SV-HUC-1 cells (ATCC, CRL-9520) were obtained from the Cell Bank of Chinese Academy of Sciences (Shanghai, China) and cultured in F-12k medium (Gibco, USA) supplemented with 12.5% fetal bovine serum (Gibco, USA). M. penetrans GTU-54-6Al was a kind gift from Professor Jiwen Zhao of Southeast University and has been preserved by the Institute of Pathogenic Biology of Hengyang Medical School University of South China. M. penetrans was cultured using SP-4 medium at 37°C.

The recombinant protein coded by the DNA sequence of the M. penetrans P35 lipoprotein (protein ID: AAC16392.1) was expressed and purified as follows. The full-length DNA segment of M. penetrans P35 lipoprotein gene was codon-optimized and inserted into the pET-30a(+) vector. The recombinant prokaryotic expression vector pET-30a(+)/P35 was transformed into E. coli Rosetta™2 (DE3), which was induced by Isopropyl β-D-1-thiogalactopyranoside (IPTG; Solarbio, 0.5 mM) to express the recombinant hexahistidine-tagged P35 (rP35). After centrifugation at 10 000 rpm for 15 min, the induced expression solution was precipitated with PBS containing 0.1 mM PMSF and lysed by ultrasound at low temperature until the solution became clear. Then the bacteria solution was centrifuged at 12 000 rpm for 15 min, the supernatant and precipitation were collected, and the presence of protein was analyzed by SDS-PAGE. The supernatant was loaded into a column containing Ni-NTA immunoaffinity chromatography and incubated for 6 hours by shock on ice. The effluent was collected and eluted with a linear gradient of 20 to 150mM imidazole. The effluents were analyzed by SDS-PAGE electrophoresis. The purified rP35 was washed with PBS and concentrated on a regenerated cellulose membrane (Millipore, USA) with a molecular weight cutoff of 10 kDa.

Based on secondary structure and hydrophilicity analyses of the P35 lipoprotein, as well as epitope mapping and recognition rates by human serum reported in literatures (Neyrolles et al., 1999b; Sasaki et al., 2002), five potential binding-related peptides were synthesized. Each truncated P35 peptide was tagged with a Flag epitope (DYKDDDDK) using a flexible linker (GGGGSGGGGS). A control peptide containing only the Flag tag and linker sequences (GGGGSGGGGSDYKDDDDK) was also synthesized ( Table 1 ). Peptide synthesis, identification, and purification were performed by Nanjing GenScript Co., Ltd. Rabbit-derived anti-Flag antibody was purchased from Biopm (PMK101M, China).

Three 8-week-old female New Zealand rabbits were purchased from Hunan Taiping Biotechnology Co., LTD. The New Zealand rabbits were immunized using the purified rP35 (150μg) mixed with Freund’s adjuvant. After four immunizations, rabbit serum was collected and antibody titer was determined. The collected rabbit serum was centrifuged 12,000 rpm for 30 minutes, and the supernatants were collected. The supernatants were roughly purified by continuous ammonium sulfate precipitation, and then further purified by CNBr-activated Sepharose 4B (GE Healthcare, Sweden) as described by our group (Peng et al., 2023; Zeng et al., 2012). All antibodies were diluted with TBST (0.05% Tween 20 in Tris-buffered saline) containing 5% skim milk.

When SV-HUC-1 cells were cultured to approximately 10⁵-10⁶ cells, the nutrient medium was changed into fresh serum-free F-12K medium and continued to culture overnight. Then, the ACTG1-siRNA (20 mM, Origin, SR300049, USA) was transfected into the cells using Lipofectamine 3000. After 48 hours of transfection, the medium was supplemented with F-12K containing 10% fetal bovine serum for subsequent experiments.

Protein samples were mixed with loading buffer and boiled for 10 minutes. Proteins were then separated by 10–15% gradient acrylamide gel electrophoresis and transferred to polyvinylidene fluoride (PVDF) membranes (Millipore, R0BB23469, USA). The membranes were blocked for 2 hours at room temperature with TBST containing 5% skim milk, followed by three washes. Membranes were incubated overnight at 4°C with the appropriate primary antibodies. Mouse IgG antibody or rabbit IgG antibody served as negative controls. After washing four times, membranes were incubated with horseradish peroxidase (HRP)-conjugated goat anti-rabbit or anti-mouse IgG secondary antibodies (1:5,000, Proteintech, SA00001-2 or SA00001-1, USA) at 37°C for 1 hour. Following final washes and drying, PVDF membranes were subjected to chemiluminescent detection using BeyoECL Star (Beyotime, China).

The SV-HUC-1 membrane proteins were extracted as described in our previous study (Deng et al., 2018). The protein samples were boiled for 5 minutes, electrophoreted with 12.5% acrylamide gel, and then stained with Coomassie blue. Molecular masses were estimated based on the Protein Ladder (Thermo, 26616, USA). To detect P35 binding proteins on the SV-HUC-1 membrane, the membrane proteins were transferred to PVDF membranes, which were then blocked in TBST containing 5% skim milk at room temperature for 2 hours, and incubated with rP35 (0.5 mg/mL) at 4°C overnight. BSA incubation was served as a negative control. After washing, the membranes were incubated with purified rabbit anti-rP35 antibody (1:50) at 37°C for 2 hours. Incubation of secondary antibody and visualization and recording of protein bands were performed as described in western blot. The protein band intensities were quantified by ImageJ.

After the molecular weight of the target bands were determined according to the results of Coomassie blue, an expected band (kDa) in the improved VOPBA experiment were cut and analyzed using high performance liquid chromatograph-mass spectrometry (HPLC-MS) by Guangzhou Fitgene Biotechnology Co. Ltd.

SV-HUC-1 cells were inoculated into the 24-wells plate (Thermo Fisher, USA) and until reaching a density of 2 × 10⁵ cells at 37°C. Cells were washed twice using PBS and then fixed with 4% paraformaldehyde at 4°C for 30 minutes. After four washes, cells were blocked with F-12K complete medium as a blocker at room temperature for 1 hour. Following additional washes, cells were incubated overnight at 4°C with either rabbit anti-ACTG1 antibody (1:2,000, Antibody System, PHF60501, France) or rabbit anti-KRT8 antibody (1:2,000, Antibody System, PHC17901, France). After PBS washes, cells were incubated with anti-rabbit Alexa-488 secondary antibody (1:200, Jackson, 111-545-003, USA) for 1 hour at 37°C. Nuclei were stained with DAPI (4′,6′-diamidino-2-phenylindole, Beyotime, China). Images were captured using a TE2000-S inverted microscope (Nikon, Japan) at 40× magnification. Fluorescence images were captured sequentially to prevent crosstalk between fluorophores.

As previously described (Peng et al., 2023), far-western blot was used to identify whether full-length rP35 specifically interacts with recombinant target proteins (ACTG1, AtaGenix, ATAP01698, France; KRT8, Antibody System, YHC17901, France) and membrane proteins from SV-HUC-1 cells. Briefly, rP35 samples were boiled for 10 minutes, and then transferred to PVDF membranes. After washing and blocking, membranes were incubated overnight at 4°C with either membrane proteins (500 μg/mL) or the corresponding recombinant target proteins (ACTG1 or KRT8). Following PBS washes, membranes were incubated with rabbit anti-ACTG1 antibody (1:2,000) or rabbit anti-KRT8 antibody (1:2,000) for 2 hours at 37°C, followed by incubation with HRP-conjugated goat anti-rabbit secondary antibody (1:5,000) for 1 hour at 37°C. Protein bands were visualized and recorded as described in the Western Blot section.

Far-western blot was also performed to confirm the interaction between synthesized P35 peptides and ACTG1. Recombinant ACTG1 protein was transferred to the PVDF membranes as described above. After washing and blocking, PVDF membranes were incubated overnight at 4°C with synthesized P35 peptides. Full-length rP35 and a negative peptide served as positive and negative controls, respectively. Following PBS washes, membranes were incubated with rabbit anti-Flag antibody (1:2,000) for 2 hours at 37°C. Secondary antibody incubation, visualization, and recording of protein bands were performed as described in the Western Blot section.

Interactions between full-length rP35 and ACTG1 or KRT8 were assayed using an indirect ELISA. Full-length rP35 (10 μg/well) was suspended in 0.1 M sodium carbonate solution (15 mM Na₂CO3, 35 mM NaHCO3, pH 9.6) and immobilized onto a 96-well polystyrene plate (Corning, USA) overnight at 4°C. After washing with PBST (0.05% Tween 20 in PBS) three times, plates were blocked with 5% skimmed milk in TBST for 2 hours at 37°C. After three additional washes with PBST, wells were incubated with either ACTG1 (0.2 mg/mL) or KRT8 (0.2 mg/mL, Antibody System, China) for 2 hours at 37°C. Wells incubated with BSA were served as negative control, and those with rabbit anti-rP35 antibodies (1:200) served as positive controls. Following washes, wells were incubated with rabbit anti-ACTG1 antibodies (1:2,000, Antibody System) or rabbit anti-KRT8 antibodies (1:2,000, Antibody System) for 2 hours at 37°C. After five washes with PBST, wells were incubated with HRP-conjugated goat anti-rabbit IgG antibody (1:5,000) for color development. Absorbance at 450 nm (A₄₅₀) was measured using a microplate meter (Tecan, Infinite F50, Switzerland).

ELISA was also utilized to detect the binding of truncated P35 polypeptides to immobilized ACTG1. ACTG1 (10 μg/well; AtaGenix) was immobilized on the plate and incubated with synthesized P35 polypeptides (100 μg/mL) for 2 hours at 37°C. Wells incubated with the control peptide served as negative control, and those with full-length rP35 served as positive control. Rabbit anti-Flag antibody (1:2,000, Biopm, PMK101M, China) or purified rabbit anti-P35 antibody was subsequently added, followed by incubation with HRP-conjugated goat anti-rabbit antibody (1:5,000). Absorbance at 450 nm (A450) was measured.

To evaluate the co-localization of rP35 or M. penetrans with ACTG1, SV-HUC-1 cells were pre-incubated with rP35 (50 μg/mL) or M. penetrans (1×10⁷ CCU/mL) for 2 hours at 37°C. After washing, cells were incubated overnight at 4°C with a combination of mouse monoclonal anti-ACTG1 antibody and rabbit anti-rP35 antibodies (1:200) or rabbit anti-Mpe serum (1:10,000) in F-12K medium. Following PBS washes, cells were stained with a combination of anti-rabbit Alexa-488 (1:200) and Cy3 anti-mouse antibody (1:200, Proteintech, China) for 1 hour at 37°C. Nuclei were stained with DAPI. Images were captured using a TE2000-S inverted microscope (Nikon, Japan) at 100× magnification. Fluorescence images were captured sequentially to prevent crosstalk between fluorophores.

200 μL of membrane proteins extract were mixed with 100 μL rP35 (10 μg/mL) and shaken at 4°C overnight. Subsequently, 40 μL of 50% protein A/G-agarose beads (Santa Cruz, USA) working solution was added and incubated on ice for 2 hours to remove nonspecific proteins. Following centrifugation at 14,000 g, the supernatant was divided into five portions. One sample served as positive control, while the other four samples were incubated overnight at 4°C with rabbit anti-rP35 antibody (1:25), mouse anti-ACTG1 antibody (1:50, Abcam, ab52599, UK), mouse IgG (1:50; Servicebio, GB11739, China) and rabbit IgG (1:50, Servicebio, GB111738, China), respectively. Each sample was then incubated with 10 μL of 50% protein A/G-agarose beads working solution on ice overnight to capture the antigen-antibody complexes. The precipitated protein complexs were washed three times and then re-suspended with 60 μL of PBS. Precipitated rP35-ACTG1 complexes were visualized and recorded as described in the Western Blot section.

SV-HUC-1 cells, with and without ACTG1 knockdown via siRNA transfection, were pre-incubated with rP35 (50 μg/mL) or M. penetrans (1 × 10⁷ CCU/mL) for 2 hours at 37°C. Pre-incubated cells were fixed with 4% paraformaldehyde at 4°C for 30 minutes and washed with PBS. To detect adhering rP35 or M. penetrans, cells were incubated overnight at 4°C with either rabbit anti-rP35 antibodies (1:200) or rabbit anti-Mpe serum (1:10,000). Following PBS washes, cells were stained with Cy3 anti-rabbit secondary antibody (1:200) for 1 hour at 37°C and nuclei were stained with DAPI. Images were captured using a TE2000-S inverted microscope (Nikon, Japan) at 40× magnification.

The M. penetrans suspension was incubated for 2 hours at 37°C with rabbit anti-P35 antibody or pre-immune serum, followed by overnight incubation with SV-HUC-1 cells. Cells were then fixed and blocked as described previously. After washing, cells were incubated overnight at 4°C with rabbit anti-Mpe serum (1:10,000). Secondary antibody incubation and image capture were performed as described in the Adhesion Detection Assay.

To further determine the role of ACTG1 in the adhesion of M. penetrans and rP35 to host cells, two sets of adhesion inhibition assays were implemented. In one group, M. penetrans or rP35 suspension was pre-incubated with ACTG1 for 2 hours at 37°C before incubation of the SV-HUC-1 cells. In another group, cells were pre-incubated with ACTG1 antibody (1:2000, Abcam) for 2 hours at 37°C before incubation with M. penetrans or rP35 suspension. The adhesion detection group served as the positive control. The addition of primary and secondary antibodies and the capture of images were described in adhesion detection experiments.

Adhesion inhibition assay also examined whether the pre-incubation of the synthesized P35 polypeptides affected the adhesion of rP35 to SV-HUC-1 cells. The cells were pre-incubated with synthesized P35 polypeptides for 2 hours at 37°C before incubation of the rP35 suspension. The adhesion detection group was the positive control, and the control peptide incubation group was the negative control. The cells were washed and then incubated with rabbit anti-rP35 antibody (1:200) overnight at 4°C. The addition of primary and secondary antibodies and the capture of images are described in adhesion detection experiments. The average integrated optical density of all groups of images was obtained using ImageJ, and then the staining intensity of rP35 or M. penetrans was quantitatively analyzed.

All experiments were performed in triplicate, and data are presented as mean ± standard deviation (SD). Statistical significance was determined using Student’s t-test or one-way ANOVA, as appropriate. A p-value of <0.05 was considered statistically significant. Data analysis was conducted using GraphPad Prism.

The full length P35 protein of M. penetrans was successfully expressed, identified and purified to facilitate the screening of interacting proteins. As shown in Supplementary Figure S1 there were expected bands in the supernatant and precipitation near the molecular weight 35 kDa, indicating that the rP35 protein was expressed successfully. The purified rP35 using Ni-NTA immunoaffinity chromatography was verified with anti-His6x antibody, and the results showed that polyclonal antibody was successfully generated against rP35.

Polyclonal rabbit antibodies against rP35 were generated by immunizing New Zealand rabbits with rP35 and subsequently purified using CNBr activated Sepharose 4B. As depicted in Supplementary Figure S2 of the Supplementary Material , the titer of specific antibody reached 1:2,560,000 after four immunizations. SDS-PAGE analysis revealed the heavy and light chains of the anti-rP35 antibody at approximately 55 kDa and 25 kDa, respectively, confirming the successful purification of the antibody. Furthermore, Western blot validated the successful preparation of the rP35-specific antibody.

Indirect immunofluorescence was employed to determine whether P35 mediates the adhesion of M. penetrans to SV-HUC-1 cells as an adhesion-associated protein. As illustrated in Figures 1B, D , significant red fluorescence was observed on the surface of SV-HUC-1 cells incubated with rP35 or M. penetrans. In contrast, no obvious red fluorescence was detected in the blank control group ( Figures 1A, C ). These observations indicate that both rP35 and M. penetrans are capable of adhering to the surface of SV-HUC-1 cell.

Adhesion inhibition assays further demonstrated that compared with the M. penetrans adhesion group ( Figure 1D ), the red fluorescence on the membrane surface of SV-HUC-1 cells pretreated with pre-immune serum did not change significantly ( Figure 1E ). Conversely, the red fluorescence on the membrane surface of cells pretreated with the anti-rP35 antibody was markedly reduced ( Figure 1F ). Quantitative analysis of the average integrated optical density ( Figure 1G ) confirmed that the rP35 antibody effectively inhibited the adhesion of M. penetrans to SV-HUC-1 cells. Collectively, these findings establish P35 as an adhesion-associated protein facilitating the interaction between M. penetrans and SV-HUC-1 cells.

SDS-PAGE analysis and modified VOPBA conducted to screen the interacting proteins of rP35. As shown in the Figure 2A , the molecular weights of SV-HUC-1 membrane proteins were mainly in the range of 10 kDa~100 kDa. The results of the modified VOPBA manifested that PVDF membrane incubated with rP35 demonstrated two distinct bands in the molecular weight range of 40 kDa to ~55 kDa ( Figure 2B ), while no obvious band appeared in the control group ( Figure 2C ). These results indicated that SV-HUC-1 membrane proteins in the range of 40 kDa to ~55 kDa might be the target proteins specifically binding to rP35.

To identify proteins that specifically bound to rP35, strips near 40 kDa~55 kDa were cut for HPLC-MS analysis. After protein matching and comparison search through NCBI database, it was found that ACTG1 (Actin Gamma 1) near 40 kDa had the highest score ( Supplementary Table 1 ), and KRT8 (Keratin, type II cytoskeletal 8) near 50 kDa had the highest score ( Supplementary Table 1 ). Therefore, ACTG1 and KRT8 are likely membrane proteins that specifically interact with rP35.

The localization of ACTG1 and KRT8 within SV-HUC-1 cells was assessed using Western blot and indirect immunofluorescence. As depicted in Figure 3A , both ACTG1 and KRT8 were present in the membrane components of SV-HUC-1 cells. Indirect immunofluorescence assays further revealed that green fluorescence representing ACTG1 ( Figure 3B ) and KRT8 ( Figure 3C ) was localized in the cytoplasm and on the cell membrane. These results indicate that ACTG1 and KRT8 are predominantly distributed in the cytoplasm and membrane of SV-HUC-1 cells.

To determine whether KRT8 and ACTG1 interact with rP35, four independent approaches were used. First, the results of far-western blot were shown in Figure 4A , there were bands at 35 kDa for the recombinant ACTG1 protein incubated group, while there was no band in the control group, indicating that ACTG1 can directly bind to rP35. However, no obvious band appeared in the KRT8 protein incubated group, indicating that KRT8 protein could not directly bind to rP35. Additionally, we further verified whether rP35 could bind to ACTG1 and KRT8 on SV-HUC-1 cell membranes using far-western blot. As shown in Figure 4B , an obvious band at 35 kDa was observed for the ACTG1 antibody group, while no band was appeared for both the KRT8 antibody group and the control group. These results substantiated that rP35 could interact with ACTG1 on SV-HUC-1 cell membrane, but not with KRT8.

Second, the interaction of ACTG1 or KRT8 with rP35 were assayed by indirect ELISA. As shown in Figures 4C, D , the A₄₅₀ values of rP35 and ACTG1 incubation group was significantly higher than that of the negative control group, while there was no significant difference between rP35-KRT8 incubation group and negative control group, indicating that rP35 specifically bound ACTG1 but did not interact with KRT8, which was consistent with the results of far-western blot.

Indirect immunofluorescence assay was performed to prove whether P35 can colocalize with ACTG1 in SV-HUC-1 cells. Representative fluorescence images in Figures 5B, C revealed that M. penetrans and the rP35 protein (green fluorescence) were detected on the cell membrane and co-localized with ACTG1 (red fluorescence), as shown in the merged images (yellow fluorescence).

The interaction between ACTG1 and rP35 was further verified using immunoprecipitation. The results showed that both the anti-rP35 antibody and anti-ACTG1 antibody precipitated samples had obvious band at 42 kDa ( Figure 5A-3 ) and 35 kDa ( Figure 5A-6 ), respectively, while the control IgG precipitation group had no obvious band ( Figures 5A-1, 5A-4 ). Collectively, these data corroborated that P35 could interact with ACTG1 on the cell membrane of SV-HUC-1 cells.

Adhesion and adhesion inhibition assays were conducted to evaluate whether ACTG1 can inhibit the adhesion of rP35 and M. penetrans to SV-HUC-1 cells. As shown in Figure 6 , both rP35 ( Figure 6A ) and M. penetrans ( Figure 6D ) could adhere to the membrane surface of SV-HUC-1 cells. However, the adhesion of rP35 ( Figure 6B ) and M. penetrans ( Figure 6E ) to ACTG1-treated cells was decreased. Similarly, the adhesion of rP35 ( Figure 6C ) and M. penetrans ( Figure 6F ) to anti-ACTG1 antibody pre-incubated cells were partly inhibited. Quantitative analysis revealed that the average fluorescence intensities for the ACTG1 pre-treatment group and the anti-ACTG1 antibody pre-incubation group were significantly lower than those of the positive control group (P < 0.001; Figures 6G, H ). Additionally, as shown in Figure 7A , the expression of ACTG1 in SV-HUC-1 cells transfected with ACTG1-siRNA was significantly decreased and the adhesion of rP35 ( Figure 7C ) and M. penetrans ( Figure 7E ) to cells was also decreased significantly compared with positive controls ( Figures 7B, D ). Quantitative staining intensity analysis confirmed that rP35 and M. penetrans exhibited significantly decreased adhesion after ACTG1-siRNA treatment (P < 0.001; Figures 7F, G ). Collectively, these data indicate that ACTG1 plays a crucial role in mediating the adhesion of rP35 and M. penetrans to SV-HUC-1 cells.

Indirect ELISA was carried out to identify the binding functional domains of P35 to ACTG1. As shown in Figure 8A , the A₄₅₀ values of P35-1 and P35-4 polypeptide groups were significantly higher than those of the P35-2, P35-3, P35-5 and control peptide group, indicating that P35-1 and P35-4 might bind to ACTG1 in vitro. To confirm the results of the ELISA assay, the binding of ACTG1 to rP35 and truncated polypeptides were also detected using far-western blot. As shown in Figure 8B , the PVDF membrane incubated with rP35 (Lane 1) and P35-4 (Lane 5) showed obvious bands at 42 kDa, while other polypeptides and control peptide did not have obvious band, which demonstrated that P35-4 peptides can bind to ACTG1.

Results of adhesion inhibition experiments were shown in Figure 9 . Compared with the positive control group ( Figure 9A ), the adhesion of rP35 to SV-HUC-1 cells pre-incubated with P35-1 peptide ( Figure 9B ) was relatively reduced, and the adhesion of rP35 to cells pre-incubated with P35-4 ( Figure 9E ) was significantly reduced. After pre-incubated with P35-2 ( Figure 9C ), P35-3 ( Figure 9D ), P35-5 ( Figure 9F ) and control ( Figure 9G ) peptides, the adhesion of rP35 to SV-HUC-1 cells was not significantly reduced compared with the positive control group ( Figure 9A ). Compared with the control group, the mean fluorescence intensity of rP35 attached to cells was statistically significant for P35-1 and P35-4 pre-incubation group ( Figure 9H ). In summary, P35-4 could significantly interact with ACTG1, and peptide P35-1 might interact with ACTG1 with low affinity, while other peptides and control peptide could not bind to ACTG1. These results substantiated that the 35-42 (P35-1) and 179-186 (P35-4) amino acids of P35 lipoprotein were possible binding functional domains that mediate P35 interacting with ACTG1.