A total of 30 human SACC tissues and 20 normal and adjacent tissues were collected at the Second Affiliated Hospital of Dalian Medical University from June 2017 to September 2019. The patients diagnosed with SACC were included with no restrictions in terms of age and sex, but excluded with treatment of radiotherapy or chemotherapy. The related clinical information is listed in Table SI. The procedures of the present study concerning human subjects were approved (approval no. DY2021-005) by the Medical Ethics Committee of Dalian Medical University (Dalian, China), and written informed consent was provided by each patient.

Immunohistochemical staining was performed according to procedures described in a previous study by the authors (24). Briefly, 4-µm thick sections were deparaffinized in dimethylbenzene and rehydrated in graded ethanol after baking at 65°C for 1 h. Following antigen retrieval, 3% hydrogen peroxide solution was used to inactivate endogenous peroxidase for 15 min at room temperature. After being blocked with goat serum for 30 min at 37°C, tissues were incubated overnight at 4°C with a primary antibody: LIS1 (1:200; product code ab2607; Abcam); Ki67 (1:3,000; cat. no. 27309-1-AP; ProteinTech Group, Inc.); Pan-CK (1:3,000; cat. no. 26411-1-AP; ProteinTech Group, Inc.) MMP2 (1:200; product code ab86607; Abcam); MMP9 (1:200; product code ab38898; Abcam); and MMP14 (1:100; product code ab78738; Abcam). The following day, the secondary antibody and horseradish peroxidase streptavidin (both contained in a kit; 1:1; cat. no. SAP-9100; ZSGB-BIO, Inc.) were added to the slices for 30 min at 37°C. The sections were then incubated with 3,3′-diaminobenzidine (DAB; ZSGB-BIO, Inc.) as a chromogen substrate and counterstained with hematoxylin for 3 min at room temperature. Finally, the slices were dehydrated with graded ethanol, cleared with xylene and mounted with neutral gum (Solarbio Science & Technology, Co., Ltd.). For the evaluation, images were obtained using a phase microscope (Olympus Corporation), and the integral optical density (IOD) and the positive immunohistochemical staining region were measured with Image-Pro Plus 7.0 software (Media Cybernetics, Inc.). Protein expression was semi-quantitatively calculated and classified based on the intensity of staining and the percentage of stained cells as previously described (25).

The SACC-LM cell line was a kind gift from Professor Liu (Affiliated Stomatological Hospital of Fudan University, Shanghai, China) (3) and was cultured in DMEM/F12 supplemented with 10% fetal bovine serum (FBS; ScienCell Research Laboratories, Inc.), 100 U/ml penicillin and 100 U/ml streptomycin (Gibco; Thermo Fisher Scientific, Inc.) at 37°C in a humidified atmosphere with 5% CO₂. The plasmids, including the pc-DNA3.1 vector and plasmids encoding LIS1 and CLIP170 were constructed by Changsha YouBio Co., Ltd. The small interfering RNAs (siRNAs) were purchased from Shanghai GenePharma Co., Ltd. and the target sequences are provided in Table SII. The cells were seeded into 6-well plates at 4×10⁵ cells per well and cultured to 70-80% confluence. Lipofectamine 3000 Transfection Reagent (3.75 µl; cat. no. L3000015; Invitrogen; Thermo Fisher Scientific, Inc.) was used for cell transfection with 2,500 ng plasmids or 37.5 pmol siRNAs. The transfection system was incubated for 8 h at 37°C. The subsequent experiment was performed at 24 h after transfection. In addition, siRNA scramble (si-control, Shanghai GenePharma Co., Ltd.) was used as the negative control.

Total RNA was extracted from SACC-LM cells with RNAiso Plus (TRIzol; Takara Bio, Inc.), and then, reverse transcription into cDNA was performed using a PrimeScript RT reagent Kit (Takara Bio, Inc.) according to the manufacturer's instructions. The primer sequences for RT-qPCR are provided in Table SIII. GAPDH was used as the internal control for each experiment. RT-qPCR was performed with SYBR Premix Ex Taq (Tli RNaseH Plus; Takara Bio, Inc.). The PCR conditions were as follows: Initial denaturation at 95°C for 30 sec, followed by 40 cycles of denaturation at 95°C for 5 sec, annealing at 55°C for 30 sec and elongation at 72°C for 30 sec, and then a final extension at 72°C for 5 min. The data were collected with a TP800 system (Takara Bio, Inc.) and analyzed using the 2^−ΔΔCq comparative method (26).

Cell proliferation was assessed using Cell Counting Kit-8 (CCK-8; APeXBio Technology, LLC). According to the manufacturer's instructions, cells were seeded in a 96-well plate at a density of 1×10³ cells/well in 100 µl of culture medium for 0-48 h. CCK-8 solution (10 µl) was added to each well of the plate, and then, the plate was incubated for 1 h in an incubator at 37°C. Finally, the absorbance was measured at 490 nm using a microplate reader (Shanghai Flash Spectrum Biotechnology Co., Ltd.).

An EdU assay kit (product code ab219801; Abcam) was used to assess cell proliferation. Following cell dissociation, the cells were diluted to 1×10⁶ cells/ml with culture medium and labeled with EdU solution for 4 h. The cells were then fixed with 4% paraformaldehyde for 15 min at room temperature, permeabilized in permeabilization buffer and incubated with EdU reaction mix at room temperature for 30 min. Finally, the number and proportion of EdU-incorporated cells were analyzed on a flow cytometer (FACSVerse; BD Biosciences) at Ex/Em=491/520 nm. The data were analyzed using Cell Quest software v5.1 (BD Biosciences).

For cell samples, following transfection, the cells were seeded onto coverslips. After adherence, the cells were fixed with 4% paraformaldehyde for 20 min at 37°C and then permeabilized with 0.25% Triton X-100 in PBS for 5 min. For 5 µm paraffin sections, after deparaffinization and rehydration, the tissues were permeabilized with 20 µg/ml Proteinase K in PBS for 5 min according to the instructions of a TUNEL Bright Apoptosis Detection Kit (cat. no. A112-01; Vazyme Biotech, Co., Ltd.) and a previous study by the authors (27). Briefly, the sections were incubated with the Equilibration Buffer for 30 min at room temperature and then marked with the labeling mix buffer for 1 h at 37°C. Finally, the sections were mounted with 4′,6-diamidine-2′-phenylindole dihydrochloride (DAPI; 1:1,000; cat. no. 10236276001; Roche Diagnostics) at 37°C for 10 min, and five visual fields were randomly selected for observation using fluorescence microscopy (Olympus Corporation).

Following transfection with LIS1 or si-CLIP170, SACC-LM cells were seeded in 6-well plates at a density of 5×10⁵ cells per well with or without treatment of ML141 (Cdc42 inhibitor; MedChemExpress). A 200-µl pipette tip was used to create scratches in the cell monolayer when cells were merged to 75-80% confluence. The cells were then washed with PBS and cultured in serum-free medium. Images were obtained using a light microscope (Olympus Corporation) at 0, 12 and 24 h to record the wound area. Finally, the area of cell migration was quantified using Image-Pro Plus 7.0 software.

For the cell migration and invasion assays, cells (5×10³/well) in 200 µl serum-free medium were directly seeded into the upper chambers (pore size, 8 µm; Corning, Inc.) coated with (at 37°C for 1 h; for the invasion assay) or without (for the migration assay) Matrigel (Corning, Inc.). Subsequently, 800 µl medium with 20% FBS was added to the lower chambers. Following culture for 12 or 24 h in an incubator, the cells were fixed with 4% paraformaldehyde (Solarbio Science & Technology, Co., Ltd.) for 20 min at 37°C and then stained with 1% crystal violet (Beijing Coolibo Technology Co., Ltd.) for 10 min at room temperature. The upper surface of the membranes was wiped with a cotton swab to remove the non-penetrating cells. Under a phase microscope at a high magnification (×200), five fields in each sample were randomly selected, and the number of migrated or invasive cells was counted.

Invadopodia of tumor cells were detected by colocalization of cortactin and F-actin according to a previous study by the authors (28). Cells were fixed with 4% paraformaldehyde for 20 min at 37°C, washed three times with PBS, and then permeabilized with 0.25% Triton X-100 in PBS for 15 min at room temperature. Following washing, the cells were blocked with 2.5% bovine serum albumin (BSA; Solarbio Science & Technology, Co., Ltd.) for 1 h at 37°C and stained with the primary antibody, cortactin (1:1,000; product code ab81208; Abcam) overnight at 4°C. Following three washes, the cells were incubated with Alexa Fluor 488 secondary antibody (1:400; cat. no. A-11034; Thermo Fisher Scientific, Inc.) and rhodamine-conjugated phalloidin (1:400; product no. 40734ES75; Yeasen Biotechnology Co., Ltd.) at 37°C for 1 h. Finally, the cell nuclei were counterstained with DAPI (Roche) for 5 min at room temperature away from light, and the coverslips were mounted with fluorescence decay resistance medium (Solarbio Science & Technology, Co., Ltd.). The sections were observed and images were captured using a confocal microscope (Leica Microsystems, Inc.). For evaluation, five random high-power fields were selected, and the threshold was set based on the staining of the negative control. Colocalization of cortactin and F-actin was considered to indicate invadopodia formation. The area of invadopodia per cell was quantitatively calculated using Image-Pro Plus 7.0 software.

Cells (2×10⁶) were lysed using radioimmunoprecipitation assay lysis buffer (RIPA; Solarbio Science & Technology, Co., Ltd.) and assessed with a bicinchoninic acid protein assay kit (BCA; Beyotime Institute of Biotechnology). Subsequently, 20 µg protein was loaded into a 10% polyacrylamide gel, separated by electrophoresis and transferred to a polyvinylidene difluoride membrane (PVDF; Millipore; Merck KGaA). The membrane was blocked with 5% non-fat milk for 1 h at room temperature and incubated with a primary antibody at 4°C overnight: GAPDH (1:6,000; cat. no. 60004-1-lg; ProteinTech Group, Inc.); LIS1 (1:1,000; product code ab2607; Abcam); CLIP170 (1:1,000; product code ab61830; Abcam) MMP2 (1:1,000; product code ab86607; Abcam); MMP9 (1:1,000; product code ab38898; Abcam); MMP14 (1:500; product code ab78738; Abcam); BCL2 (1:1,500; cat. no. 12789-1-AP; ProteinTech Group, Inc.); cyclin D1 (1:250; cat. no. sc-8396; Santa Cruz Biotechnology, Inc.); Cdc42 (1:1,000; product code ab187643; Abcam) and then the secondary HRP-conjugated antibody (1:2,000; cat. no. ZB-2305; ZSGB-BIO, Inc.) at room temperature for 1 h. Finally, the proteins on the membranes were visualized using a Supersignal West Femto Kit (cat. no. 34094; Thermo Fisher Scientific, Inc.) and scanned with an imager (ChemiDoc™ Touch Imaging System; Bio-Rad Laboratories, Inc.). The ImageJ software (version 1.51j8; National Institutes of Health) was used for densitometric analysis.

Full-length cDNA encoding human LIS1 was cloned into pLenti-CMV-EGFR-Puro with an HA tag. The LIS1 short hairpin RNA (sh-LIS1) and shRNA scramble as the negative control (sh-control) were constructed using GV248-EGFP-Puro. The sh-LIS1 target sequence was 5′-GCA TGT GGT AGA ATG CAT T-3′, and the sh-control target sequence was 5′-TTC TCC GAA CGT GTC ACG T-3′. The 3rd generation system was used to recombine lentiviruses, and 2.5 µg transfer plasmid, 1.25 µg pMDLg/RRE, 0.75 µg VSVG and 0.5 µg pRSV-Rev were transfected into 293FT cells (cat. no. R70007; Thermo Fisher Scientific, Inc.) in a 6-cm dish for 8 h at 37°C. The culture medium with recombinant lentiviruses was collected after 24 and 48 h of transfection. Lentiviruses with plasmids and sh-RNA were provided by Shanghai GeneChem Co., Ltd. and used to infect SACC-LM cells at a multiplicity of infection (MOI) of 10 for 48 h at 37°C. Stable cell lines were selected with 10 µg/ml puromycin (Beijing Coolibo Technology Co., Ltd.) for 4 weeks. The expression of LIS1 in SACC-LM cells was detected by qPCR and western blotting.

A total of 40 female nude mice (BALB/c; 4 weeks old; weight, 18-20 g) were obtained from Dalian Medical University Laboratory Animal Center. The animal experiments were approved (approval no. AEE20017) by the Animal Experimental Ethics Committee of Dalian Medical University. The 40 mice were housed at 23±2°C and 50±10% humidity, and under a 12-h light/dark cycle with access to food and water ad libitum. The mice were divided into eight groups randomly (4 groups for the xenograft models and 4 groups for the metastasis models). For the xenograft models, 1×10⁷ SACC cells stably transfected with vector, LIS1, sh-control and sh-LIS1, respectively, in 100 µl PBS were subcutaneously injected into the right flank of the mice (n=5 per group) under anaesthetization by an intra-peritoneal injection of sodium pentobarbital (30 mg/kg). The volume of the tumor (length x width²/2) (29) and the weight of the mice were monitored every two days throughout the study period. In compliance with the Guide for the Animal Care and Use Committee (ACUC) from the US National Institutes of Health, all tumor burdens were not >10% body weight of the mice, and the maximum tumor volume was 751.04 mm³. After 1 month, the 20 mice were euthanized through the intravascular administration of an overdose of sodium pentobarbital (>100 mg/kg) followed by cervical dislocation. Euthanasia was confirmed by the loss of vital signs (respiration and heartbeat cessation), and the harvested tumors were separated and fixed with 4% paraformaldehyde at 4°C for 48 h for subsequent histochemical staining. For the metastasis models, 5×10⁶ SACC cells stably transfected with vector, LIS1, sh-control and sh-LIS1, respectively, in 100 µl PBS were injected into the mice via the tail vein (n=5 per group) under anaesthetization by an intraperitoneal injection of sodium pentobarbital (30 mg/kg). The mice were weighed and observed every other day. After one month, an IVIS Spectrum system (PerkinElmer, Inc.) was used to detect tumor metastasis by tracking cells with stable green fluorescent protein. Living Image Software (version 4.4; PerkinElmer, Inc.) was used to quantify tumor metastasis. According to the humane endpoints in ACUC, among the 20 metastasis model mice, 10 mice were euthanized through the intravascular administration of an overdose of sodium pentobarbital (>100 mg/kg) followed by cervical dislocation in advance due to adverse complications (loss of body weight >20% and uncoordinated movement) after tail vein injection of tumor cells, and the other 10 mice were euthanized at two months after injection. The lungs of the mice were then isolated and fixed for subsequent immunohistochemical staining.

Following transfection, the cells in 6-well plates were lysed with 300 µl IP lysis buffer (Thermo Fisher Scientific, Inc.) per well at 4°C for 5 min. The lysis buffer was then centrifuged for 13,000 × g at 4°C for 10 min to obtain the supernatant. All samples were processed using a Pierce Co-Immunoprecipitation (Co-IP) Kit (cat. no. 88804; Thermo Fisher Scientific, Inc.) according to the manufacturer's recommendations. Cell lysates were incubated with a primary antibody: LIS1 (1:50; product code ab2607, Abcam); or CLIP170 (1:50; product code ab61830; Abcam), overnight at 4°C, and then, the antigen and antibody complexes were combined with 250 µg protein A/G magnetic beads at room temperature for 1 h. After the beads were collected by a magnetic stand, the complexes were eluted for 10 min, and the expression of immunoprecipitated proteins was detected by western blotting.

All statistical analyses were performed using GraphPad Prism 7 (GraphPad Software, Inc.). Unpaired two-tailed t-test was used for two groups, and one-way analysis of variance followed by Tukey's post hoc test was used for multiple comparisons. Kaplan-Meier with log-rank testing was performed for the survival analysis of the mice. All experiments were performed at least in triplicate. All data are presented as the mean ± SEM. P<0.05 was considered to indicate a statistically significant difference.

To evaluate whether LIS1 is a therapeutic target for a promising SACC treatment strategy, the expression and distribution of LIS1 in the tissues and adjacent normal tissues (ANTs) of patients with SACC were examined using IHC staining. As shown in Fig. 1A, the expression of LIS1 was markedly increased in most SACC tissues compared with ANTs. The results also revealed that the positive staining area of LIS1 was mainly localized in the cytoplasm of cancer cells rather than in the nucleus. By analyzing the mean density of LIS1, it was determined that there was significantly higher LIS1 expression in SACC tissues than in ANTs. SACC consists of two main cell types: Ductal and modified myoepithelial cells. These two types of cells exhibited three basic growth patterns: Cribriform, tubular, and solid. Among the three patterns, it was determined that the expression level of LIS1 was higher in solid SACC tissues (low differentiation) than in the other two pattern types (high differentiation) (Fig. 1B).

In addition, the proportion of high and low LIS1 expression levels were analyzed in ANTs and SACC (Table SI). The results revealed that in ANTs, the proportion with low and high LIS1 expression was 80 and 20%, respectively. By contrast, low expression of LIS1 accounted for only 36.67%, and high expression of LIS1 was as high as 63.33% in SACC samples (Fig. 1C). Moreover, the expression level of LIS1 was examined in metastatic and non-metastatic samples and it was determined that the expression of LIS1 was significantly higher in metastatic samples than in non-metastatic samples (Fig. 1D). In summary, the aforementioned results indicated that LIS1 was highly expressed in SACC and that its level was positively associated with the malignancy of SACC.

To investigate the effect of LIS1 on SACC-LM cells, LIS1 overexpression plasmids were first constructed for transient transfection and siRNAs targeting the human LIS1 sequence were employed to overexpress and knock down LIS1 in SACC-LM cells, respectively (Fig. S1A-D). As shown in Fig. 2A and D, CCK-8 assays revealed that overexpression of LIS1 significantly enhanced the proliferation ability of tumor cells compared with the control group, while knockdown of LIS1 (si-LIS1) significantly suppressed cell proliferation. EdU incorporation assays revealed that the percentage of EdU-positive cells was higher at 48 h after transfection with the LIS1 plasmid than after transfection with the vector plasmid (Fig. 2B and C). In the si-LIS1 group, the percentage of EdU-positive cells was significantly decreased compared with that in the si-control group (Fig. 2E and F). The effect of LIS1 was further explored on the apoptosis of SACC-LM cells using TUNEL assays at 48 h after transfection. The results showed that the ratio of apoptotic cells was significantly reduced in tumor cells with overexpressing LIS1 in contrast to the control group. Conversely, the ratio of apoptotic cells was increased in tumor cells with knockdown of LIS1 expression (Fig. 2G and H). Therefore, the expression level of LIS1 was positively associated with the proliferation of tumor cells and negatively associated with apoptosis, indicating that LIS1 can promote tumor cell proliferation and anti-apoptosis in SACC.

As the key cellular microtubule dynein, LIS1 is highly relevant to cell movement (30,31). To identify the effect of LIS1 on the invasion and metastasis of SACC cells, wound healing assays were first used to detect horizontal cell migration. The results revealed that the migration rate of LIS1 cells was faster than that of vector cells, while that of si-LIS1 cells was markedly slower than that of si-control cells (Fig. 3A-C), and the differences were statistically significant. Moreover, the chemotactic migration and invasion abilities of SACC-LM cells were assessed using a Transwell system. As revealed in Fig. 3D-F, the number of cells that penetrated the Matrigel-coated or uncoated membrane increased following overexpression of LIS1 and decreased after knockdown of LIS1 expression in SACC-LM cells.

MMPs are enzymes that can degrade the components of the extracellular matrix (ECM). Studies have demonstrated that MMP2 and MMP9 are secreted into the extracellular space via invadopodia (32). MMP14 is usually located on the surface of the invadopodia of tumor cells, degrades the ECM and is required for the activation of MMP2 and MMP9 (33). Therefore, the formation of invadopodia is an important factor that promotes tumor cell migration and invasion. Next, through double IF staining, it was observed whether LIS1 could regulate invadopodia in SACC-LM cells. As revealed in Fig. 3G, colocalization of cortactin and F-actin was considered to indicate the presence of invadopodia in SACC-LM. The results revealed that the number of invadopodia in LIS1-overexpressed cells was significantly increased and that in LIS1-knockdown cells it was decreased compared with the relative control groups. Quantitative analysis of the invadopodia-positive area/cells further confirmed that LIS1 promoted the formation of invadopodia in SACC-LM cells (Fig. 3H). The expression levels of MMPs was then detected via western blotting. As expected, LIS1 markedly enhanced the expression of MMP2, MMP9 and MMP14. The expression of MMP14 in the LIS1 overexpression group was nearly double that in the vector group. By contrast, the levels of MMPs were suppressed when the expression of LIS1 was knocked down in SACC-LM cells (Fig. 3I and J). These results indicated that LIS1 may regulate the migration and invasion of tumor cells by mediating the formation of invadopodia in SACC.

To corroborate the role of LIS1 in the promotion of tumor cells in vivo, SACC xenograft models were established via subcutaneous injection of SACC-LM cells in nude mice (Fig. 4A). LIS1 overexpression and knockdown stably transfected cell lines were primarily constructed with green fluorescent protein via recombinant lentiviruses. The overexpression and knockdown efficiencies of the stable cell lines are presented in Fig. S2. Following injection in the xenograft models, the tumor volumes and mouse weights were measured every other day (Fig. 4C). The tumor growth curve began to exhibit different trends at approximately 15 days after injection. The xenografts with LIS1-overexpressing cells grew faster than those with vector cells. Subsequently, one month later, the mice were euthanized, and the tumors were completely isolated. As revealed in Fig. 4B, the tumor volumes with high LIS1 expression were larger than those with low LIS1 expression. Next, the xenografts were weighed and statistically analyzed, and the results further indicated that the weight of tumors in the LIS1-overexpressed group was approximately 1.5 times that in the vector group, while the weight of tumors in the sh-control group was almost twice that in the sh-LIS1 group (Fig. 4D). Moreover, TUNEL assays and Ki67 immunohistochemical staining were performed to detect the tumor tissue sections. The results of the TUNEL assay demonstrated that knockdown of LIS1 significantly induced tumor cell apoptosis in vivo, and the number of apoptotic tumor cells was approximately 5 times that in the sh-control group. Compared with the vector group, the number of Ki67-positive cells was significantly increased in the LIS1 overexpression group (Fig. 4E-G). Collectively, the in vivo results demonstrated that LIS1 promoted the proliferation and inhibited the apoptosis of SACC cells, which was consistent with the in vitro results.

To further clarify whether LIS1 can promote the metastasis of SACC cells in vivo, a SACC metastasis model was constructed using mice. Following injection of SACC-LM cells into nude mice via the tail vein, the survival rate of the mice was continuously monitored for two months. The results showed that the mortality of mice in the LIS1 overexpression group was the highest, and this was the first experimental group with the death of a mouse. In the LIS1 overexpression group, the first mouse succumbed at approximately one month after injection, and the mouse mortality was as high as 80% at two months. Among the four groups, the survival rate was the highest in the sh-LIS1 group, and the first mouse death occurred approximately one and a half months after injection (Fig. 5A). Therefore, the survival curve revealed that reducing the level of LIS1 in SACC cells significantly prolonged the life of mice with metastatic SACC. Because the constructed stable cell line exhibited green fluorescence, live animal imaging was performed one month after the injection to observe the metastasis of tumor cells in each group. As shown in Fig. 5B, after tail vein injection of SACC-LM cells, the tumor cells mainly metastasized to the lungs of mice. The fluorescence intensity in the LIS1 overexpression group was significantly stronger than that in the vector group, while the fluorescence intensity in the sh-LIS1 group was significantly weaker than that in the sh-control group. A total of two months after establishment of the model, the lung tissue of mice was completely isolated for observation. As revealed in Fig. 5C, in the LIS1 overexpression group, numerous metastatic and necrotic areas were found in the lungs, and the number of metastatic nodules in the lungs was the highest. The morphology of the lung lobes was destroyed, and the color of the lungs was gray. In the sh-LIS1 group, the number of lung metastases was decreased compared with the sh-control group, and the morphology and color of the lungs were normal.

After embedding and sectioning, hematoxylin and eosin (H&E) and Pan-CK IHC staining of mouse lung tissue was performed. Compared with the vector group, the tumor cells in the LIS1 overexpression group exhibited severe lung metastasis, forming multiple large cancer nests. Compared with the sh-control group, the cancer nests in lung tissue in the sh-LIS1 group were rare, and the pulmonary alveolar structure was normal (Fig. 5D and E). Next, the expression of MMPs in mouse lung tissues was detected via IHC staining. As revealed in Fig. 5F and G, the levels of MMP2, MMP9 and MMP14 in the LIS1 overexpression group were higher than those in the vector group, while the levels of MMP2, MMP9 and MMP14 in the sh-LIS1 group were lower than those in the sh-control group, and the differences were significant. These results indicated that LIS1 significantly regulated the expression of MMPs and promoted lung metastasis of SACC tumor cells in vivo.

Cdc42 is a well-known member of the Ras homolog (Rho) family. It regulates crucial cellular processes, including the cell cycle, and cell cytoskeleton organization (34). To investigate whether LIS1 can regulate the Cdc42 pathway, the expression of Cdc42 in SACC cells was assessed using western blotting. The results revealed that the expression of LIS1 was positively associated with Cdc42 in SACC-LM cells (Fig. 6A). Following overexpression of LIS1, the inhibitor of Cdc42 was added in SACC-LM cells. The results of the CCK-8 assay revealed that the proliferation ability of tumor cells was significantly decreased compared with the LIS1 overexpression group, indicating that Cdc42 may be a pivotal downstream factor of LIS1 in SACC-LM cells (Fig. 6B). Studies have shown that the WD-40 repeat domain of the LIS1 protein can interact with the second zinc finger motif at the COOH-terminus of CLIP170 to regulate the dynamic balance of microtubules and microfilaments, thus playing an essential role in cell division and cell migration (13,16). To verify whether LIS1 and CLIP170 can interact with each other in SACC-LM cells, Co-IP was performed. When LIS1 was overexpressed in SACC-LM cells, the level of CLIP170 expressed with IP-LIS1 was significantly higher than that in the vector group, and vice versa, indicating that LIS1 and CLIP170 could bind together in SACC-LM cells (Fig. 6C). Next, the siRNA-CLIP170 sequence was used to knock down CLIP170 in SACC-LM cells (Fig. S1E and F). Using CCK-8 and TUNEL assays, when the expression of CLIP170 was knocked down after overexpression of LIS1, the proliferation of tumor cells was significantly decreased, and the apoptosis of tumor cells was increased compared with the LIS1 overexpression group. Interestingly, when Cdc42 inhibitor was added in SACC-LM cells with double-transfection of LIS1 and si-CLIP170, the proliferation of cells was not significantly inhibited, and the apoptosis level was also not increased compared with the LIS1 and si-CLIP170 double-transfected group, suggesting that LIS1 needs to interact with CLIP170 to activate Cdc42, so as to enhance the proliferation and anti-apoptosis ability of tumor cells (Fig. 6D-F). Moreover, the expression levels of the proliferation-related gene cyclin D1 and the apoptosis-related gene BCL2 were further detected via western blotting (Fig. 6G and H). Overexpression of LIS1 in SACC cells upregulated cyclin D1 and BCL2, while knockdown of CLIP170 after overexpression of LIS1 in tumor cells significantly reduced the levels of cyclin D1 and BCL2. However, compared with the LIS1 and si-CLIP170 double-transfection group, the expression of cyclin D1 and BCL2 in tumor cells was not significantly decreased in the LIS1 + si-CLIP170 + ML141 group. These results demonstrated that LIS1 interacted with CLIP170 to form the LIS1/CLIP170 protein complex, which could activate the Cdc42 signaling pathway to promote the proliferation and anti-apoptosis of SACC cells.

Numerous studies have demonstrated that Cdc42 is closely related to cell migration. Cao et al (35) revealed that Cdc42 regulates the invasive ability of breast cancer cells by regulating the formation of lamellipodia and expression of MMPs. First, wound healing and Transwell assays were used to verify whether the LIS1/CLIP170 complex could regulate the migration and invasion of SACC cells via the Cdc42 pathway. The results revealed that compared with the LIS1 overexpression group, knockdown of CLIP170 after overexpression of LIS1 in SACC cells significantly reduced the migration and invasion of tumor cells, while the Cdc42 inhibitor could not significantly decrease the migration and invasion of tumor cells in the double-transfection of LIS1 and si-CLIP170 group (Fig. 7A-D). In addition, through double IF staining, changes in the formation of invadopodia were detected in SACC cells after treatment. The results showed that when CLIP170 was knocked down after overexpression of LIS1, the number of invadopodia formed by tumor cells was significantly reduced. Concurrently, it was determined that the number of invadopodia in LIS1 + si-CLIP170 + ML141 group was not less than that in the LIS1 + si-CLIP170 group (Fig. 7E and F). Moreover, the expression of MMP proteins was assessed via western blotting, and the results were consistent with the aforementioned results (Fig. 7G and H). Therefore, these results demonstrated that LIS1 may interact with CLIP170 to form the LIS1/CLIP170 protein complex and then activate Cdc42 to regulate the expression of MMPs by regulating the formation of invadopodia in SACC cells, thus promoting degradation of the ECM to enhance metastasis.