The suppliers for all materials are provided in Table I.

A total of three human pituitary tumor specimens were collected (November 2024 to June 2025) during neurosurgery at the First Affiliated Hospital of Shihezi University, Shihezi, China. The present study was approved by the Scientific and Technological Ethics Committee of the First Affiliated Hospital of Shihezi University (approval no. KJ2024-476-01). Patients with pituitary tumors confirmed by preoperative computed tomography and postoperative pathology were included Exclusion criteria were as follows: Patients with incomplete clinical data; patients who undergone radiotherapy or chemotherapy prior to surgery; samples with poor tissue preservation or insufficient RNA/protein quality. All patients signed a written informed consent form. Patient information is summarized in Table II.

Rat pituitary tumor cell lines (GH3 and MMQ) were obtained from Wuhan Procell Biotechnology Co., Ltd. Both GH3 and MMQ cells were cultured in a GH3-specific complete medium (82.5% Ham's F-12K medium, 15% horse serum, 2.5% FBS and 1% penicillin-streptomycin) in a sterile, humidified incubator at 37°C with 95% air and 5% CO₂. GH3 cells grew in a loosely adherent manner, while MMQ cells grew in suspension. When confluency reached 70-80%, the cells were passaged at a ratio of 1:3 to maintain exponential growth.

AS-IV was purchased from MedChemExpress (purity, 99.93%) and dissolved in DMSO according to the manufacturer's instructions. GH3 cells were divided into the AS-IV treatment group (n=6) and DMSO-treated negative control group (NC, n=3). Transcriptomics sequencing was performed at Hangzhou Lianchuan Biotechnology Co., Ltd. RNA was isolated using TRIzol (Invitrogen; Thermo Fisher Scientific, Inc.). The total RNA was quantified and quality-controlled using a NanoDrop ND-1000 (NanoDrop). rRNA was removed using the Epicentre Ribo-Zero Gold Kit (Illumina, Inc.), and cDNA was synthesized from the fragmented RNA using reverse transcriptase (Invitrogen SuperScript™ II Reverse Transcriptase, cat. no. 1896649). Then, E. coli DNA polymerase I (NEB, cat. no. m0209) and RNase H (NEB, cat. no. m0297) were used for double-strand synthesis to convert the composite double-stranded RNA-DNA hybrid into double-stranded DNA. The double-stranded DNA was digested with UDG (NEB, cat. no. m0280), followed by PCR (pre-denaturation at 95°C for 3 min; 8 cycles of denaturation at 98°C for 15 sec each; annealing at 60°C for 15 sec; extension at 72°C for 30 sec, followed by a final extension at 72°C for 5 min) to generate a library consisting of fragments ranging from 300±50 bp. Finally, the library was sequenced using Illumina Novaseq™ 6000 (LC Bio Technology Co., Ltd.) in a paired-end (PE150) sequencing mode. The library concentration was measured using a Qubit fluorometer (Thermo Fisher Scientific, Inc.) and a High Sensitivity DNA Chip on a Bioanalyzer 2100 (Agilent Technologies), with the final loading concentration adjusted to 2 nM for sequencing.

Low-quality sequences in the raw sequencing data were filtered out via Cutadapt (v1.9; cutadapt.readthedocs.io/). The processed data were used for transcriptome assembly and quantitative analysis. Differentially expressed genes (DEGs) were screened with the limma package (v3.5x; bioinf.wehi.edu.au/limma/) in R (v4.2.1; r-project.org/) with the threshold defined as fold-change (FC) ≥2 or ≤0.5 and q-value <0.05 (|log₂FC|≥1 and q<0.05). Gene Ontology (GO) enrichment analysis (geneontology.org/), Gene Set Enrichment Analysis (GSEA; gsea-msigdb.org/), Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment (kegg.jp/) and protein-protein interaction (PPI) analysis (string-db.org/) were performed using WebGestalt (webgestalt.org/). Partial bioinformatics data were obtained from the Gene Expression Omnibus (GEO) database (accession no. GSE136781; ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE136781) and the STRING database (cn.string-db.org; TUBB4B and STMN1).

The lentiviral vectors used for OE included LV5(EF-1α/GFP&Puro)-TUBB4B (OE-TUBB4B) and LV5 (EF-1α/GFP&Puro)-NC (OE-shNC; both 1×10⁹ TU/ml). The lentiviral vectors used for knockdown LV3-GFP&Puro-TUBB4B-141 (KD1-TUBB4B), LV3-GFP&Puro-TUBB4B-199 (KD2-TUBB4B) and LV3-GFP&Puro-NC (KD-shNC; all 1×10⁸ TU/ml) and polybrene were purchased from Suzhou GenePharma Co., Ltd. The sequences of lentiviral vectors are listed in Table III. The lentiviral packaging system used in this study is a third-generation system. Lentiviral particles were prepared in 293T cells (purchased from GenePharma). For transfection, 10 μg of lentiviral transfer plasmids (expressing TUBB4B or shRNA) were mixed with the packaging plasmids pGag/Pol, pRev, and pVSV-G in a 4:2:2:1 ratio and transfected using Lipofectamine 3000 (Thermo Fisher Scientific, Inc.) at 37°C for 48 h. After transfection, the supernatant containing lentiviral particles was collected, filtered through a 0.45 μm filter, and concentrated by ultracentrifugation. The viral titer was determined by qPCR and expressed in TU/ml.

GH3/MMQ cells were seeded into 6-well plates at a density of 2×10⁵ cells/well. When cell confluency reached 70%, lentiviral vectors and polybrene were added to 6-well plates containing complete medium (MOI for the GH3 cell line was ~50, and for the MMQ cell line, ~100) and allow transfection to proceed for 24-48 h at 37°C and 5% CO₂. After 72 h, the cells were selected with 1 μg/ml puromycin for 48 h at 37°C and 5% CO₂ to establish stably transfected cell lines, and 1 μg/ml puromycin was used to maintain the continued culture of stably transfected cells. The transfection efficiency of the stably transfected cell lines was verified by reverse transcription-quantitative (RT-q)PCR and western blotting. Cells stably transfected for up to 2 weeks were used for subsequent experiments.

Total RNA was extracted from the GH3/MMQ cells using Total RNA kit I, and the RNA concentration was quantified with a NanoDrop One device. cDNA was synthesized via one-step RT using the RevertAid First Strand cDNA Synthesis kit following the manufacturer's instructions. Target cDNA fragments were amplified and detected using UltraSYBR Mixture PCR reagents on a Gentier 96R RT-qPCR system (Xi'an Tianlong Technology Co., Ltd.) according to the manufacturer's instructions. The thermocycling conditions were as follows: initial denaturation at 95°C for 10 min, followed by 40 cycles of 95°C for 15 sec and 60°C for 60 sec). GAPDH was adopted as the internal reference to normalize the relative expression of target mRNAs, and the 2^−∆∆Cq method (21) was employed for quantitative analysis of relative nucleic acid expression. All primer sequences are listed in Table IV.

The animal experiments were approved by the Bioethics Committee of Shihezi University, Shihezi, China (approval no. A2024-353). Thirty female BALB/c nude mice (age, 3-4 weeks old; weight, 15-18 g at the start of the study) were purchased from Helilai Co., Ltd. and housed in a specific pathogen-free environment (temperature, 24±2°C; humidity, 50-60%; 12 h light/dark cycle) with ad libitum access to sterilized food and water. At the start of the experiment, each nude mouse was injected with 2×10⁶ GH3 cells suspended in 100 μl PBS into the left axilla. Mice in the Vector + AS-IV, OE-TUBB4B + AS-IV and KD2-TUBB4B+AS-IV) with n=5 per subgroup, were intraperitoneally injected with AS-IV at a half-maximal inhibitory concentration (IC₅₀) dose of 10-20 mg/kg daily, whereas the mice in the control group, consisting of three subgroups (Vector, OE-TUBB4B, and KD2-TUBB4B) with n=5 per subgroup (total n=15),were intraperitoneally injected with the same volume of normal saline daily. Humane endpoints, including tumor diameter exceeding 15 mm, significant weight loss, and visible signs of tumor ulceration or infection. The body weight and the length/width of the subcutaneous tumors were measured every 3 days. The mice were euthanized by cervical dislocation following 21 days of continuous feeding; death was confirmed by the absence of a heartbeat, cessation of breathing and fixed, dilated pupils. Tumor samples were weighed and stored for further analysis.

GH3/MMQ cells were seeded into 96-well plates at a density of 8×10³ cells/well and cultured in a 37°C for 24 h. The cells were cultured in 100 μl GH3 complete medium containing AS-IV at (0 to 140 μM in a 37°C, 5% CO₂ incubator for 0-72 h. At 0, 24, 48, and 72 h, 10 μl Cell Counting Kit 8 (CCK-8) reagent and 90 μl F12K medium were added to each well, followed by incubation for 2 h. The optical density at 450 nm was measured using a microplate reader.

To investigate the mechanism of TUBB4B, GH3 cells were transduced with lentivirus to establish OE-TUBB4B and Vector stable cell lines. Following the manufacturer's instructions, the cells were treated with 60 nM U0126 (a MEK/ERK pathway inhibitor) at 37°C with 5% CO₂ for 48 h, after which cell viability was assessed using the CCK-8 assay.

GH3 cells were seeded into 24-well plates at a density of 1×10⁵ cells/well. After incubating for 24 h at 37°C and 5% CO₂, the GH3 complete medium was replaced with GH3 complete medium containing AS-IV ranging from 0 to 160 μM, and the cells were cultured at a 37°C for 48 h. The cells were washed three times with PBS and 500 μl 4% paraformaldehyde was added for 15 min at room temperature. The cells were permeabilized with 0.4% Triton X-100 for 10 min at room temperature. A TUNEL apoptosis detection kit was used according to the manufacturer's instructions. The nuclei were stained at room temperature in the dark for 20 min with DAPI at a final concentration of 1 μg/ml. Finally, the slides were mounted with 50% glycerol in PBS, cells were observed using a fluorescence microscope and the number of TUNEL-positive cells was counted using ImageJ software (v1.54p; National Institutes of Health).

GH3 cells were seeded into 24-well plates at a density of 1×10⁵ cells/well. After incubating for 24 h at 37°C, the medium was replaced with a GH3 complete medium containing 112.3 μM AS-IV, and the cells were cultured at 37°C for 48 h. An EdU Imaging kit was used according to the manufacturer's instructions to stain nuclei in the S phase. Nuclei were stained at room temperature in the dark for 20 min with DAPI at a final concentration of 1 μg/ml. The results were observed using a fluorescence microscope, and the number of EdU-positive cells was counted using ImageJ software.

GH3 cells were seeded into 6-well plates at a density of 500 cells/well. The medium was changed every 3 days, and the cells were cultured at 37°C for 12-15 days. At room temperature, a total of 1 ml 4% paraformaldehyde was added to each well to fix the cells for 15 min, followed by staining with 0.1% crystal violet for 15 min. Excess crystal violet was removed by washing with water. Colonies were defined as clusters containing >50 cells. Images were captured using a light microscope, and the number of colonies was counted using ImageJ software.

The cells from the Vector, Vector + AS-IV, OE-TUBB4B+AS-IV, and KD2-TUBB4B+AS-IV groups were collected, resuspended in precooled PBS at a density of 1×10⁶ cells/ml and centrifuged at 400 × g and 4°C for 5 min. The supernatant was discarded, and the cells were resuspended in 200 μl PBS. A total of 5 μl Annexin V-PE and 5 μl 7-AAD apoptosis detection reagents were added and incubated at 4°C in the dark for 30 min. Following addition of 300 μl PBS, detection was performed using an ACEA NovoCyte (Agilent Technologies) flow cytometer. The total number of cells undergoing early and late apoptosis was calculated, using NovoExpress software (v1.5.6; Agilent Technologies).

GH3/MMQ cells and tumors from the nude mice were collected. Protein samples were extracted using RIPA lysis buffer, and the protein concentration was quantified with an enhanced Bicinchoninic Acid Protein Assay kit. 50 μg proteins were loaded into each lane; after separation by 10% SDS-PAGE and transferred to a 0.45 μm PVDF membrane. The PVDF membrane was blocked with 5% (w/v) non-fat dry milk in Tris-buffered saline containing 0.1% Tween-20 at room temperature for 1 h, incubated with the primary antibody at 4°C overnight and incubated with the corresponding horseradish peroxidase-conjugated secondary antibody at room temperature for 1 h. Western blots were visualized using ECL reagent, and analyzed using ImageJ software. Antibodies were as follows: TUBB4B 1:1,000; PCNA 1:2,000; Bcl-2 1:1,000; Bax 1:1,000; pro-caspase-3 1:1,000; cleaved-caspase-3 1:1,000; T-PARP 1:1,000; cleaved-PARP 1:1,000; STMN1 1:500; p-STMN1 1:1,000; cPLA2 1:1,000; p-cPLA2 1:500; CCND1 1:1,000; ERK 1:1,000; p-ERK 1:1,000; JNK (1:1,000); p-JNK 1:1,000; GAPDH 1:20,000; Goat anti-rabbit 1:20,000; Goat anti-mouse 1:20,000.)

Human pituitary and mouse tumor specimens were fixed with a 4% paraformaldehyde solution at room temperature for 24 h, followed by paraffin embedding and serial sectioning (thickness, 4 μm). For H&E staining, the sections were rehydrated and stained with hematoxylin solution (0.5%) for 3-5 min, followed by eosin Y solution (0.5%) for 15-20 sec at room temperature. For IHC staining, the sections were rehydrated and antigen retrieval was performed in high-temperature citric acid solution at 160°C for 10 min and washed with PBS. Sections were blocked with 3% H₂O₂ at room temperature for 10 min, followed by blocking with PBS containing 5% BSA at room temperature for 30 min. Primary antibodies against TUBB4B (1:100), PCNA (1:100), Bcl-2 (1:200), Bax (1:100) and cleaved-caspase-3 (1:200) were added at 4°C in the dark overnight. The sections were incubated with a universal secondary antibody (HRP-conjugated goat anti-mouse/rabbit IgG, Servicebio; catalog No. TBAG0036) at a dilution of 1:200 at room temperature for 1 h. DAB chromogen was added for 20 min. For nuclear counterstaining, the sections were stained with hematoxylin at room temperature for 1-2 min. The sections were observed and photographed using a light microscope (Olympus BX53).

GH3 cells stably transfected with lentiviruses were seeded into confocal dishes at a density of 1×10⁵ cells/dish. Subsequently, cells in the Vector, OE-TUBB4B+AS-IV, and KD2-TUBB4B+AS-IV groups were treated with AS-IV at a concentration of 112.3 μM at 37°C for 48 h, fixed with 4% paraformaldehyde at room temperature for 15 min, and permeabilized with 0.4% Triton X-100 at room temperature for 10 min. The cells were stained with phalloidin at room temperature in the dark for 20 min, placed in blocking solution for 30 min, incubated with TUBB4B primary antibody (1:100) at 4°C overnight, and then incubated with fluorescent secondary antibody [goat anti-rabbit IgG (H+L), conjugated with Alexa Fluor-488; 1:200] at room temperature for 1 h. The nuclei were stained with Hoechst 33342 (10 μg/ml) for 20 min at room temperature. Finally, images were captured using a confocal microscope.

AutoDockTools software (v1.5.7; http://autodock.scripps.edu/) was used to preprocess the conformations of AS-IV and TUBB4B and appropriate docking sites were selected. AutoDock Vina software (v1.1.2; vina.scripps.edu/) was used to select the optimal molecular docking conformation. PyMOL software (v2.3.0; pymol.org/) and LigPlot+ software (v2.2.8; ebi.ac.uk/thornton-srv/software/LigPlus/) were used for analysis and visualization.

Amber24 software (ambermd.org/) was used for molecular dynamics simulation purposes, with the ff19SB force field and Optimal Point Charge (OPC) water model selected. The AS-IV-TUBB4B complex was placed in a cubic water box. The cutoff distance for electrostatic and van der Waals interactions was set to 1.0 nm, the time step was 2 femtosecond (fs) and the particle-mesh Ewald method (22) used for long-range correction of electrostatic interactions. Energy minimization was performed, followed by 200 ps Microcanonical ensemble (NVE) equilibrium dynamics and 100 ps number of particles, pressure and temperature) equilibrium dynamics. The V-rescale method was used for temperature coupling of the thermal bath system (23) and the Parrinello-Rahman method was used for pressure control (24). Finally, 100 nsec molecular dynamics sampling was performed. Indicators such as root mean square fluctuation (RMSF), root mean square deviation (RMSD), radius of gyration (Rg) and solvent-accessible surface area (SASA) were calculated using Amber modules (AmberTools24; https://ambermd.org), Cpptraj (v5.1.0; github.com/Amber-MD/cpptraj) and Python scripts (v3.10; https://www.python.org).

GH3 cells in the logarithmic growth phase were treated with 112.3 μM AS-IV at 37°C for 24 h in the shNC, OE-TUBB4B and KD2-TUBB4B groups. Following trypsin digestion, the cells were resuspended and aliquoted at 1×10⁶ cells/well into eight EP tubes. The tubes were heated for 10 min across a temperature gradient (35-70°C). NP-40 lysis buffer was added and the sample was centrifuged at 4°C, 20,000 × g for 20 min. The protein supernatant was collected. The TUBB4B protein expression was detected via western blot analysis as aforementioned. Melting curves were plotted by analyzing changes in TUBB4B solubility or abundance at different temperatures.

Statistical analysis was performed using GraphPad Prism 10 (Dotmatics). All experiments were repeated at least three times, and all data are expressed as the mean ± SD. One- or two-way ANOVA followed by Dunnett's t test was used for comparisons. P<0.05 was considered to indicate a statistically significant difference.

CCK-8 assay demonstrated that AS-IV significantly inhibited the activity of GH3 and MMQ cells in a concentration- and time-dependent manner (Fig. 1D-K). The chemical structure of AS-IV is shown in Fig. 1A. The IC₅₀ values detected after 48 h AS-IV treatment in GH3 and MMQ cells were 112.3 and 247.7 μM, respectively (Fig. 1B and C). To ensure consistency in subsequent experiments, 48 h was selected as the treatment duration and IC₅₀ values were used as the drug concentration for AS-IV intervention.

TUNEL assay was performed to confirm that AS-IV promotes apoptosis in GH3 cells in a concentration-dependent manner (Fig. 1L). The apoptosis rate of the GH3 cells treated with 160 μM AS-IV was significantly greater than that of the control cells (Fig. 1M). Western blot analysis of apoptosis-associated proteins (Fig. S1A) revealed that the expression of cleaved-PARP (Fig. S1C) significantly increased in AS-IV-treated cells. The expression of the 113 kDa total PARP isoform decreased but that of the 89 kDa form increased, however there was no significant difference in total protein levels (Fig. S1B and C). Western blotting revealed that pro-caspase3 protein levels remained largely unchanged in AS-IV-treated cells, whereas the relative expression of cleaved-caspase3 protein (Fig. S1D-F) was significantly elevated. Transmission electron microscopy revealed increased numbers of vesicles, apoptotic bodies, fragmented mitochondrial tubular networks and loss of mitochondrial cristae fusion in AS-IV-treated cells (Fig. S1G). These findings confirmed that AS-IV induces pituitary tumor cell apoptosis.

To explore the key signals and related pathways underlying the AS-IV-mediated inhibition of GH3 cell proliferation, single-cell sequencing of AS-IV-treated GH3 cells was performed. Compared with the NC group, 931 up- and 829 downregulated DEGs were identified (Fig. 2A and B). GO enrichment analysis in the biological process category (Fig. 2C) revealed that the functions of the DEGs were enriched in 'DNA replication' and 'cell cycle'. A heatmap of the transcriptomic DEGs (Fig. 2D) revealed that the tubulin family (which synthesizes the cytoskeleton) (25), especially TUBB, was significantly downregulated in the AS-IV group. GSEA revealed that genes were significantly downregulated in 'mitotic cell cycle', 'cyclin-dependent protein kinase holoenzyme complex', 'double-stranded RNA binding' and 'structural constituent of cytoskeleton' (Fig. 2E-H). As these enriched gene sets are associated with tubulin-associated functions, it was hypothesized that there was a potential link between cytoskeletal organization and RNA binding, with TUBB4B possibly serving as a bridging molecule. To explore this hypothesis, we selected the gene lists of 'double-stranded RNA binding' and 'structural constituent of cytoskeleton' to construct a Venn diagram (Fig. 2I), which revealed that TUBB4B was present in the intersection. It was hypothesized that TUBB4B may be a key gene mediating AS-IV-induced inhibition of pituitary tumor proliferation and promotion of apoptosis. GEO data demonstrated that TUBB4B expression in pituitary tumor tissues was significantly higher than that in normal pituitary tissue (Fig. 2J), which was consistent with the IHC staining results of pituitary tumor tissue from patients (Fig. 2K). These results confirm that TUBB4B is highly expressed in pituitary tumors.

Stably transfected cell lines with lentivirus-mediated OE-TUBB4B or knockdown were established (Fig. 3A). The transfection efficiency was verified by RT-qPCR and western blotting (Fig. 3B, C and E-G). The results confirmed the successful establishment of stably transfected cell lines. In both cell lines, the knockdown efficiency of KD2-TUBB4B was greater than that of KD1-TUBB4B; therefore, the KD2-TUBB4B stable cell line was selected for subsequent experiments.

Cell viability was assessed with the CCK-8 assay. In both cell lines, viability in the OE-TUBB4B group was greater than that in the shNC group, which was significantly greater than that in the KD2-TUBB4B group (Fig. 3D and H). EdU assays revealed that the number of proliferating GH3 cells (in the S phase of DNA synthesis; red fluorescence) was highest in the OE-TUBB4B group and lowest in the KD2-TUBB4B group (Fig. 3I and K). Colony formation assay (Fig. 3J and L) revealed that the number of colonies in the OE-TUBB4B group was significantly greater than that in the shNC group, which was greater than that in the KD2-TUBB4B group. Flow cytometry confirmed that OE-TUBB4B decreased the proportion of apoptotic cells, whereas KD-TUBB4B increased the percentage of apoptotic cells in both cell lines (Fig. 3M and O). Western blotting was used to verify the expression of proliferation-associated protein (PCNA) and apoptosis-related proteins (Bcl-2, Bax, pro-caspase-3 and cleaved-caspase-3) in both cell lines (Fig. 3N and P-T). OE-TUBB4B cells had increased expression of PCNA and antiapoptotic protein (Bcl-2) and decreased expression of proapoptotic proteins (Bax and cleaved-caspase-3); in KD2-TUBB4B cells, decreased expression of PCNA and Bcl-2 was observed, along with increased expression of Bax and cleaved-caspase-3. Cells in the OE-TUBB4B group exhibited increased expression of PCNA and Bcl-2 and decreased expression of Bax and cleaved caspase-3; in contrast, cells in the KD2-TUBB4B group exhibited decreased expression of PCNA and Bcl-2, and increased expression of Bax and cleaved caspase-3, confirming the regulatory role of TUBB4B. In summary, TUBB4B promotes the proliferation and inhibits the apoptosis of pituitary tumor cells.

To verify that AS-IV inhibits pituitary tumors by targeting TUBB4B, molecular conformations of TUBB4B and AS-IV were generated. The surface electrostatic potential of the TUBB4B protein is shown in Fig. 4A, and molecular docking was performed between AS-IV and TUBB4B (Fig. 4B). The binding energies of multiple TUBB4B protein conformations with AS-IV indicated a high degree of fit in the binding pocket and a binding energy of −8.2 kcal/mol (Fig. 4C). These findings confirmed that AS-IV and TUBB4B had strong binding activity and bind under natural conditions. On the basis of the molecular docking results, a 100 nsec molecular dynamics simulation was performed on the AS-IV-TUBB4B complex to investigate its binding stability and dynamic behavior. RMSF (Fig. 4D) of the AS-IV-TUBB4B complex fluctuated at residues 56, 80, 175, 277-282 and 427, which may be associated with the function of TUBB4B affected by AS-IV; the RMSD (Fig. 4E) fluctuated in the early stage (0-20 nsec) but stabilized at 20-100 nsec, indicating that the structure of the complex was relatively stable; the Rg value (Fig. 4F) remained stable at ~2.1 nm, indicating that the structure of the complex was compact and the SASA (Fig. 4G) of the entire system was relatively stable. Free energy landscape of the complex demonstrated that the conformation of the AS-IV-TUBB4B complex was stable when the Rg values ranged from 2.07 to 2.09 nm and the RMSD was 0.03-0.17 nm (Fig. 4H).

The present study demonstrated the targeted interaction between AS-IV and TUBB4B by verifying whether TUBB4B regulation directly influences the pharmacological efficacy of AS-IV. Immunofluorescence demonstrated the shNC + AS-IV group exhibited partial cell shrinkage, nuclear condensation and blurred or fragmented cytoskeletal structures (Fig. 4J). In the OE-TUBB4B + AS-IV group, some nuclei were markedly lysed and the cytoskeletal components were fragmented and degraded in the cytoplasm, indicating that OE-TUBB4B enhances the inhibitory effect of AS-IV on pituitary tumors. Conversely, in the KD2-TUBB4B + AS-IV group, although nuclei shrank, the cytoskeletal structure remained intact, suggesting that KD-TUBB4B attenuated the inhibitory effect of AS-IV on pituitary tumors. The present study validated the targeted interaction between AS-IV and the TUBB4B protein via CETSA. TUBB4B protein gradually denatured with increasing temperature (Fig. S2A). In AS-IV-treated cells, the thermal denaturation temperature of the TUBB4B protein increased. Compared with that in the shNC group (Fig. S2B), the melting curve in the OE-TUBB4B group shifted to the right (Fig. S2C), whereas the KD2-TUBB4B group showed the opposite trend (Fig. S2D and E), indicating direct binding of AS-IV to TUBB4B. The present study determined the effect of TUBB4B on AS-IV drug sensitivity via viability assays. In the shNC group, the IC₅₀ was 112.8 μM, reflecting baseline sensitivity, whereas in the OE-TUBB4B group, the IC₅₀ was 60.4 μM and that in the KD2-TUBB4B group was 292.1 μM (Fig. S2F). The cells overexpressing TUBB4B exhibited increased sensitivity to AS-IV, whereas KD-TUBB4B expression decreased AS-IV sensitivity. Flow cytometry (Fig. 4I and L), EdU assay (Fig. 4K and M) and western blot analysis were performed for both cell lines (Fig. 4N-S). The cells in the OE-TUBB4B + AS-IV group proliferated more slowly and exhibited a higher apoptosis rate than those in the Vector + AS-IV group; in contrast, cells in the KD2-TUBB4B+AS-IV group proliferated more rapidly and exhibited a lower apoptosis rate compared to the Vector + AS-IV group. These results collectively confirmed that OE-TUBB4B enhanced AS-IV efficacy, whereas KD-TUBB4B attenuated it, indicating that TUBB4B modulates cellular sensitivity to AS-IV. These findings collectively demonstrated that AS-IV targeted TUBB4B to exert its inhibitory effect on pituitary tumors.

To clarify the mechanism by which TUBB4B affects the proliferation and apoptosis of pituitary tumors, the cell cycle of GH3 cells was analyzed. OE-TUBB4B accelerated the G1-to-S phase transition (Fig. 5A and B). RT-qPCR was used to assess cell cycle-related protein expression (Fig. 5C); expression levels of CCND1, CDK4 and CDK6 were significantly increased in the OE-TUBB4B group, consistent with the KEGG pathway enrichment (Fig. 5D). PPI analysis (Fig. 5E) revealed that the TUBB4B protein interacted with STMN1, which was consistent with the data from the STRING database (Fig. 5F). Therefore, the present study focused on the regulation of the ERK/MAPK pathway by the STMN1 protein.

To verify whether changes in STMN1 expression directly relied on TUBB4B regulation, immunofluorescence colocalization of the TUBB4B and STMN1 protein was performed (Fig. S3A). Upon OE of TUBB4B, the green (TUBB4B) and the red fluorescence signal (STMN1) increased. Following KD of TUBB4B expression, the fluorescence signals of both TUBB4B and STMN1 simultaneously decreased, with a corresponding decrease in colocalization. These results indicate that STMN1 expression was associated with TUBB4B levels and that STMN1 colocalized within the microtubule network. Western blot analysis demonstrated OE-TUBB4B increased expression of the STMN1, p-STMN1, ERK and p-ERK proteins (Fig. S3B), whereas KD-TUBB4B decreased expression (Fig. S3C-F). These findings indicate that STMN1 served as an effector molecule in the TUBB4B downstream pathway and was associated with ERK pathway activation.

CCK-8 assay revealed no significant difference in cell viability between the OE-TUBB4B and the shNC group following U0126 treatment (Fig. 5G). Western blot analysis (Fig. 5H-Q) indicated that OE-TUBB4B led to increased expression of p-STMN1, p-cPLA2, p-ERK and CCND1. Additionally, the expression of total STMN1, cPLA2 and ERK proteins increased, and no significant differences in the levels of the JNK and p-JNK protein were detected. Following U0126 treatment, the STMN1, p-STMN1, cPLA2, p-cPLA2, ERK, p-ERK, CCND1, JNK, and p-JNK protein levels in the Vector + AS-IV group or the OE-TUBB4B+AS-IV group did not significantly change. These findings indicated that the ERK/MAPK pathway is involved in proliferation and that blocking this pathway inhibited positive feedback. These findings also suggest that OE-TUBB4B may promote pituitary tumor proliferation through the activation of positive feedback among downstream proteins.

To verify the reliability of the in vitro cell experiments in vivo, a nude mouse subcutaneous xenograft model was established using TUBB4B-regulated GH3 cell lines (Fig. 6A). There was no significant difference in body weight between any groups (Fig. 6B), which was attributed to the benign nature of the GH3 cell line. Compared with the Vector group, both tumor volume and weight were significantly increased in the OE-TUBB4B group, whereas they were decreased in the KD2-TUBB4B group (Fig. 6C and C). Compared with the Vector + AS-IV group, both tumor volume and weight were decreased in the OE-TUBB4B + AS-IV group, whereas they were increased in the KD2-TUBB4B + AS-IV group. Overall, the trends in tumor volume and weight were consistent across the groups. The largest tumor was observed in the OE-TUBB4B group, with a maximum tumor diameter of 12.7 mm, a volume of 892 mm³ and a mass of 1,671 mg. Compared with the shNC group, the OE-TUBB4B group presented notable blood sinuses and the tumor cells exhibited the highest density, large in size, and have a high nucleus-to-cytoplasm ratio, consistent with a state of high proliferation; the KD2-TUBB4B and shNC + AS-IV group presented partial nuclear fragmentation and cytoskeletal fragmentation/degradation in tumor tissue; the OE-TUBB4B + AS-IV group presented notable nuclear pyknosis/fragmentation and severe apoptosis and necrosis of tumor tissue and the KD2-TUBB4B + AS-IV group presented partial nuclear shrinkage but relatively regular tissue morphology (Fig. 6E). Western blotting (Figs. 6F, H-J and S4A-C) and IHC staining (Fig. 6G and K-O) results for the xenograft tumor tissue were consistent with those of the in vitro experiments; compared with the Vector group, the OE-TUBB4B group exhibited increased expression of PCNA and Bcl-2 proteins and decreased expression of Bax and cleaved-caspase-3 proteins; while the KD2-TUBB4B group exhibited the opposite trends in these protein expressions compared to the Vector group; Following AS-IV intervention, the OE-TUBB4B+AS-IV group exhibited lower PCNA and Bcl-2 protein expression compared to the Vector + AS-IV group, while Bax and cleaved-caspase-3 protein expression was higher; conversely, the KD2-TUBB4B+AS-IV group showed the opposite trends in these protein expressions compared with the Vector + AS-IV group. These results were consistent with the aforementioned findings that OE-TUBB4B increased the effectiveness of AS-IV, while KD-TUBB4B attenuated effectiveness. Fig. 7 shows the schematic of the overall mechanism by which AS-IV regulates proliferation and apoptosis in pituitary tumors by targeting of TUBB4B. In summary, AS-IV inhibited proliferation and promoted apoptosis of xenograft tumors in animals by targeting of TUBB4B.