The present study was approved by The Clinical Medical Research Ethical Committee of Tianjin Third Central Hospital (Tianjin, China; approval no. IRB2023-039-01) for the use of clinical biopsy specimens, and informed written consent was obtained from the patients. Tissues from a total of 60 cases of primary BC (including invasive ductal carcinoma and invasive lobular carcinoma), and 22 cases of benign breast tissues were collected from Tianjin Third Central Hospital (Tianjin, China). The inclusion criteria were as follows: i) Patients with pathological diagnosis of BC; and ii) patients who underwent radical mastectomy with no adjuvant treatment prior to and after surgery. Patients with serious heart, liver and kidney injuries, serious infections, endocrine and blood system diseases were excluded. In addition, the clinicopathological parameters of patients were all collected.

The human normal mammary epithelial cell line MCF-10A, the BC cell lines MDA-MB-231 and MCF7, and the embryonic kidney cell line 293T were purchased from the American Type Culture Collection. Cells were grown in Dulbecco's modified Eagle medium (DMEM; Biological Industries) supplemented with 10% Hyclone fetal bovine serum (FBS; Cytiva) at 37˚C in an atmosphere of 5% CO₂.

293/293T cells (American Type Culture Collection) were transfected with the lentiviral packaging Lenti-Mix vector (pMDL, pVSVG or pRSV-Rev, 5:3:2; Thermo Fisher Scientific, Inc.) carrying specific small hairpin RNA (shRNA) targeting nucleotides to PTX3 (PTX3 shRNA) to generate stable cell lines. The sequence of the selected PTX3 shRNA oligomer was 5'-GCATCCTGTGAGACCAATGAA-3'. The plasmid backbone was pLVX-shRNA (Thermo Fisher Scientific, Inc.) and 12 µg lentiviral plasmid plus 12 µg Lenti-Mix were used for transfection. A third generation system was used and a MOI of ~40 was used to infect cells. Briefly, lentiviral supernatants were collected, centrifuged at 3,000 x g at 4˚C for 10 min and filtered through a 0.45-µm filter. MDA-MB-231 and MCF7 cells were subsequently infected with lentiviral supernatant, incubated at 37˚C overnight and selected with 1.5 µg/ml puromycin (Biosharp Life Sciences) for 14 days and maintained with 0.625 µg/ml puromycin for 7 days for use for subsequent experiments. Cells transfected with an unrelated sequence (scrambled siRNA) were used as the negative control. The sequence of the negative control siRNA is listed in Table SI.

Cell viability was estimated by Cell Counting Kit-8 (CCK-8) assay kit (Beijing Solarbio Science & Technology Co., Ltd.), according to the manufacturer's instructions. Briefly, cells from the control and shPTX3 groups were seeded in 96-well plates and cultured for different time periods (0, 1, 2, 3, 4 or 5 days). Aliquots (15 µl) of CCK-8 were subsequently added to each well and incubated at 37˚C for 2 h. Finally, the absorbance was measured at 450 nm.

Briefly, cells were inoculated into a 6-well plate, and after the cells had reached 80-90% confluency, the cells were starved for 12 h and scratched evenly at the bottom of the orifice plate with a 10 µl tip. The distances of cell migration calculated by changes in the widths of the scratches were recorded at different time points (0, 3, 6, 9 or 12 h). A total of three measurements at different positions along each scratch-line were recorded using an Olympus BX53 light microscope (Olympus Corporation).

Briefly, the chemoattractant (EGF; PeproTech, Inc.) was loaded into the lower chamber, and cells (5x10⁵/ml) suspended in binding medium (DMEM containing 0.1% bovine serum albumin and 25 mM Hepes) were subsequently loaded into the upper chamber (Neuro Probe Inc.). The two chambers were segregated with a 10 µm membrane (Neuro Probe Inc.) that had been pretreated with 10 µg/ml fibronectin (MilliporeSigma) at 4˚C overnight. Subsequently, the chambers were incubated in an atmosphere containing 5% CO₂ at 37˚C for 3 h. The filter membrane was then washed, and the attached cells were fixed and stained with trypan blue for 10 min at room temperature (RT). The number of migrating cells was counted in three separate fields under an Olympus BX53 microscope, and the chemotaxis index was calculated according to the following formula: Number of migrating cells in a chemoattractant gradient/the number of migrating cells in the control group.

After having starved the cells with serum-free medium containing 50 ng/ml EGF, the cells were first plated in 35 mm dishes containing a glass coverslip that had been coated with 10 µg/ml fibronectin overnight at 4˚C, and then air-dried. Cells were allowed to attach to the coverslips for time periods of 5, 15 or 30 min, and subsequently were washed gently twice with cold PBS, prior to being fixed with 4% paraformaldehyde for 10 min at 4˚C. Finally, all cells attached to each coverslip were counted with a hemocytometer under a light microscope with five random fields (magnification, x200).

The upper surfaces of Transwell chambers containing polycarbonate membranes with 8-µm pores were coated with 20 µl diluted Matrigel™ (BD Biosciences) for 1 h at 37˚C. A total of lx10⁵ cells were starved in serum-free medium for 3 h and subsequently added to the upper chamber, whereas the lower chamber was filled with the same medium as in the upper chamber, but with 10% FBS. After an incubation for 20 h at 37˚C, the cells remaining on the upper layer of the chambers were wiped off, whereas the cells attached to the submembrane surface were fixed with neutral formaldehyde for 10 min at 4˚C and stained with Giemsa for 10 min at 4˚C. Finally, the average numbers of invaded cells per field of view were obtained from five random fields under a light microscope.

Briefly, cells were planted in 35 mm dishes overnight, pretreated with serum-free medium for 3 h and subsequently stimulated by adding 50 ng/ml EGF at 37˚C for time periods of 0, 15 sec, 30 sec, 1, 2 or 5 min. Thereafter, cells were fixed, permeabilized and stained in the dark with Molecular Probes™ Oregon Green 568 phalloidin (Thermo Fisher Scientific, Inc.) diluted in F-buffer (10 mM Hepes containing 20 mM KH₂PO₄, 5 mM EGTA and 2 mM MgCl₂ in Dulbecco's PBSl pH 6.8) at RT for 1 h. The bound phalloidin was extracted with methanol for 1 h, and the F-actin content was subsequently measured with a fluorescence reader at an excitation wavelength of 578 nm and an emission wavelength of 600 nm. The results were expressed as the relative F-actin content, where F-actin^Δt/F-actin0=(fluorescence^Δt mg/ml)/(fluorescence 0/mg/ml).

Cells were starved with serum-free medium for 3 h and stimulated by 50 ng/ml EGF for various time periods. The reactions were terminated and the cells washed twice with cold PBS, before subsequently being lysed in 1X SDS lysis buffer [Tris-HCl; pH 6.8; containing 62.5 mM (2%) SDS and 10% glycerol]. The protein determination method was via bicinchoninic acid assay. Equal amounts of protein (20 µg/lane) were separated by SDS-PAGE (10% gels) and blotted onto PVDF membranes (MilliporeSigma). The membranes were blocked with 5% non-fat milk for 1 h at RT and incubated overnight at 4˚C with the primary antibodies raised against β-actin (cat. no. 4967S; 1:3,000; Cell Signaling Technology), PTX3 (cat. no. ab190838; 1:3,000; Abcam), PKCζ (cat. no. 9372S; 1:3,000; Cell Signaling Technology), Akt (cat. no. 9272S; 1:3,000; Cell Signaling Technology), integrin β1 (cat. no. 4706S; 1:3,000; Cell Signaling Technology), cofilin (cat. no. 5175S; 1:3,000; Cell Signaling Technology) and their phosphorylated proteins (Cell Signaling Technology), followed by an incubation at RT with HRP-conjugated secondary antibody (cat. no. 7074S; 1:1,000; Cell Signaling Technology) for 1 h. Finally, the western blots were visualized using enhanced chemiluminescence reagents (Pierce; Thermo Fisher Scientific, Inc.) and analyzed with ImageJ 1.53a (National Institutes of Health).

After washing the MDA-MB-231 cells three times with cold PBS, the cells were lysed in 1 ml ice-cold cell lysis buffer (40 mM Hepes containing 120 mM NaCl, 1% Triton X-100, 10 mM pyrophosphate, 10 mM glycerophosphate, 50 mM NaF, 1.5 mM Na₃VO₄, 1 mM EDTA, 1X cocktail; pH 7.5) on ice for 30 min. Following centrifugation at 12,000 x g for 15 min at 4˚C, the supernatants were removed and precleared through the addition of 50 µl Invitrogen^® protein A (Thermo Fisher Scientific, Inc.) for 1 h. Supernatants (50 µl) were immunoprecipitated using anti-PKCζ antibody (cat. no. ab59364; 1:3,000; Abcam). Finally, the immunocomplexes were captured by adding 50 ml protein A, and the immunocomplexes were then subjected to western blotting analysis, which was performed as aforementioned.

Briefly, 293T cells were transfected with exogenous PKCζ and lysed with 1 ml ice cold cell lysis buffer (40 mM Hepes containing 120 mM NaCl, 1% Triton X 100, 10 mM pyrophosphate, 10 mM glycerophosphate, 50 mM NaF, 1.5 mM Na3VO4, 1 mM EDTA, 1X cocktail; pH 7.5) after 24 h. The supernatants were removed after centrifugation at 12,000 x g for 15 min at 4˚C and the PKCζ antibody was subsequently added to pull down the PKCζ-bound proteins. The protein determination method was use of the bicinchoninic acid assay, and 10% SDS gel electrophoresis was performed. Gels stained with Coomassie brilliant blue were decolorized with glacial acetic acid for 1 h at RT. The bands on the gel were then visualized and finally, proteins interacting with PKCζ were identified by LTQ-orbitrap XL MS (Thermo Fisher Scientific, Inc.). The LTQ Orbitrap XL MS was set for continuous monitoring of positive ions, and data were collected over 15 min in centroid mode over the mass range 50-1,000 m/z. The nitrogen gas temperature was set 350˚C, and cone voltage was 150 V, 30 arb of flow rate.

Briefly, MDA-MB-231 cells were plated in 12-well plates containing sterile coverslips and incubated for 24 h at 37˚C. Cells were subsequently fixed with 4% paraformaldehyde, quenched with 50 mM NH₄Cl, permeabilized with 0.2% Triton X-100 in PBS and blocked with 3% bovine serum albumin (AbMole Bioscience, Inc.). Cells were incubated with anti-PTX3 (cat. no. ab190838; 1:3,000; Abcam) and anti-PKCζ (cat. no. AB59364; 1:3,000; Abcam) antibodies overnight and stained with Invitrogen™ Alexa-Fluor 488 (cat. no. S11223; 1:3,000) and 594 (cat. no. S11227; 1:3,000) conjugated secondary antibodies for 1 h at 4˚C (Thermo Fisher Scientific, Inc.). Finally, coverslips were mounted and visualized using a confocal microscope (Nikon A1R; x400 magnification).

Paraffin tissue sections (5 µm in thickness) were firstly fixed with 4% paraformaldehyde for 12 h at 4˚C, then deparaffinized, rehydrated through graded alcohol solution and subsequently boiled (100±5˚C) in 10 mM sodium citrate for 3 min in an autoclave, followed by subsequent cooling to RT for antigen retrieval. After that, the tissues were treated with 3% hydrogen peroxide and blocked with 2% BSA for non-specific antigenic sites for 1 h at RT. The sections were subsequently incubated with the primary antibody against PTX3 (cat. no. ab190838; 1:500; Abcam) overnight at 4˚C. After washing three times with PBS, sections were incubated with the peroxidase-labeled secondary antibody (cat. no. TA373083L; OriGene Technologies, Inc.) for 30 min at RT. The reactions were visualized using an HRP DAB Detection Kit (OriGene Technologies, Inc.), with subsequent counterstaining with hematoxylin for 5 min at RT. The images of stained sections were captured under an optical microscope, and subsequently evaluated by three experienced pathologists. The results of staining were scored as follows: 0 points, unstained; 1 point, light brown; 2 points, brown; and 3 points, dark brown. Scores ≥1 point were considered to represent positive staining.

All animal experiments were performed in accordance with the Guidelines for the Care and Use of Laboratory Animals of Nankai University (Tianjin, China; approval no. 2023-SYDWLL-000545). Briefly, 4-6-week-old female severe combined immunodeficient (SCID) mice (weight, 15-20 g) from Wei Tong Li Hua Experimental Animal Co., Ltd. were randomly divided into two groups (n=10) with one group acting as the control group (transfected with pLKO.1-scrambled siRNA-luc), and another acting as the PTX3 knockdown group (transfected with pLKO.1-shPTX3-luc). The mice were bred and housed singly using a 12:12 reverse light-dark cycle in the SPF level Animal Experimental Laboratory of Nankai University, with the temperature of 22˚C and 50% atmospheric humidity. Water and food were available ad libitum. Luciferase labelling of control cells and PTX3 knockdown of MDA-MB-231 cells were first performed (giving rise to MDA-MB-231-pLKO.1-scrambled siRNA-luc and MDA-MB-231-pLKO.1-shPTX3-luc cells, respectively), and 3x10⁶ cells were subsequently injected into the mammary fat pads of the SCID mice. The body weight of the mice was measured individually every 3 days, and the survival times were monitored. After 8 weeks, mice were intraperitoneally injected with luciferin (Amresco, Inc.) and live IVIS imaging was performed to detect the bioluminescence intensity of the tumors. The mice were sacrificed by cervical dislocation, after which the tumors were excised, and tumor volumes were measured. To examine spontaneous metastasis, the lungs were fixed with formalin for 12 h at 4˚C, embedded in paraffin, then cut into 4- to 6-µm thick sections and stained with H&E stain for 5 min at RT. Finally, the metastatic tumor nodules were counted under a Zeiss inverted microscope.

SPSS version 20.0 software (IBM Corp.) was used for data analysis. Semi-quantification of the western blotting data was performed using ImageJ software (version 1.8; National Institutes of Health). Unpaired student's t-test and one-way ANOVA were used to determine statistical significance for comparisons of 2 and ≥2 groups, respectively. Data are shown as the mean ± SD and P<0.05 was considered to indicate a statistically significant difference. Tukey's test was applied for post-hoc comparisons.

First, the expression level of PTX3 was examined in 60 cases of invasive BC tissues (including invasive ductal carcinoma and invasive lobular carcinoma) and 22 cases of benign breast tissues by immunohistochemical analysis in order to assess the importance of PTX3 in BC. The results obtained showed that positive staining of PTX3 was detected in 52 tumor samples, but only in two benign breast tissue samples (Table I), suggesting that the expression level of PTX3 is increased during tumorigenesis (Fig. 1). It is noteworthy that PKCζ is a cancer-promoting factor that is also highly expressed in BC tissues. In addition, the statistical analysis revealed that PTX3 expression was detected in 51 of 52 lymph node metastasis cases (P=0.003; Table II). However, only one tumor tissue was identified as being positive in eight cases of non-lymph node metastasis. Taken together, these results strongly suggested that upregulation of PTX3 is associated with lymph node metastasis of BC.

The initial assumption was made that PTX3 fulfills an important role in the proliferation and migration of BC cells. First, the expression levels of PTX3 in the human normal mammary epithelial cell line (MCF-10A) and two BC cell lines (MDA-MB-231 and MCF7) were evaluated by western blotting assay. The results obtained showed that the expression levels of PTX3 in the BC MDA-MB-231 and MCF7 cell lines were higher compared with that in MCF-10A cells (Fig. 2A), and this difference may have been associated with the malignant behavior of BC cells. Subsequently, stable PTX3 downregulation of MDA-MB-231 and MCF7 cells was performed using lentiviral packaging vectors, and the respectively transformed cells were used for the following experiments (Fig. 2B). Cell viability assay revealed that the proliferation rate of the shPTX3 cells was decreased compared with control BC cells (Fig. 2C and D). Subsequently, the migratory capability of the BC cells was examined using a wound healing assay, and changes in the scratch widths were recorded at different times. As shown in Fig. 2E and F, shPTX3-transformed cells exhibited much lower directional migratory rates compared with the control BC cells, suggesting that knockdown of PTX3 led to a marked inhibition of the migratory rate of BC cells. In addition, western blotting analysis showed a blockade at EGF-induced Akt phosphorylation at Ser473, indicating that the function of Akt was inhibited in shPTX3 cells (Fig. 2G). Taken together, the aforementioned findings revealed that knockdown of PTX3 inhibited the proliferative and migratory capabilities of the BC cells through the suppression of p-Akt expression.

Subsequently, the chemotaxis of BC cells was investigated to evaluate whether PTX3 exerts a pivotal role in the metastasis of BC cells, and the results of these experiments are shown in Fig. 3A. EGF-induced chemotaxis of MDA-MB-231 and MCF7 cells was found to be markedly inhibited following PTX3 knockdown compared with the control groups (Fig. S1). As shown in Fig. 3B and D, both EGF-induced cell attachment and actin polymerization were decreased to a statistically significant extent (P<0.05) in shPTX3 cells compared with the control BC cells. Furthermore, PTX3 downregulation also led to a marked attenuation of the invasive capability of MDA-MB-231 and MCF7 cells compared with control cells (Fig. 3C). As shown in Fig. 3E, EGF-induced phosphorylation of both cofilin and integrin β1 was severely impaired in shPTX3 cells. Taken together, these findings suggested that downregulation of PTX3 inhibits EGF-induced BC cell metastasis through inhibiting cofilin and integrin β1, which are two downstream effectors of PKCζ.

The results from the MS experiments first identified PTX3 as a PKCζ-interacting protein (Fig. 4A). Therefore, it was surmised that PTX3 could regulate EGF-induced BC cell chemotaxis and metastasis through regulating the activation of PKCζ. As shown in Fig. 4B, co-IP assay confirmed that there was an interaction between PTX3 and PKCζ in MDA-MB-231 cells. Confocal microscopy analysis further revealed that both PTX3 and PKCζ were distributed in the cytosol of resting cells. Upon stimulation with 50 ng/ml EGF, however, both PTX3 and PKCζ were co-translocated to the cell membrane, suggesting the activation of proteins (Fig. 4C). The translocation of PKCζ was also analyzed in experiments that involved PTX3 knockdown of the MDA-MB-231 cells. Compared with the Scr cells, EGF-induced translocation of PKCζ was markedly inhibited in shPTX3 cells (Fig. 4D). Furthermore, according to the findings of western blotting analysis experiments, the phosphorylation of PKCζ was also reduced in shPTX3 cells upon knocking down PTX3 (Fig. 4E). Taken together, these results suggested that PTX3 co-localizes with PKCζ and regulates its activation.

Finally, the present study aimed to explore the metastatic capabilities of shPTX3 cells in a BC xenograft model. shPTX3 and control cells were implanted into the mammary fat pads of SCID mice. After 8 weeks, the tumors were measured, and the maximum tumor size was found to be ~990 mm³. The tumor growth in mice with transplanted shPTX3 cells was found to be markedly inhibited compared with the control group, suggesting that PTX3 exerts an important role in tumor growth in vivo (Fig. 5A). Subsequently, lung metastasis of the human BC cells was evaluated using H&E staining (Fig. 5B). A clear and dramatic decrease in the number of pulmonary metastatic nodules in shPTX3-implanted mice was detected compared with the control groups (Fig. 5C). Moreover, the mice with implanted shPTX3 cells also exhibited enhanced survival rates (Fig. 5D) compared with the control mice. Therefore, these results demonstrated that PTX3 is required for the metastasis of human BCs in vivo, and PTX3 knockdown led to an inhibition of the growth and metastasis of breast tumors.