Seven-day-old newborn Sprague-Dawley (SD) rats (randomly male and female) were obtained from Hunan SJA Laboratory Animal Co., Ltd. The newborn rats were allowed to nurse freely by female rats in a 12-hour alternating light/dark cycle, with a temperature of 24 $\pm$ 2 °C and humidity kept at 55 $\pm$ 5%. The neonatal rats were randomly assigned to Sham and Model groups, each consisting of 18 rats. Rice-Vannucci method was employed to induce HIBD model [15]. Isoflurane induced anesthesia was usually administered by inhalation. The rats were placed in a closed anesthesia box, the switch of the connected anesthesia machine was turned on, and the anesthetic effect could be achieved at an interval of 4-6 min, the isoflurane dose was 4%, and the oxygen flow rate was 1 liter per min. After the rats were induced into anesthesia, the rats were taken out and the limbs were quickly fixed on the operating plate, and the rats’ mouth and nose were covered with a breathing catheter with anesthetic gas to maintain anesthesia. The isoflurane dose was 1.5% and the oxygen flow rate was maintained at 1 liter per min. A 0.5 cm midline incision was made to expose left carotid artery after sterilising neck skin with 75% alcohol. The vessels were ligated at both ends and disconnected in middle. The incision was then closed using sutures, and the surgical site was disinfected. Following procedure, neonatal rats were exposed to a nitrogen-oxygen mixture (8% O₂ + 92% N₂) at 36 $\pm$ 1 °C for 2.5 h [16]. Blood vessels were separated without additional actions, ligation or hypoxia treatment in Sham group. The mortality rate of rats was 3–5%, and this occurred mostly while inhaling hypoxically. The induction concentration of isoflurane during surgery was 2–2.5%, and the maintenance concentration was 1–1.5%. Animals were placed on an electric blanket throughout the operation. During the hypoxic period, animals were placed in a closed, temperature-controlled box at 37 °C [16]. Rectal temperature was measured using an electronic thermometer. Hypoxia began 2 h after surgery. Assessment of neurological function was carried out 48 h after brain damage. Brain tissue was then removed for measurement of the cerebral infarct volume and brain water content (BWC). The rat was euthanized using spinal dislocation: one hand pressed the head and neck of the rat with a hard rod, the other hand grabbed the tail or hind leg of the rat, and quickly and forcefully pulled back and up to dislocate the cervical vertebra.

Negative geotropism was tested 48 h after HIBD by two blinded investigators in an unbiased setting, with 18 rats used in each group. The test involved placing the neonatal rats head-down on a tilt board set at a 45° angle, and then recording the time taken by the animal to turn and assume a head-up position. The test was limited to a maximum duration of 60 s [17].

TTC staining was performed 48 h after HIBD in order to evaluate cerebral infarct volume, with 6 rats being allocated to each group. A coronal section (2 mm thickness) of the brain was immersed in 1% TTC solution (AWI0490, Abiowell, Changsha, Hunan, China), stained for 15–30 min at 37 °C, and then photographed for evaluation. After staining with TTC, the infarct volume of brain tissue was measured using a computerized pathological image analysis system. Infarct volume = (volume of the normal hemisphere – non-infarct volume of the infarct hemisphere)/volume of the hemisphere $\times$ 100%.

The BWC was examined 48 h after HIBD as previously described [18, 19]. Six rats were used in each group, with the BWC (%) calculated using formula: (wet weight – dry weight)/wet weight $\times$ 100%. The wet weight of brain tissue was measured after removal from the animal. It was placed in a constant-temperature oven for approximately 48 h at 60 °C in order to obtain the dry weight.

Neonatal SD rats (within 24 h after birth) were soaked in 75% alcohol, disinfected for 3 min, and then transferred to a sterile operating table. The fetal rat cortical tissue was separated aseptically on a super-clean workbench, cut up with scissors, and placed into a 15 mL centrifuge tube. Pancreatic enzyme digestion solution (AWC0232, Abiowell, Changsha, Hunan, China) was then added (5 mL, 0.25%), mixed well, and allowed to digest the tissue for 15 min at 37 °C. To stop the digestion, Dulbecco’s modified Eagle medium (DMEM, D5796, Sigma, Saint Louis, MO, USA) with 10% fetal bovine serum (FBS, 10099141, Gibco, Grand Island, NY, USA) was added. Mixture was then centrifuged at 300 g for 5 min. Supernatant was discarded. DMEM with 10% FBS was added, and a cell suspension made by gently blowing 20 times with a Pasteur pipette. The filtrate was collected using a 70 µm filter, centrifuged at 300 g for 5 min. Pellet was re-suspended in DMEM with 10% FBS and used to seed 6-well plates that had been pre-coated with poly-lysine. On the second day, the medium was replaced with complete neuron culture medium (PriMed-iCell-005, icellbioscience, Shanghai, China) with 2% B27, with replacement every 2–3 days.

For identification of PCNs, cell crawling tablets were cleaned 2–3 times with phosphate buffered saline (PBS), fixed with 4% paraformaldehyde (AWI0062, Abiowell, Changsha, Hunan, China), and permeated at 37 °C for 30 min with 0.3% Triton X-100 (AWH0299, Abiowell, Changsha, Hunan, China). Cell crawling tablets were blocked with a 5% solution of bovine serum albumin (BSA, V900933, Merck, Beijing, China) for 1 h at 37 °C, and incubated with an appropriately diluted primary antibody for neuron-specific enolase (NSE, AWA10959, Abiowell) overnight at 4 °C. This was followed by incubation with 50–100 µL of anti-rabbit-IgG-labeled fluorescent antibody (AWS0005a, Abiowell) for 90 min. Slides were incubated with 4^′,6-diamidino-2-phenylindole (DAPI) working solution (AWC0292a, Abiowell, Changsha, Hunan, China) for 10 min at 37 °C. They were then sealed with buffered glycerin, preserved away from light, and observed with an epi-fluorescence microscope (BA210T, Motic, Fujian, China).

Pheochromocytoma (PC12) cells (undifferentiated, AW-CCR443, Abiowell) were cultured in a humidified atmosphere at 37 °C with 5% CO₂ in Roswell Park Memorial Institute (RPMI) 1640 medium (M3817, sigma, Saint Louis, MO, USA) supplemented with 10% horse serum, 5% FBS, and 100 U/mL penicillin-streptomycin. The PC12 cells were authenticated Quantitative polymerase chain reaction (QPCR), and all experiments were conducted using mycoplasma-free cells. For differentiation induction, PC12 cells were treated with 50 ng/mL of nerve growth factor-beta and 1% FBS for 4 days. A cell OGD/reoxygenation (OGD/R) model was first established. PC12 cells or PCNs were cultured at 37 °C for 24 h in serum- and glucose-free medium in the atmosphere of 93% N₂, 5% CO₂ and 2% O₂. After OGD/R, PC12 cells or PCNs were transferred to medium containing glucose, brought back to normal oxygen levels, and incubated at 37 °C with 5% CO₂ for 24 h to reoxygenate [20]. In addition, cells were treated with OGD/R following knockdown or overexpression of IGF2BP3 or PDCD4. Short hairpin negative control (sh-NC), overexpression negative control (oe-NC), sh-IGF2BP3, oe-IGF2BP3, sh-PDCD4, and oe-PDCD4 were synthesized by HonorGene (Changsha, Hunan, China). Cells were transfected with Lipofectamine 2000 (11668019, invitrogen, Carlsbad, CA, USA) for 48 h.

Protein extraction was performed from cells and damaged brain tissue (6 rats in each group) with Radio Immunoprecipitation Assay Lysis buffer (P0013B, Beyotime, Shanghai, China) and the protein content measured. Proteins underwent separation on Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) with the use of MB2479 from meilunbio (Dalian, Liaoning, China), followed by transfer onto polyvinylidene fluoride membranes. Incubation with primary antibody was then conducted overnight at 4 °C using following antibodies: PDCD4 (ab80590, 1:1000, Abcam, Cambridge, UK), IGF2BP3 (14642-1-AP, 1:5000, Proteintech, Chicago, IL, USA), Bcl-2 (60178-1-Ig, 1:5000, Proteintech), Bax (ab32503, 1:5000, Abcam), caspase3 (19677-1-AP, 1:1000, Proteintech), and $\beta$-actin (66009-1-Ig, 1:5000, Proteintech). This was followed by incubation with Goat Anti-Mouse IgG(H+L) (SA00001-1, 1:6000, Proteintech) or Goat Anti-Rabbit IgG(H+L) (SA00001-2, 1:6000, Proteintech) for 90 min at 37 °C. Proteins were subsequently detected using a luminescent solution (K-12045-D50, advansta, Beijing, China) and an imaging system (Chemiscope6100, CLINX, Shanghai, China). $\beta$-actin was employed as internal reference, and Image J software (National Institutes of Health, Bethesda, MD, USA) was utilized to analyze the intensity of target bands.

RNA was isolated through Trizol (15596018, Invitrogen, Carlsbad, CA, USA) and converted into cDNA through reverse transcription with the kit (CW2569, CWBIO, Bejing, China). The cDNA was then amplified with Ultra SYBR Mixture (CW2601, CWBIO) on ABI 7900 system (ABI, Carlsbad, CA, USA). Gene expression levels were measured through 2^{-Δ⁢Δ⁢Ct} method with $\beta$-actin as internal reference. Primer sequences were: IGF2BP3-F: GCCAAGGCTGAGGAGGAAAT, IGF2BP3-R: GAGGAGTCATGGCTGAAGGG; $\beta$-actin-F: ACATCCGTAAAGACCTCTATGCC, $\beta$-actin-R: TACTCCTGCTTGCTGATCCAC.

Cells were trypsinized, counted, and subsequently plated into 96-well plates (1 $\times$ 10⁴ cells/well, 100 µL per well) for further incubation in a 5% CO₂ atmosphere at 37 °C. Ten µL of CCK-8 (NU679, DOJINDO, Kumamoto, Japan) was then added and and incubated with 5% CO₂ for 4 h at 37 °C. Absorbance at 450 nm was evaluated with microplate reader (MB-530, Heales, Shenzhen, Guangdong, China).

Cells were detached using ethylenediaminetetraacetic acid (EDTA)-free trypsin, followed by centrifugation for 5 min at 1000 rpm to collect cell pellets. The pellets were resuspended, washed with PBS, and centrifuged again for 5 min at 1000 rpm. The cells were subsequently suspended in 500 µL of binding buffer and incubated with 5 µL of Annexin V-APC (KGA108, KeyGen, Nanjing, Jiangsu, China). Subsequently, 5 µL of propidium iodide (537060, sigma, Saint Louis, MO, USA) was added to mixture, mixed, and then incubated in dark for 10 min. Flow cytometry analysis was conducted with a Beckman instrument (A00-1-1102, Beckman Coulter, Brea, CA, USA) within 1 h.

Interleukin 6 (IL-6), Interleukin 1 $\beta$ (IL-1$\beta$), and tumor necrosis factor-$\alpha$ (TNF-$\alpha$) levels were evaluated using IL-6 (CB-E04640R, CUSABIO, Wuhan, Hubei, China), IL-1$\beta$ (RLB00, R&D Systems, Minneapolis, MN, USA), and TNF-$\alpha$ (CB-E11987R, CUSABIO) kits, respectively, and the Bio Tek microplate reader according to instructions.

The interaction between IGF2BP3 protein and PDCD4 mRNA was studied with EZMagna RIP kit (17-701, Merck Millipore, Darmstadt, Germany). PC12 cells were dissolved in RIP solution and incubated with beads coupled to Ago2 or anti-rabbit immunoglobulin G (IgG, negative control). Precipitated RNA was assessed using qRT-PCR [21].

To confirm the interaction between IGF2BP3 protein and PDCD4 mRNA, RNA pull-down detection was performed using previously described methods [22, 23]. Specifically, the cell extract was mixed with biotinylated RNA and then added to rinsed streptavidin agarose beads and incubated. The microbeads were subsequently washed and boiled, and the proteins analyzed by SDS-PAGE and Western blot.

Statistical analysis of data was performed using Graphpad Prism 8.0 software (GraphPad Software, Inc., San Diego, CA, USA), and results were presented as the mean $\pm$ standard deviation. Tests were conducted to assess normality and homogeneity of variance. Group comparisons were done using non-paired t-tests. For multiple group comparisons, one-way analysis of variance (ANOVA) was used, followed by Tukey’s post hoc test. A significance level of p 0.05 was considered statistically significant.

We first constructed a neonatal rat model of HIBD. At 48 h after HIBD, the geotaxis reflex time, infarct volume and BWC of Model group were all higher compared to Sham group (Fig. 1A–C). We next investigated the possible mechanism of action of PDCD4 in HIBD. catRAPID online analysis predicted that PDCD4 interacts with IGF2BP3. PDCD4 and IGF2BP3 levels were evaluated both in vivo and in vitro. In neonatal rats, PDCD4 and IGF2BP3 levels were higher in Model group than Sham group (Fig. 1D). In vitro, PDCD4 and IGF2BP3 levels were also elevated in OGD/R group than Control group (Fig. 1E). These results demonstrated PDCD4 and of the m6A reader protein IGF2BP3 levels were up-regulated in HIBD neonatal rats.

Fig. 1.

Upregulation of PDCD4 and m6A reader protein IGF2BP3 in HIBD neonatal rats. (A) Assessment of short-term neurological function (geotaxis reflex time). (B) TTC staining for cerebral infarct volume. (C) Assessment of BWC. (D) Western blot analysis of PDCD4 and IGF2BP3 levels in Sham Sand model groups. (E) Western blot detection of PDCD4 and IGF2BP3 levels in Control and OGD/R groups. *p 0.05 vs. Pre-surgery or Sham or Control. PDCD4, programmed cell death 4; IGF2BP3, insulin-like growth factor 2 mRNA binding protein 3; HIBD, hypoxic-ischemic brain damage; TTC, 2,3,5-Triphenyl tetrazolium chloride; OGD/R, oxygen glucose deprivation/reoxygenation; BWC, brain water content.

Knockdown of IGF2BP3 was performed to study its function. First, the efficiency of knockdown was verified (Fig. 2A). IGF2BP3 mRNA and protein levels in OGD/R+sh-IGF2BP3 group were lower than OGD/R+sh-NC group, indicating successful knockdown of IGF2BP3. Moreover, cell viability and Bcl-2 levels were repressed in OGD/R group than control group, whereas apoptosis was increased. Bax, cleaved-caspase3, IL-6, IL-1$\beta$, and TNF-$\alpha$ levels were also elevated in OGD/R group. Knockdown of IGF2BP3 reversed these effects (Fig. 2B–E). These results showed that knockdown of IGF2BP3 reduced OGD/R-induced injury in PC12 cells.

Fig. 2.

Knockdown of IGF2BP3 reduced OGD/R-induced injury in pheochromocytoma PC12 cells. (A) IGF2BP3 mRNA and protein levels. (B) Cell viability was determined by means of the CCK-8 assay. (C) Flow cytometry detection of apoptosis. (D) Cellular Bax, Bcl-2, and cleaved-caspase3 protein levels. (E) Evaluation of IL-6, IL-1$\beta$, and TNF-$\alpha$ levels. *p 0.05 vs. Control, ^#p 0.05 vs. OGD/R+sh-NC. CCK-8, Cell Counting Kit-8; sh-NC, short hairpin negative control; IL-6, Interleukin-6; IL-1$\beta$, Interleukin-1$\beta$; TNF-$\alpha$, Tumor necrosis factor-$\alpha$.

PDCD4 protein levels were higher OGD/R group than control group, but decreased following knockdown of IGF2BP3 (Fig. 3A). Furthermore, RIP and RNA pull-down revealed an interaction between IGF2BP3 protein and PDCD4 mRNA (Fig. 3B,C). These results indicated that IGF2BP3 interacted with PDCD4 mRNA to regulate PDCD4 levels.

Fig. 3.

IGF2BP3 interacted with PDCD4 mRNA to regulate the PDCD4 level. (A) PDCD4 protein levels. (B) RIP identification of interaction between IGF2BP3 protein and PDCD4 mRNA. (C) RNA pull-down was utilized to confirm the interaction between IGF2BP3 protein and PDCD4 mRNA. *p 0.05 vs. Control or IgG, ^#p 0.05 vs. OGD/R+sh-NC. RIP, RNA immunoprecipitation; IgG, immunoglobulin G; IP, immunoprecipitation.

We conducted further mechanistic investigations by knocking down IGF2BP3 and overexpressing PDCD4. Compared with the sh-NC group, PDCD4 levels decreased following knockdown of IGF2BP3. In contrast, PDCD4 levels increased following the overexpression of PDCD4 (Fig. 4A). In comparison to sh-NC group, sh-IGF2BP3 group showed increased cell viability, reduced apoptosis, lower levels of Bax and cleaved-caspase3, higher levels of Bcl-2, as well as decreased IL-6, IL-1$\beta$, and TNF-$\alpha$ contents. Overexpression of PDCD4 reversed these effects (Fig. 4B–E). The above results indicated that knockdown of IGF2BP3 reduced OGD/R-induced cell damage through PDCD4.

Fig. 4.

Knockdown of IGF2BP3 reduced OGD/R-induced cell damage via PDCD4. (A) PDCD4 protein levels. (B) Cell viability assessment through CCK-8 assay. (C) Flow cytometry detection of apoptosis. (D) Cellular Bax, Bcl-2, and cleaved-caspase3 protein levels. (E) Assessment of IL-6, IL-1$\beta$, and TNF-$\alpha$ contents. *p 0.05 vs. sh-NC, ^#p 0.05 vs. sh-IGF2BP3+oe-NC. CCK-8, Cell Counting Kit-8; OD, Optical density; IL-6, Interleukin-6; IL-1$\beta$, Interleukin-1$\beta$; TNF-$\alpha$, Tumor necrosis factor-$\alpha$; sh-NC, short hairpin negative control; oe-NC, overexpression negative control.

Next, we investigated the roles of IGF2BP3 and PDCD4 in PCNs. PCNs were first isolated and identified, as shown in Fig. 5A. Following the knockdown of IGF2BP3 in PCNs, PDCD4 levels were lower compared to sh-NC group. However, PDCD4 levels increased in PCNs following the overexpression of PDCD4. PDCD4 levels in PCNs increased following overexpression of IGF2BP3 compared to oe-NC group. PDCD4 levels decreased in PCNs following knockdown of PDCD4 (Fig. 5B). Furthermore, compared to sh-NC group, cell apoptosis was decreased, Bax and cleaved-caspase3 levels were supressed, Bcl-2 levels were elevated, and IL-6, IL-1$\beta$, and TNF-$\alpha$ contents were supressed in sh-IGF2BP3 group of PCNs. Overexpression of PDCD4 reversed these effects. Conversely, compared to oe-NC group, cell apoptosis was increased, Bax and cleaved-caspase3 levels were elevated, Bcl-2 levels were supressed, and IL-6, IL-1$\beta$, and TNF-$\alpha$ contents were elevated in oe-IGF2BP3 group of PCNs. Knockdown of PDCD4 reversed these effects (Fig. 5C–E). These results suggested IGF2BP3/PDCD4 axis reduced OGD/R-induced cell damage in PCNs.

Fig. 5.

The IGF2BP3/PDCD4 axis reduced OGD/R-induced cell damage in PCNs. (A) Identification of PCNs. Scale bar: 50 µm. (B) PDCD4 protein levels. (C) Flow cytometry for the analysis of apoptosis. (D) Cellular Bax, Bcl-2, and cleaved-caspase3 protein levels. (E) IL-6, IL-1$\beta$, and TNF-$\alpha$ contents. *p 0.05 vs. sh-NC or oe-NC, ^#p 0.05 vs. sh-IGF2BP3+oe-NC or oe-IGF2BP3+sh-NC. PCNs, primary cortical neurons; DAPI, 4^′,6-diamidino-2-phenylindole.