. Scale bar, 50 μm. f Pro-inflammatory factor mRNA levels were examined in WT and Usp43-KO mice, n = 4 mice per group. g ELISA showed the levels of serum cytokines in WT and Usp43-KO mice, n = 6 mice per group. h Western blotting reflected the relative level of NF-κB pathway-related proteins in WT and Usp43-KO mice, n = 3 mice per group. i TUNEL staining demonstrated apoptotic cells in WT and Usp43-KO group. TUNEL (green), DAPI (blue). Right, TUNEL-positive cells were quantified. n = 4 mice per group, Scale bars, 50 μm. j RT-PCR showing the mRNA levels of apoptosis-related factors in WT and Usp43-KO mice, n = 4 mice per group. k Western blotting reflected the protein levels of apoptosis-related factors in Usp43-KO and WT mice. n = 3 mice per group. #p < 0.05, ##p < 0.01 vs. WT Sham, and *p < 0.05, **p < 0.01 vs. WT I–R

USP43 knockdown ameliorates neuron injury, inflammatory response and apoptosis during OGD/R

Neurons are the most vulnerable cell types in the brain during ischemia–reperfusion, so we wanted to further investigate the function of Usp43 in neurons. Primary neurons were infected with Usp43 knock down adenovirus (AdshUsp43) or its control adenovirus (AdshRNA). Usp43 knock down were confirmed by western blotting (Fig. 3a). It was showed that OGD/R treatment reduced cell activity and promoted the production of LDH, while neurons in AdshUsp43 OGD/R group exhibited higher cell viability than the shRNA OGD/R group, and the cells in AdshUsp43 OGD/R group released less LDH than the shRNA OGD/R group (Fig. 3b, c). Treatment with OGD/R resulted in increased transcript levels of Tnf, Il6 and Il1b compared to the Control group. A decline in elevation was observed in the AdshUsp43 OGD/R group in comparison with the AdshRNA OGD/R group (Fig. 3d). Furthermore, OGD/R treatment caused NF-κB pathway activation, which was inhibited by USP43 knockdown (Fig. 3e). Similarly, by examining mRNA and protein expression, it was found that the expression of pro-apoptotic molecules was elevated and the expression of inhibitory apoptotic molecules was decreased after OGD/R, and USP43 knockdown suppressed this trend. (Fig. 3f, g). The findings from series of experiments demonstrated that the knockdown of USP43 did not result in alterations to cell viability, inflammation or apoptosis in untreated OGD/R neurons. As the result, Usp43 knockdown plays a protective role in OGD/R-stimulated neurons.

Fig. 3.

USP43 knockdown ameliorates cell injury, inflammatory response and apoptosis during OGD/R. a Usp43 knockdown in primary neurons was validated by western blotting, n = 3 mice per group. b Cell counting kit-8 (CCK-8) assays showed cell viability of neurons infected with AdshUsp43 (Usp43 knockdown primary neurons) and AdshRNA after OGD/R treatment. n = 3 of independent cell culture preparations. c Lactate dehydrogenase (LDH) kit showed LDH released by primary neurons infected with AdshRNA or AdshUsp43 after OGD/R treatment. n = 3 of independent cell culture preparations. d The mRNA levels of proinflammatory factors in primary neurons infected with AdshRNA or AdshUsp43 at 6 h after OGD/R, n = 4 of independent cell culture preparations. e. Western blotting reflected the relative level of NF-κB pathway-related proteins in primary neurons infected with AdshRNA or AdshUsp43 at 6 h after OGD/R, n = 3 of independent cell culture preparations. f RT-PCR showing the mRNA levels of apoptosis-related factors in primary neurons infected with AdshRNA or AdshUsp43 at 6 h after OGD/R compared with the control, n = 4 of independent cell culture preparations. g Western blotting reflected the protein levels of apoptosis-related factors in primary neurons infected with AdshRNA or AdshUsp43 at 6 h after OGD/R. n = 3 of independent cell culture preparations. #p < 0.05, ##p < 0.01vs. AdshRNA Control group, and *p < 0.05, **p < 0.01 vs. AdshRNA OGD/R group

USP43 overexpression aggravates neuron injury, inflammatory response and apoptosis during OGD/R

To further validate the role of Usp43 in neurons, primary neurons were infected with USP43 overexpressed adenovirus (AdUsp43) or its control adenovirus (AdGfp). Usp43 overexpression was validated by western blotting (Fig. 4a). Similarly, OGD/R treatment decreased neuronal cell viability and released more LDH, while USP43 overexpression was found to exacerbate this tendency. (Fig. 4b, c). In addition, as shown as the results, Usp43 overexpression exacerbated the upregulation of the transcription levels of Tnf, Il6 and Il1b after OGD/R stimulation (Fig. 4d). USP43 overexpression was also found to enhance activation of the NF-κB signaling pathway (Fig. 4e). Besides, Usp43 overexpression promoted an increase in the expression levels of Bax and C-Caspase3, as well as a decrease in the expression level of Bcl2 in OGD/R-treated group (Fig. 4f, g). The above experimental results indicated that USP43 overexpression does not exert an effect on cell viability, inflammation, or apoptosis at the baseline level. Therefore, Usp43 overexpression promotes neurons injury, inflammatory response and apoptosis in primary neurons during OGD/R.

Fig. 4.

USP43 overexpression aggravates cell injury, inflammatory response and apoptosis during OGD/R. a Usp43 over-expression in primary neurons was validated by western blotting, n = 3 mice per group. b CCK-8 assays showed cell viability of neurons infected with AdUsp43 (Usp43 overexpression primary neurons) and AdGfp after OGD/R treatment. n = 3 of independent cell culture preparations. c LDH kit showed LDH released by primary neurons infected with AdGfp or AdUsp43 after OGD/R treatment. n = 3 of independent cell culture preparations. d The mRNA levels of proinflammatory factors in primary neurons infected with AdGfp or AdUsp43 at 6 h after OGD/R, n = 4 of independent cell culture preparations. e. Western blotting reflected the relative level of NF-κB pathway-related proteins in primary neurons infected with AdGfp or AdUsp43 at 6 h after OGD/R, n = 3 of independent cell culture preparations. f RT-PCR showing the mRNA levels of apoptosis-related factors in primary neurons infected with AdGfp or AdUsp43 at 6 h after OGD/R compared with the control, n = 4 of independent cell culture preparations. g Western blotting reflected the protein levels of apoptosis-related factors in primary neurons infected with AdGfp or AdUsp43 at 6 h after OGD/R. n = 3 of independent cell culture preparations. #p < 0.05, ##p < 0.01 vs. AdGfp Control group, and *p < 0.05, **p < 0.01 vs. AdGfp OGD/R group

Usp43 regulates the TAK1-JNK/p38 signaling pathway during cerebral I–R injury

The MAPK pathway regulates inflammatory response and apoptosis and plays an important role in cerebral ischemia–reperfusion [28, 29], so we speculated that Usp43 might be involved in modulating the MAPK pathway. To further explore the mechanism of action of Usp43, western blotting was used to test total and phosphorylation levels of extracellular regulated protein kinases (ERK), c-Jun N-terminal kinase (JNK), and p38 mitogen-activated protein kinases (p38) in mice brain tissues and primary neurons. Under normal conditions, knockdown or overexpression of USP43 does not cause changes in the expression of MAPK signaling pathway proteins. The phosphorylation levels of ERK, JNK and p38 were elevated in the brain tissues of mice in the surgical group compared to the Sham group (Fig. 5a). In vitro, similar results have been observed (Fig. 5b, c). In vivo, the phosphorylation levels of JNK and p38 were significantly lower in the KO group than in the WT group after t/MCAO, but there was no difference in ERK phosphorylation levels (Fig. 5a). Besides, the same results were observed in the AdshUsp43 group, while the opposite results were observed in the USP43 overexpression group after ODG/R treatment, in vitro (Fig. 5b, c). Therefore, Usp43 could promote cerebral I–R injury by increasing phosphorylation of JNK and p38. TAK1 and ASK1 are upstream molecules of JNK and p38, and play an important role in the MAPK signaling pathway. For this reason, the total and phosphorylation levels of TAK1 and ASK1 were detected by western blotting. The phosphorylation levels of TAK1, and ASK1 were elevated in the brain tissues of mice in the surgical group compared to the Sham group (Fig. 5d). In vitro, similar results have been observed (Fig. 5e, f). The total and phosphorylation levels of TAK1 were lower in the KO group than that in the WT group after t/MCAO surgery, but there was no difference in ASK1 total and phosphorylation levels (Fig. 5d). Meanwhile, there are the same results in the AdshUsp43 group and the opposite results in the Usp43 overexpression group after ODG/R stimulation (Fig. 5e, f).

Fig. 5.

USP43 regulates the TAK1-JNK/p38 signaling pathway during cerebral I–R injury. a Western blotting showed total and phosphorylation levels of c-Jun N-terminal kinase (JNK), extracellular signal-regulated kinase (ERK) and p38 in WT and Usp43-KO mice at 24 h after I–R. n = 3 mice per group. b Western blotting showed the phosphorylation levels and total protein levels of JNK, ERK and p38 in primary neurons infected with AdshRNA and AdshUsp43 at 6 h after OGD/R. n = 3 of independent cell culture preparations. c Western blotting showed the phosphorylation levels and total protein levels of JNK, ERK and p38 in primary neurons infected with AdGfp and AdUsp43 at 6 h after OGD/R. n = 3 of independent cell culture preparations, n.s. p > 0.05. d Western blotting showed total and phosphorylation levels of TAK1 and Ask1 in WT and Usp43-KO mice at 24 h after I–R. n = 3 mice per group. e Western blotting showed total and phosphorylation levels of TAK1 and Ask1 in primary neurons infected with AdshRNA and AdshUsp43 at 6 h after OGD/R. n = 3 of independent cell culture preparations. f Western blotting showed total and phosphorylation levels of TAK1 and Ask1 in primary neurons infected with AdGfp and AdUsp43 at 6 h after OGD/R. n = 3 of independent cell culture preparations. #p < 0.05, ##p < 0.01 vs. WT Sham, AdshRNA Control or AdGfp Control group, and n.s. p > 0.05, **p < 0.01 vs. WT I–R, AdshRNA OGD/R or AdGfp OGD/R group

USP43 overexpress promotes cerebral I–R injury via TAK1- JNK/p38 pathway

To further determine whether Usp43 overexpression induced I–R damage is mediated by TAK1 activation, a specific TAK1 inhibitor was used to inhibit TAK1 function in USP43 overexpression neurons. It was observed that inhibitor reduced the phosphorylation levels of TAK1, JNK and p38 in the Usp43 overexpression group after OGD/R (Fig. 6a). Compared with the dimethyl sulfoxide (DMSO)-treated group, the overexpressing neurons in the inhibitor-treated group showed better cell activity and lower LDH content (Fig. 6b, c). It was revealed that the transcription level of Tnf, Il1b, and Il6 were reduced in the inhibitor-treated group (Fig. 6d). In addition, inhibitor abolished the promoting effect of Usp43 overexpression on the activation of NF-κB signaling pathway during OGD/R stimulation (Fig. 6e). It was also observed that the regulation of Usp43 overexpression on the expression of Bax, C-Caspase3 and Bcl2 was rescued with inhibitor treated (Fig. 6f, g). These results suggest that inhibition of TAK1 could prevent the regulation effects of Usp43 overexpression on OGD/R-treated neurons.

Fig. 6.

USP43 overexpress promotes cerebral I–R injury via TAK1- JNK/p38 pathway. a Western blotting reflected the phosphorylation levels of TAK1, JNK and p38 in primary neurons infected with AdGfp or AdUsp43 and then treated with dimethyl sulfoxide (DMSO) or iTAK1 1 h before exposure to OGD/R for 6 h, n = 3 of independent cell culture preparations. b CCK-8 assays showed the cell viability in primary neurons infected with AdGfp or AdUsp43 and then treated with DMSO or iTAK1 1 h before exposure to OGD/R for 6 h, n = 3 of independent cell culture preparations. c LDH kit showed LDH released by primary neurons infected with AdGfp or AdUsp43 and then treated with DMSO or iTAK1 1 h before exposure to OGD/R for 6 h. n = 4 of independent cell culture preparations. d RT-qPCR showed the mRNA levels of proinflammatory factors in primary neurons infected with AdGfp or AdUsp43 and then treated with DMSO or iTAK1 1 h before exposure to OGD/R for 6 h, n = 4 of independent cell culture preparations. e Western blotting reflected the relative level of NF-κB pathway-related proteins in primary neurons infected with AdGfp or AdUsp43 and then treated with DMSO or iTAK1 1 h before exposure to OGD/R for 6 h, n = 3 of independent cell culture preparations. f RT-PCR showing the mRNA levels of apoptosis-related factors in primary neurons infected with AdGfp or AdUsp43 and then treated with DMSO or iTAK1 1 h before exposure to OGD/R for 6 h, n = 4 of independent cell culture preparations. g Western blotting reflected the relative level of apoptosis-related proteins in primary neurons infected with AdGfp or AdUsp43 and then treated with DMSO or iTAK1 1 h before exposure to OGD/R for 6 h, n = 3 of independent cell culture preparations. #p < 0.05, ##p < 0.01 vs. AdGfp group treated with DMSO (AdGfp + DMSO), and *p < 0.05, **p < 0.01 vs. AdUsp43 group treated with DMSO (AdUsp43 + DMSO)

USP43 binds to TAK1 and inhibits the degradation of TAK1

To further clarify the mechanism by which USP43 regulates the expression and phosphorylation of TAK1, Flag-tagged USP43 and HA-tagged TAK1 were overexpressed in HEK293T cells, respectively or simultaneously. IP experiments demonstrated that USP43 co-immunoprecipitated with TAK1 and vice versa (Fig. 7a). In Usp43-overexpressed primary neurons, TAK1 also binds to USP43 (Fig. 7b). In addition, it was found that USP43 and TAK1 have co-localized regions in the cytoplasm (Fig. 7c). Furthermore, the GST pull-down assay confirmed a direct interaction between USP43 and TAK1 (Fig. 7d). Molecular mapping analyses were then performed to identify protein structural domains associated with USP43 and TAK1 binding. The results suggest that the N-terminus of TAK1 [TAK1(1-300)] is required for its direct interaction with USP43. In contrast, both segments 1-712 and 713-1123 of USP43 have structural domains that can bind to TAK1 (Fig. 7e). Subsequently, neurons were infected with different concentrations of AdUsp43 and stimulated with OGD/R, the amount of TAK1 protein increased as the amount of infection increased (Fig. 7f). Besides, neurons infected with Usp43 overexpression adenovirus or control virus and stimulated by OGD/R were treated with the protein synthesis inhibitor cycloheximide (CHX) for different periods of time, it was found that the content of TAK1 in the Usp43 overexpression group was always higher than that in the control group (Fig. 7g). This result indicates that USP43 inhibits the degradation of TAK1.

Fig. 7.

USP43 interacts with TAK1 and inhibits its degradation. a Co-IP assays were performed to show the interaction between USP43 and TAK1 in 293 T cells transfected with the indicated plasmids. b USP43 in primary neurons infected with AdUsp43 was precipitated by Flag antibody. The interaction between endogenous USP43 and TAK1 was examined by Western blot analysis. c USP43 and TAK1 were transfected into 293 T cells for immunofluorescence assays. Colocalization analysis showed that USP43 and TAK1 colocalized in the cytoplasm. Scale bars, 7.5 μm. d GST pull-down assays showed the direct interaction of USP43 and TAK1 in 293 T cells transfected with the indicated plasmids. e 293 T cells were transfected with truncated mutants of USP43 or TAK1 and a full-length plasmid of another molecule. The results of Co-IP experiments revealed that the N-terminus of TAK1 [TAK1(1–300)] is its direct interaction with USP43. Both segments 1–712 and 713–1123 of UPS43 have structural domains that can bind to TAK1. f Primary neurons were infected with increasing concentration of AdUsp43 and increased TAK1 concentration was detected by Western blot analysis. g Primary neurons were infected with AdGfp or AdUsp43. Cells were treated with 50 μM CHX for 0, 6, 12 h and then harvested at the indicated time points. Western blot analysis showed prolonged TAK1 half-life in neurons infected with AdUsp43

USP43 inhibits the K48-linked ubiquitination degradation of TAK1

The proteasome pathway and the lysosomal pathway are two important pathways for protein degradation. Our results showed that the proteasome inhibitor MG132, but not lysosomal inhibitors chloroquine (CQ), reversed the degradation of TAK1 in AdshUsp43 infected neurons (Fig. 8a). This finding suggests that USP43 mediated regulation of TAK1 homeostasis largely depends on the proteasome system. Since USP43 is a member of the deubiquitinating enzyme family, we hypothesize that USP43 affects the function of TAK1 through deubiquitination. It was found that USP43 interacted with TAK1 and inhibited the ubiquitination on TAK1 (Fig. 8b). Moreover, USP43 dramatically decreased the K48-linked ubiquitination on TAK1 indicating that USP43 predominantly removes the K48-linked ubiquitination on TAK1 (Fig. 8c). USP43 reduced the ubiquitination of WT and K48O, but it failed to remove the ubiquitination of the K48R mutant (Fig. 8d). Furthermore, it was found that the ubiquitinating enzyme activity mutation of Usp43 (Usp43-M) nullified the regulatory effect of Usp43 overexpression on TAK1, JNK and p38 stimulated by OGD/R (Fig. 8e). Meanwhile, neurons that overexpressed Usp43-M showed higher cell viability and lower LDH content compared with Usp43 overexpressed group after OGD/R (Fig. 8f, g). Consistently, inhibition of the enzymatic activity of Usp43 was observed that could reverse the promoting effects of inflammatory response and apoptosis induced by USP48 overexpression (Fig. 8h, i). This suggests that the role of USP43 in brain I–R damage depends on its deubiquitinating enzyme activity.

Fig. 8.

USP43 affects the stability of TAK1 through inhibiting K48-linked ubiquitination. a Primary neurons were infected with AdshRNA or AdshUsp43. Cells were treated with 25 μM MG132 or CQ or DMSO and 50 μM CHX for 12 h before harvest, and then TAK1 was detected by Western blot analysis. b–d 293 T cells were transfected with the corresponding plasmids, ubiquitination experiments were performed to analyze polyubiquitination of TAK1. e Western blotting reflected the phosphorylation levels of TAK1, JNK and p38 in primary neurons infected with AdGfp, AdUsp43 or AdUsp43 mutant, n = 3 of independent cell culture preparations. f CCK-8 assays showed the cell viability in primary neurons infected with AdGfp, AdUsp43 or AdUsp43 mutant, n = 3 of independent cell culture preparations. g LDH kit showed LDH released by primary neurons infected with AdGfp, AdUsp43 or AdUsp43 mutant. n = 3 of independent cell culture preparations. h RT-qPCR showed the mRNA levels of proinflammatory factors in primary neurons infected with AdGfp, AdUsp43 or AdUsp43 mutant, n = 4 of independent cell culture preparations. i RT-qPCR showing the mRNA levels of apoptosis-related factors in primary neurons infected with AdGfp, AdUsp43 or AdUsp43 mutant, n = 4 of independent cell culture preparations. **p < 0.01 vs. the AdGfp group, ##p < 0.01 vs. AdUsp43 group

Discussion

Although USP43 has been reported to be associated with a variety of cancers and related to their prognosis, its role in the neurological field has not been investigated. Our study identified for the first time the role of USP43 in cerebral ischemia–reperfusion injury and related mechanisms. USP43 KO mice were utilized in our study, which exhibited smaller infarct volumes and milder neurological deficits. In vitro, primary cortical neurons were cultured and subjected to OGD/R. USP43 knockdown ameliorated neuronal inflammation and apoptosis in primary neurons, while USP43 overexpression exacerbated these processes. Mechanistically, we found that USP43 activates the TAK1-JNK/p38 axis. USP43 regulates TAK1 protein stability by interacting directly with TAK1 and acting as a deubiquitinating enzyme in the K48 ubiquitin chain. In conclusion, we suggest that USP43 is a key molecule in the regulation of cerebral ischemia–reperfusion injury.

USP43 has been shown to be up-regulated in cancer tissues and cells such as bladder, breast, cervical, osteosarcoma and pancreatic ductal adenocarcinoma [15, 16, 18–20, 30]. And the upregulation of USP43 was associated with cancer proliferation, migration and poor prognosis. In our study, we found that USP43 protein expression was increased in both t/MCAO and neuronal OGD/R models. We suggest that USP43 is a disease-stimulating molecule that is up-regulated in response to organismal responses in the presence of disease-causing factors. Our results showed no significant changes in USP43 at the mRNA level and a significant elevation at the protein level. This suggests to us that USP43 may be subject to the regulatory role of a post-translational modification. USP43 may be upregulated by deubiquitinating enzymes when pathogenic factors are present, leading to changes in its protein stability. But the specific mechanisms involved are still unclear and need to be explored in further research.

USP43 knockout mice were used in our study to construct the t/MCAO model, and we found that USP43 can affect neural function, increase ischemic volume, and activate inflammatory responses and apoptotic processes in mice. However, it is not clear what kind of cells play a role in ischemia–reperfusion injury. Since neuronal energy metabolism is almost entirely dependent on aerobic glycolysis, when the supply of oxygen and glucose is terminated by inadequate blood supply, neuronal cells are the most vulnerable. Therefore, the protection of neurons is crucial in cerebral ischemia–reperfusion injury. Mao et al. found that LIG protects hippocampal neurons from cerebral I–R injury [31]. Wang et al. found that ferroptosis could be inhibited by targeting CHAC1 in neurons [32]. Elimination of damaged mitochondria by mitophagy is beneficial in protecting against neuronal death [33]. Jin et al. found that the STAT3/HIF-1α/PTRF axis in neurons modulates brain I–R injury by regulating the activity and stability of PLA2G4A [34]. It is demonstrated that Shank3 regulates neuronal oxidative stress and inflammatory activity after I–R via the STIM1/Nrf2 pathway [35]. In vitro, we isolated primary rat neurons and found that USP43 promotes inflammatory responses and neuronal apoptosis using the OGD/R model. This suggests that neurons are indeed important target cells for the cerebral ischemia–reperfusion process. The limitation of this study is that in vivo, we did not specifically study the role of USP43 in neurons. To further clarify the global role of neuronal injury on cerebral ischemia–reperfusion, in subsequent studies we will further explore the role of USP43 in vivo using neuron-specific knockout adeno-associated virus or neuron-specific knockout mice.

Our study revealed a direct interaction of USP43 with TAK1. TAK1 belongs to the mitogen-activated protein kinase kinase kinase (MAP3K) family, which is a key molecule in the classical MAPK signaling pathway and plays an important role in cerebral ischemia–reperfusion injury. Larochelle et al. find that RIPK2 deficiency leads to downregulation of TAK1 and inhibits neuroinflammatory responses after stroke injury [36]. Xia et al. found that TRIM45 promotes TAK1 activation and activates the NF-κB signaling pathway [37]. Our previous study also found that TAK1 and ASK1, as members of the MAP3K family, play an important role during cerebral ischemia–reperfusion injury [38, 39]. It has been found that the injection of TAK1 inhibitor in the early stage of ischemia can improve the memory function and reduce the ischemic volume of mice [40, 41]. Our study found that the inflammatory response and neuronal apoptosis aggravated by USP43 overexpression were effectively ameliorated using TAK1 inhibitors. Thus, TAK1 is an activating molecule for cerebral ischemia–reperfusion injury, and USP43, as an upstream activating molecule for TAK1, also has an activating effect on cerebral ischemia–reperfusion.

The post-translational modification of TAK1 has been reported to play an important role in multiple inflammation-related diseases. It has been demonstrated that OTUD5 can inhibit TAK1-mediated podocyte inflammatory response and damage by interacting with TAK1 and reducing the polyubiquitination of TAK1 linked to K63 at residue K158 [44]. TIGAR promotes TAK1 ubiquitination by inducing the interaction of TRAF6 with TAK1, thereby activating the NF-κB signaling pathway to accelerate the progression of sepsis [45]. USP18 catalyzed the deubiquitination of the TAK1-TAB1 complex, regulates T cell activation and Th17 production [46]. TRIM56 enhances ubiquitination of TAK1, in particular the polyubiquitin chain linking TAK1 to M1, leading to a tight interaction of the TAK1-IKKα complex, which positively regulates TNFα-induced NF-κB signaling [47]. Down-regulation of THBS1 destabilizes TAK1 by activating NEDD4-mediated ubiquitination of K48-linked TAK1, promoting necrosis and release of DAMPs from trophoblast cells [48]. CHIP positively regulates NF-κB signaling by targeting TAK1 and enhancing its ubiquitination to the K63 linkage [49]. Upon overexposure to ROS, Roquin-2 promotes ubiquitination and degradation of TAK1, which inhibits the activation of Nrf2, thereby contributing to efficient cell death [50]. All of the above studies demonstrate that ubiquitination modifications play an important role in the regulation of protein stability in TAK1. Our study suggests that USP43 interacts with TAK1 through deubiquitination of the K48 link during cerebral ischemia–reperfusion injury, activating the downstream JNK/p38 signaling pathway. It has been shown that USP43 is an important regulatory molecule at the level of TAK1 protein during cerebral ischemia–reperfusion injury.

USP43 has been shown to be located upstream of multiple signaling pathways and is involved in the regulation of protein stability of several molecules. In bladder tumor, USP43 increased c-Myc protein levels in a dose-dependent manner. USP43 deubiquitinates c-Myc at the K148 and K289 sites [19]. In colorectal cancer SW480 cell line, USP43 can stabilize ZEB1 protein by deubiquitination [15]. In breast cancer, USP43 mediates Cav2.2 function in cortical actin stabilization, invasive pseudopod formation, ECM degradation and metastasis through deubiquitination functions [51]. In ovarian cancer, USP43 inhibited HDAC2 degradation through its deubiquitination function, activated the HDAC2-mediated Wnt/β-catenin signaling pathway, and reduced the sensitivity of EOC cells to cisplatin [20]. In cervical cancer, USP43 prolongs TAZ protein stability and activates the Hippo/TAZ pathway to promote CC cell proliferation, migration, invasion and epithelial-to-mesenchymal transition [30]. Our study also found that USP43 regulates TAK1 protein stability by exerting its function as a deubiquitinating enzyme. We designed USP43 enzyme activity mutant adenoviruses with which we infected primary neurons and found that the upregulation of TAK1 by USP43 was lost. The USP43 protein with mutated enzymatic activity was unable to activate the downstream TAK1-p38/JNK pathway, causing inflammatory response and neuronal apoptosis. USP43 can be used as a potential intervention site for acute stroke treatment.

Conclusions

In summary, our study identifies USP43 as a novel regulator in CIRI. Overall, USP43 regulates the protein stability of TAK1 through its deubiquitinating enzyme activity, activates the downstream JNK/p38 signaling pathway, and is an important regulatory molecule in cerebral ischemia–reperfusion injury (Fig. 9). Targeted inhibition of USP43 could be a potential therapeutic modality.

Fig. 9.

Proposed mechanism of USP43 mediated cerebral I/R injury. In response to cerebral ischemia–reperfusion injury, USP43 expression increases in neurons. USP43 directly interacts with TAK1, removing the K48 ubiquitin chain and increasing TAK1 phosphorylation level. This activates the downstream JNK and p38 signaling pathways, promoting neuronal inflammation and apoptosis. Created in BioRender. dong, y. (2025) https://BioRender.com/3fr7dwg

Abbreviations

Bcl2

B-cell lymphoma-2

Bax

Bcl2-associated X protein

CCK-8

Cell counting kit-8

Ccl2

C–C motif chemokine ligand 2

Cxcl2

Chemokine (C-X-C motif) ligand 2

DMSO

Dimethyl sulfoxide

DUB

Deubiquitinating enzyme

I–R

Ischemia–reperfusion

ERK

Extracellular signal-regulated kinase

Gfp

Green fluorescent protein

Il1b

Interleukin1beta

Il6

Interleukin 6

Ikkβ

Inhibitor of nuclear factor kappa-B kinase beta

IκBα

Inhibitor of nuclear factor kappa-Bα

JNK

C-Jun N-terminal kinase

LDH

Lactate dehydrogenase

MAPK

Mitogen-activated protein kinase

NF-κB

Nuclear factor kappa-B

OGD/R

Oxygen and glucose deprivation/reoxygenation

p65

Nuclear factor kappa-B RelA

p38

P38 mitogen-activated protein kinase

TAK1

Transforming growth factor-β-activated kinase 1

t/MCAO

Transient middle cerebral artery occlusion

Tnf

Tumor necrosis factor

TUNEL

Terminal deoxynucleotidyl transferase dUTP nick end labeling

USP43

Ubiquitin specific peptidase 43

Author contributions

Peng Zhang, Wenyuan Zhao and Jincao Chen designed the experiments. Yuhan Dong, Sha Hu, Tingbao Zhang and Rui Zhang performed the experiments. Yuhan Dong, Yu Feng and Weiyu Sun analyzed the data. Yuhan Dong and Sha Hu wrote the original draft. Peng Zhang, Jincao Chen, and Wenyuan Zhao edited and reviewed the final manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by Key Research and Development Program of Hubei Province (No. 2022BCA003).

Data availability

The data that support the findings of this study are available on request from the corresponding author.

Declarations

Ethics approval and consent to participate

The animal experiment protocol was approved by the Experimental Animal Welfare Ethics Committee, Zhongnan Hospital of Wuhan University. (APPROVAL#ZN2024275).

Consent for publication

Not applicable.

Competing interests

We certify that we have no affiliations with any organization or entity with any financial interest or non-financial interest.