An ascorbic acid-decorated nanostructured surface on titanium inhibits breast cancer development and promotes osteogenesis

The chest wall is the most frequent metastatic site of breast cancer (BC) and the metastasis usually occurs in a solitary setting. Chest wall resection is a way to treat solitary BC metastasis, but intraoperative bone defects and local tumor recurrence still affect the life quality of patients. Titanium-based prostheses are widely used for chest wall repair and reconstruction, but their inherent bio-inertness makes their clinical performance unfavorable. Nanostructured surfaces can give titanium substrates the ability to excellently modulate a variety of cellular functions. Ascorbic acid is a potential stimulator of tumor suppression and osteogenic differentiation. An ascorbic acid-decorated nanostructured titanium surface was prepared through alkali treatment and spin-coating technique and its effects on the biological responses of BC cells and osteoblasts were assessed. The results exhibited that the nanorod structure and ascorbic acid synergistically inhibited the proliferation, spreading, and migration of BC cells. Additionally, the ascorbic acid-decorated nanostructured surface significantly promoted the proliferation and osteogenic differentiation of osteoblasts. This work may provide valuable references for the clinical application of titanium materials in chest wall reconstruction after the resection of metastatic BC.


Introduction
Breast cancer (BC) metastases most frequently to the chest wall, especially in the advanced stage and local recurrence of BC, but bone metastasis typically takes place in an isolated setting.Although there is no curative treatment for metastatic BC, in BC patients with isolated metastasis, chest wall resection (CWR) is an effective way to prevent further BC metastasis and relieve pain and other symptoms, thereby improving the life quality of the patients [1].However, CWR can leave large chest wall defects, which expose major pleural structures and pericardial surfaces and affect normal cardiopulmonary function.Therefore, it is routine to implant a prosthesis for post-CWR chest wall reconstruction.Due to excellent biocompatibility and mechanical properties, lightweight, and low radiological interference, titanium-based materials are widely applied for chest wall repair and reconstruction [2,3].Whereas, currently applied titanium-based prostheses have mainly demonstrated their mechanical properties and support performances, without considering their anti-tumor and osteogenesis-enabling properties, especially the efficacy in inhibiting tumor recurrence.Although chest wall reconstruction surgery may be followed by a combination of local radiotherapy to clear residual cancer cells and prevent BC recurrence [3,4], endowing prostheses with their anti-tumor capabilities is a promising strategy.Because of the intrinsic bio-inertness, titanium-based materials cannot perform satisfactory biological functions.Fortunately, the functionality of titanium-based materials can be improved by a practicable surface modification.
A growing number of studies emphasize the importance of the synergistic effects of surface structure and chemical modification to enhance the biological function of titanium-based materials.Micron/nano-scale pore/tube structures can be constructed on titanium implants by 3D printing and anodizing technology, which can greatly promote osteogenesis and osseointegration and facilitate bone trauma repair [5][6][7][8].The multi-scale pore/tube structures provide platform for the loading and release of anticancer elements/drugs such as doxorubicin, apoptosis-inducing ligand (Apo2L/TRAIL), paclitaxel, and selenium, enabling local treatment to prevent tumor recurrence.Rare-earth elements/drugs/bioactive molecules-decorated nanostructured titanium surfaces can significantly promote osteogenesis in combination with light-triggered therapy and scavenge bacteria and osteosarcoma cells by released ions/drugs and light-excited generation of reactive oxygen species (ROS) and nitric oxide [9][10][11].In addition, modification of the active components of the bone matrix to enhance the ability of titanium implants to promote osteogenesis is receiving increasing attention [12,13].As a simple and effective surface treatment technique, alkali treatment has been extensively applied to the surface modification of titanium-based materials, which can engrave nanoscale structures on titanium surfaces similar to bone matrix [14,15].Alkali-treated nanostructured surfaces can tremendously energize the osteogenic potential of osteoblastic-lineage cells.In addition, the alkali-treated nanostructured surfaces also effectively modulate the immune response and angiogenesis, making it more conducive to new bone production [13,15].The favorable effects of alkali-treated nanostructured titanium surfaces on the functions of multifarious cells were highlighted by the promotion of cell adhesion and spreading.
Ascorbic acid (Vitamin C) is an essential dietary nutrient that cannot be synthesized by humans and other primates and must be obtained from diet [16].Ascorbic acid exhibits strong reducibility and participates in complex metabolic processes in the body, such as growth promotion and immunity enhancement.In addition, ascorbic acid has been proposed as a promising anti-cancer agent when combined with radiotherapy and chemotherapy [17][18][19].Ascorbic acid can selectively induce oxidative stress in tumor cells, impede their glucose metabolism, and modulate the expression of hypoxia-inducible factor, thus leading to energy deficiency and cell death of cancer cells and inhibiting tumor development [20][21][22].Moreover, Magrì et al revealed that high-dose ascorbic acid could mediate anti-tumor immune responses and cooperate with immune checkpoint therapy (ICT) in several cancer types to suppress tumor development [17].The combination of ascorbic acid and anti-tumor agents not only improves the therapeutic efficacy against the primary tumor, but also overcomes the major limitations of conventional therapy, such as drug resistance, post-surgery recurrence, and tumor metastasis [23].Ascorbic acid is an important supplement to induce osteogenic differentiation in the culture of multiple osteoblasticlineage cells [10,13,24,25].Potent stimulation by ascorbic acid elicits collagen synthesis and secretion by osteoblasts, allowing extracellular matrix (ECM) assembly and promoting focal adhesion of osteoblasts [26,27].Focal adhesion initiates its mediated intracellular signal transduction, leading to the expression of osteogenic-specific genes and promoting osteogenic differentiation of the cells [26,28].As an effective osteogenesis-inducing stimulus, ascorbic acid is widely used to modify the implantable materials to improve their-mediated osteogenic differentiation [29][30][31][32].
In this work, alkali treatment was used to construct a nanostructured surface on the pure titanium substrate, followed by spin-coating processes to graft ascorbic acid onto the nanostructured surface.The effects of the surfaces on the biological responses of BC cells and osteoblasts were subsequently investigated.The ascorbic acid-decorated nanostructured titanium surface significantly inhibited the vitality and migration of BC cells and promoted osteogenesis of osteoblasts.This work emphasizes the synergistic roles of the nanostructure and ascorbic acid in regulating BC development and osteogenesis, highlighting in particular the opposed regulatory effects of ascorbic acid on BC cells and osteoblasts.The work is expected to provide an example of the design of titanium prostheses for chest wall reconstruction.

Sample fabrication and characterization
Pure titanium sheets (99.6% purity, Φ10 × 3 mm) were cleaned sequentially in acetone, anhydrous ethanol, and deionized water for 10 min by ultrasonication and the samples were named Ti.The Ti samples were submerged in 5 M NaOH (Sinopharm Group, China) solution at 60 • C for 48 h and then cleaned ultrasonically in deionized water for 5 min.The alkali-treated samples were named TiN.To prepare the ascorbic acid-decorated titanium samples, 100 µl of ascorbic acid solution (5 mg ml −1 , Solarbio, China) was dropped onto the TiN samples, and the adsorption was allowed to take place for 10 s with an acceleration of 500 rpm on a spin coater (WS-650HZB-23B, Laurell Technologies, America).Afterwards, the samples were washed with deionized water and dried at 37 • C, and the spin-coating process was repeated three times.The alkali-treated samples with ascorbic acid decoration were named TiNA.
The surface morphology of the samples was observed using field emission scanning electron microscopy (FE-SEM, GeminiSEM 300, Zeiss, Germany) at 10 kV.The water contact angle of the sample surfaces was detected by a contact angle analyzer (JC2000C1, Powereach, China).The three-dimensional (3D) structure and roughness of the sample surfaces were evaluated using atomic force microscopy (AFM, Multimode-8HR03040111, Bruker, Germany) in a 10 × 10 µm area.An x-ray energy-dispersive spectrometry (EDS, X-Max 80, Oxford, UK) was used to measure the elemental composition and distribution of the sample surfaces.A diffuse reflectance Fourier transform infrared spectroscopy (DR-FTIR, EQUINOX 55, Bruker, Germany) was applied to detect the chemical groups on the sample surfaces.
An ascorbic acid assay kit (Solarbio, China) was used to detect the ascorbic acid immobilized on titanium surfaces and released into the environment.The TiNA samples were immersed in 1 ml of citric acid solution (0.025 mM, Sinopharm Group, China) to extract the immobilized ascorbic acid.Afterwards, the content of ascorbic acid in the resulting solution was detected according to the kit instructions.Briefly, the solution was mixed with the reaction reagents according to the operating guides and the absorbance was measured at 265 nm using an ultraviolet spectrophotometer (Thermo Fisher Scientific, America).The concentration of ascorbic acid in the solution was calculated by normalization to the standard concentration and the amount of ascorbic acid immobilized on the TiNA surface was converted to a content per unit area (nmol cm −2 ).For evaluating the releasing performance of ascorbic acid, the TiNA samples were immersed in 1 ml of simulated body fluid (SBF, pH = 7.4) at 37 • C for 2, 4, 6, 8, 10, 12, and 14 d.The concentration of ascorbic acid released into SBF at each time point was determined following the method described above.The percentage of ascorbic acid remaining on the sample surface was calculated to characterize its stability.

The influence of the samples on BC cells 2.2.1. BC cell culture
Human BC cells (hBCCs, MDA-MB-231) were provided by the Chinese Academy of Sciences (CAS) Cell Bank (Shanghai, China).The hBCCs were cultured in Leibovitz's L-15 medium (Gibco, America) supplemented with 10% fetal bovine serum (Gibco, America) and 1% penicillin-streptomycin (Solarbio, China) at 37 • C in a humidified incubator with 100% air.The hBCCs were inoculated on the samples at a density of 2 × 10 4 cells cm −2 in the subsequent experiments unless otherwise specified.

Viability, cell proliferation, and cytoskeleton of BC cells
A Live/Dead staining kit (Invitrogen, America) was used to assess the hBCC vitality according to the specifications of the manufacturer.After cultured on the samples for 1 d, the hBCCs were stained with 50 µl of working solution for 30 min.The fluorescent images were taken by a confocal laser scanning microscope (CLSM, C2 Plus, Nikon, Japan), while living cells were labeled with calcein-AM (green fluorescence) and dead cells were labeled with ethidium homodimer-1 (red fluorescence).The mean fluorescence intensity of the calcein-AM labeled cells was analyzed by ImageJ software to quantificationally characterize the cell viability.Thiazolyl blue tetrazolium bromide (MTT, 5 mg ml −1 , Sigma, America) was used to evaluate the hBCC proliferation after cultured on the samples for 1 and 3 d.Briefly, the cells were incubated in 1 ml of working solution (medium/MTT solution 9:1 (v/v)) for 4 h and 1 ml of dimethylsulfoxide (Sinopharm Group, China) was used to dissolve the generated MTT formazan.Absorbance (at 492 nm) was measured on a microplate reader (Infinite F50, TECAN, Switzerland).After the hBCCs were cultured on the samples for 1 d, the cytoskeleton was counterstained with Phalloidin/DAPI (Sigma, America) according to the reagent instructions and imaged by the CLSM.Detailed experimental procedures were introduced in the previous studies [10,11].

Migration capability of BC cells
The sample-mediated infiltration and migration capabilities of the hBCCs were evaluated by the woundhealing model.Briefly, the supernatant from the cells cultured on the samples for 2 d was collected and used to stimulate the cells inoculated in the 24-well plates, and the normally cultured hBCCs were set as Control.Subsequently, equal-width cell scratches were created using a 200 µl pipette and the cells were rinsed with phosphate buffer saline (PBS, Adamas, China) to eliminate cell debris.The medium was substituted with the serum-free Leibovitz's L-15 medium.Images of cell scratches were captured using an optical microscope (ECLIPSE Ts2, Nikon, Japan) at 0, 12, and 24 h after scratch formation.The widths of the cell scratches were measured using ImageJ software and the cell migration distance was calculated using the following formula: Migration distance = L n − L 0 (µm), where L 0 represented the initial scratch width (t = 0 h) and L n represented the residual width at the metering time point (t = n h, n = 12, 24).

Intracellular oxidative stress levels of BC cells
The ROS level of the hBCCs was determined by a Reactive Oxygen Species Assay Kit (Beyotime, China) following the manufacturer's instructions.In brief, the hBCCs were inoculated on the samples and cultured for 2 d.Then, the cells were rinsed with PBS and incubated with 500 µl of working solution for 20 min.Afterwards, nuclei were counterstained with Hoechst 33342 (Beyotime, China) for 5 min and then the fluorescent images were captured by the CLSM.The mean fluorescence intensity of ROS was quantized by ImageJ software.

The influence of the samples on osteoblasts 2.3.1. Osteoblast culture
Osteoblasts (MC3T3-E1 Subclone 14) were provided by CAS Cell Bank and cultured with α-MEM (Gibco, America) containing 10% fetal bovine serum and 1% penicillin-streptomycin at 37 • C in a humidified incubator with 5% CO 2 .In subsequent experiments, the osteoblasts were inoculated on the samples at a density of 2 × 10 4 cells cm −2 unless otherwise indicated.

Viability, cell proliferation, and cytoskeleton of osteoblasts
According to the methods presented in section 2.2.2, the vitality of the osteoblasts was assessed by fluorescent staining after cultured on the samples for 2 d.The cell proliferation of osteoblasts was tested by MTT assay after cultured on the samples for 2 and 4 d.The cytoskeleton of the osteoblasts cultured on samples for 2 d was imaged according to the above methods.

Gene expression of osteoblasts
Gene expression of osteoblasts was detected by realtime quantitative reverse transcriptase polymerase chain reaction (qRT-PCR).Osteoblasts were inoculated on the samples and normally cultured for 3 d.Afterwards, the medium was changed to osteogenic differentiation induction medium (oriCell, China) and the cells were cultured for another 7 d.According to the manufacturer's protocols, total RNA was extracted from the osteoblasts cultured on the samples using MonPure™ Universal RNA Kit (Monad, China).Subsequently, complementary DNA was synthesized from total RNA using MonScript™ RTIII All-in-One Mix with dsDNase (Monad, China).The expression levels of runtrelated transcription factor 2 (Runx2), osterix (OSX), alkaline phosphatase (ALP), type I collagen (COL1), osteopontin (OPN), osteocalcin (OCN), and bone morphogenetic protein-2 (BMP2) were analyzed by a LightCycler ® 96 Real-Time PCR system (Roche, Switzerland) using MonAmp™ SYBR® Green qPCR Mix (Monad, China).Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as a housekeeping gene to normalize the gene expression levels.The results were presented as relative fold change and quantified by the comparative CT (∆∆CT) method.The sequences of corresponding primers are listed in table S1 of supplementary information.

ALP activity of osteoblasts
ALP activity of the osteoblasts cultured on the samples was detected with BCIP/NBT Alkaline Phosphatase Color Development Kit (Beyotime, China) and Alkaline Phosphatase Assay Kit (Beyotime, China).Briefly, the osteoblasts were normally cultured for 3 d and then cultured with osteogenic differentiation induction medium for another 3 and 7 d.The cells were fixed by 4% paraformaldehyde for 30 min and stained with BCIP/NBT working solution for 30 min, followed by imaged on a stereo microscope (SMZ745, Nikon, Japan).In addition, cell lysis buffer (Beyotime, China) was used to lyse the osteoblasts, and the quantitative detection of ALP activity in cell lysis product was conducted using an Alkaline Phosphatase Assay Kit following the product instructions.

Collagen secretion of osteoblasts
Collagen secretion of the osteoblasts growing on the samples was evaluated by Picrosirius Red Staining Solution (Phygene, China).After the osteoblasts were cultured on the samples for 3 d and induced into osteogenic differentiation for 7 and 14 d, the cells were fixed by 4% paraformaldehyde for 30 min and incubated with the staining solution for 18 h.The excess dyes were removed through adequate rinse with 0.1 M acetic acid.Qualitative images were photographed using the stereo microscope.Subsequently, 1 ml of decolorization solution (0.2 M NaOH/methanol 1:1 (v/v)) was applied to dissolve the dyed cells, and absorbance (at 570 nm) was quantified on the microplate reader.

ECM mineralization of osteoblasts
The degree of ECM mineralization was assessed with Alizarin Red S Staining Solution (2%, pH 4.2, Beyotime, China).The osteoblasts were cultured on the samples for 3 d and subsequently cultured with osteogenic differentiation induction for 14 and 21 d.Afterwards, the cells were fixed with 75% ethanol for 1 h at each time point and stained with the working solution for 30 min, followed by adequately rinsing with deionized water to eliminate the excess dyes.Optical images of the dyed samples were photographed with the stereo microscope.For quantitative analysis of ECM mineralization degree, 500 µl of 10% cetylpyridinium chloride (Sinopharm Group, China) was used to dissolve the dyes, and absorbance (at 570 nm) was measured on the microplate reader.

Statistical analysis
Three repetitions were conducted in each experiment and the data were shown as mean ± standard deviation.Statistical analysis was implemented with one-way ANOVA followed by the Student-Newman-Keuls test using the SPSS 14.0 software.p < 0.05 was deemed the differences were statistical, p < 0.01 was deemed the differences were significant, and R Li et al p < 0.001 was deemed the differences were highly significant.

Physicochemical characterization of the sample surfaces
The SEM images show the relatively flat surface of the untreated titanium sample (Ti) and the nanorodstructured surface of the alkali-treated sample (TiN) (figure 1(A)).Furthermore, nanoscale morphology uniformly distributed on the surface of TiN and TiNA.In high-magnification images, the nanorods on the TiNA surface were thicker than TiN, probably due to the nanostructures being wrapped in ascorbic acid.The contact angle of the samples decreased sharply after the alkali treatment, indicating an increase in the surface hydrophilicity (figure 1(B)).The contact angle of the TiN surface was approximately 3.68 • , which suggested that the alkali-treated titanium substrate possessed high-hydrophilicity.The ascorbic acid-decorated nanostructured surface also exhibited extremely high hydrophilicity with a contact angle of 3.74 • .The introduction of hydroxyl groups (due to alkali treatment and the alcohol groups of ascorbic acid) causes the increased hydrophilicity of the sample surfaces [13].In addition, alkali treatment removed the adsorbed contaminants such as hydrocarbons, which tend to increase the hydrophobicity of the sample surfaces [33].The AFM 3D images show that pure titanium substrate presented a relatively smooth surface with low surface roughness (Ra = 24.87 ± 0.52 nm), as shown in figure 1(C).The surface of the alkali-treated titanium substrate exhibited uniformly distributed nanorodlike structures, which endowed the surface with high roughness (Ra = 46.80 ± 4.55 nm).Decoration of ascorbic acid reduced the roughness of the nanostructured surface, but the TiNA surface still exhibited the nanorod-like 3D structures.
The elemental content of the sample surfaces detected by EDS is shown in figure 2(A).The concentration of O increased while the concentration of Ti and C decreased after the alkali treatment.The C and O elements on the Ti surface might originate from the adsorption of contaminants in the air.The sharp increase in O content and the appearance of Na element were due to the formation of a nanostructured sodium titanate layer on the titanium substrate surface after alkali treatment [13,15,25].After ascorbic acid spin-coating, the increase in C content indicated that ascorbic acid was successfully grafted on the surface of the alkali-treated samples.The change in the number of light dots representing each element in EDS elemental mapping images also demonstrated the above statements (figure 2(B)).DR-FTIR was applied to further certify the successful construction of the ascorbic acid-decorated nanostructured layer and investigate the formation mechanism.The absorption band from 1000 to 500 cm −1 corresponds to the Ti-O bonds and the Ti-OH absorption peak at 940 cm −1 appeared in the TiN spectrum after alkali treatment (figure 2(C)).A broad O-H band (3600-3200 cm −1 ) and the stretching vibration absorption peaks C=O (at 1750 cm −1 ), C=C (at 1650 cm −1 ), C-H (at 1320 and 680 cm −1 ), and C-O (at 1140 and 1038 cm −1 ) in TiNA spectrum indicated the grafting of ascorbic acid on the titanium substrate surface.The absorption peaks at 875, 820, and 750 cm −1 are attributed to Ti-O-C stretching vibration.During the spin-coating process, the alcohol groups in the ascorbic acid combined with the Ti-OH on the alkalitreated titanium substrates to form hydrogen bonds, and then the hydrogen bonds were dehydrated under the heating condition to form Ti-O-C bonds, which allowed ascorbic acid to be grafted stably onto the nanostructured surfaces.
The ascorbic acid immobilized on the surface of nanostructured titanium was distributed at an approximate density of 759.23 nmol cm −2 (figure 3(A)).About 596.29 nmol (105.02µg) of ascorbic acid could be coated on the TiNA sample (disk substrate with a diameter of 10 mm).During the first 2 d of immersing the TiNA sample in SBF, ascorbic acid was released dramatically and approximately 48.11 nmol (8.47 µg) of ascorbic acid was released into the solution (figure 3(B)).The rate of ascorbic acid release slowed down during the period of 2-4 d of immersing the sample in SBF, and the amount of ascorbic acid released into the environment on the fourth day was approximately 54.64 nmol (9.63 µg).From the fourth day onward, the amount of ascorbic acid released from the TiNA sample into the environment no longer increased.After the complete release of free ascorbic acid into the environment, about 91% of ascorbic acid remained on the TiNA surface, due to the covalent immobilization of ascorbic acid onto the alkali-activated titanium surface (figure 3(C)).This result indicated that the ascorbic acid-coated nanostructured surface constructed by the spin-coating process was a stable layer, and the majority of the immobilized ascorbic acid would be not released into the humoral environment and remain grafted onto the titanium substrate surface for a long time.
In summary, alkali treatment activated the bioinert titanium substrate and endowed it with a nanorod-structured surface, which provided structural and chemical anchors for the grafting of bioactive molecules.After spin-coating with ascorbic acid, an ascorbic acid-decorated nanostructured surface was stably constructed on titanium substrate and possessed both nanostructure and ascorbic acid properties.

Biological responses of BC cells to the samples
The samples Ti, TiN, and TiNA did not exhibit cytotoxicity to hBCCs (fluorescent images did not present   red fluorescent labeling of dead cells), but the number of the living hBCCs (green fluorescence-labeled) growing on the samples decreased in sequence (figure 4(A-a)).The mean fluorescence intensity of the calcein-AM labeled cells in the TiNA group was lower than that in the Ti group, indicating that the ascorbic acid-decorated nanostructured surface depressed intracellular esterase activity and viability of hBCCs (figure 4(A-b)).This phenomenon was supported by cell proliferation assay (figure 4(B)).The hBCC proliferation on the TiN and TiNA was significantly lower than that on the Ti, especially the hBCCs cultured on the TiNA exhibited the lowest proliferation rate and the cell numbers grew very little after culturing for 3 d compared with 1 d.The result that the number of hBCCs on the TiN and TiNA was less than Ti was also confirmed by cytoskeleton staining (figure 4(C)).Moreover, the nanorodstructured surfaces (TiN and TiNA) were not conducive to hBCC spreading, as confirmed by a large number of hBCCs with curled-up cell morphology on the nanostructured surfaces on the cytoskeletal fluorescence images.The hBCC migration was significantly inhibited by the cell microenvironment mediated by the nanorod-structured surfaces, as proved by the fewer migratory cells at the cell scratches at 24 h (figure 4(D-a)).Moreover, the TiNA showed a more prominent inhibition effect, which may be caused by the ascorbic acid released into the microenvironment.As shown in figure 4(D-b), the quantitative statistics of cell migration distance indicated that the nanorodstructured surface mediated hBCC microenvironment could significantly suppress hBCC migration, and the ascorbic acid-decorated nanostructured surface enhanced the inhibition effect.The results of intracellular ROS levels are shown in figure 5.The hBCCs cultured on the TiNA surface exhibited higher ROS levels with noticeably strong fluorescent intensity (figure 5(A)).Quantitative analysis demonstrated that TiN-mediated intracellular ROS levels of hBCCs were slightly higher than Ti, while TiNA-mediated ROS levels were significantly higher than the formers (figure 5(B)).
Ascorbic acid can induce oxidative stress in cancer cells and restrain their metabolism, which exhibits the recognized anti-cancer activity [20,21].In cell culture media, ascorbic acid is oxidized to dehydroascorbate (DHA, the oxidized form of ascorbic acid) unless reducing agents are added [34].The hBCCs adhering to the TiNA sample take up the immobilized and released DHA via the GLUT1 glucose transporter.Increased DHA uptake causes oxidative stress and elevated levels of intracellular ROS due to the depletion of glutathione (GSH), thioredoxin, and nicotinamide adenine dinucleotide phosphate (NADPH) during the reduction of DHA to ascorbic acid in the cancer cells [20,21].ROS targets the active-site cysteine of GAPDH, causing it to undergo irreversible oxidation, which leads to loss of GAPDH activity.In addition, ROS can activate PARP to cause nicotinamide adenine dinucleotide (NAD+) exhaustion, which is the substrate for GAPDH-dependent oxidation.Inactivation of GAPDH and depletion of its substrate inhibit fermentative glycolysis of the cancer cells, thus inducing a significant drop in adenosine triphosphate (ATP) level and leading to energy crisis and death of the cancer cells.A study reported that alteration in mitochondrial metabolic homeostasis induced by ascorbic acid can give rise to the disruption of intracellular iron metabolism, which increases cancer cell sensitivity to ascorbic acid and results in selective toxicity of ascorbic acid [22].Disruption of mitochondrial homeostasis gives rise to elevated levels of O 2

.-
and H 2 O 2 , which are capable of perturbing redoxactive iron homeostasis, resulting in oxidative damage to intracellular macromolecules (i.e.DNA, protein, lipids) that chelate redox-active iron.Thus, ascorbic acid exhibits toxicity to the cancer cells.Decoration of the titanium substrate surface by ascorbic acid remarkably increased the level of oxidative stress in hBCCs (figure 5), thereby modulating their mitochondrial metabolism and inhibiting their proliferation (figures 4(A) and (B)).The nanorod structure exerted a slight promoting effect on intracellular ROS production in hBCCs, and its modulation of hBCCs was mainly reflected in the restriction of cell spreading (figure 4(C)), which ultimately inhibited cell proliferation.The combination of nanostructures and ascorbic acid not only directly improved the inhibition of hBCC vitality by titanium prostheses for chest wall repair, but also retarded hBCC migration and invasion in the tissue microenvironment figure 4(D).One study reported that high-dose ascorbic acid could activated CD8 + T cells and modulated infiltration of tumor tissue by cells of the immune system, thereby mediating anti-tumor immune responses and inhibiting cancer growth [17].Moreover, ascorbic acid could collaborate with immune checkpoint blockade in ICT to effectively enhance antitumor efficacy.Consequently, the ascorbic aciddecorated nanostructured titanium-based prostheses used in chest wall repair and reconstruction after CWR in response to BC metastasis can orchestrate multiple cells and systems to arrest BC recurrence and re-metastasis.

Biological responses of osteoblasts to the samples
All samples supported normal adhesion and growth of osteoblasts, as shown by the uniformly distributed and abundant green fluorescence-labeled cells (living cells) on the fluorescent images (figure 6(Aa)).There was no difference in sample-mediated osteoblast viability among the Ti, TiN, and TiNA groups (figure 6(A-b)).In addition, no dead cells (red fluorescence-labeled) were found on the fluorescent images, indicating that the samples were biocompatible to osteoblasts.As shown in figure 6(B), the cell proliferation results suggested that the nanorod structure and ascorbic acid favored osteoblast proliferation.On the fourth day, the number of osteoblasts grown on each sample increased significantly compared with the second day, especially the TiNA sample showed the highest cell proliferation rate.At each time point, the nanostructured surface promoted cell proliferation compared with the pure titanium surface, while the ascorbic acid-decorated nanostructured surface was more favorable for osteoblast proliferation compared with the formers.The cytoskeleton assembly and cell spreading were presented in figure 6(C).All samples induced abundant and satisfactory actin assembly of osteoblasts (figure 6(C-a)).Additionally, the osteoblasts cultured on the TiN and TiNA surfaces stretched out larger cell spreading morphology.Quantitative analysis of the cell spreading area revealed that the nanostructured surfaces were conducive to osteoblast spreading (figure 6(C-b)), which was attributed to the high hydrophilicity and roughness of the nanostructured surfaces after alkali treatment [13,15].The sample-mediated osteogenic capacity of osteoblasts has also been exhaustively investigated.The results of osteogenic-related gene expression showed that the nanorod-structured surface (TiN) upregulated the expression levels of Runx2, OSX, OPN, OCN, and BMP2 (figures 7(A) and S1) compared with Ti.In addition, the ascorbic aciddecorated nanostructured surface (TiNA) significantly upregulated the expression levels of Runx2, OSX, ALP, COL1, OPN, OCN, and BMP2 compared with Ti and TiN.Both Runx2 and OSX are crucial osteoblast-specific transcription regulators that mediate osteogenic-related gene expression [35,36].An essential cytosolic enzyme, ALP, is widely recognized as an early marker of osteogenic differentiation [37].COL1 is an important component of the osteoblastic matrix, and the massive secretion of COL1 by osteoblasts is regarded as a mid-term marker of osteogenic differentiation [38].OPN is an acidic glycoprotein secreted into the bone matrix by osteoblasts, which is involved in the regulation of ECM remodeling and the early and middle stages of osteogenic differentiation [39].OCN participates in the regulation of ECM mineralization, which is regarded as a marker of late osteogenic differentiation [39,40].BMP2 can regulate a variety of biological processes such as cell growth, metabolism, and differentiation, which plays an important regulatory role in osteogenesis [41,42].Collectively, the nanorodstructured surface promoted osteogenic differentiation of osteoblasts at the gene level.In addition, the introduction of ascorbic acid significantly enhanced the expression of osteogenic-related genes, which facilitated osteogenic differentiation.We further investigated the effects of the samples on osteogenic differentiation of osteoblasts at the protein level and biological processes.As shown in figure 7(Ba), blue-purple ALP-stained crystals appeared on all samples, indicating that the samples did not inhibit ALP activity in osteoblasts.The ALP-stained results were more pronounced on the TiNA sample, suggesting that TiNA mediated higher levels of ALP activity.Quantitative assays of ALP activity showed no significant difference in sample-mediated ALP activity among the three groups after 3 d of osteogenic differentiation induction, but ALP activity in osteoblasts cultured on TiNA was significantly higher than that on Ti and TiN after 7 d of osteogenic differentiation induction (figure 7(B-b)).Sirius red staining images of collagen secreted by osteoblasts are shown in figure 7(C-a).The amount of collagen secreted by osteoblasts increased with the prolongation of osteogenic differentiation induction, while the TiNA sample induced osteoblasts to secrete more collagen.The above statement could be supported by quantitative results (figure 7(C-b)).It is noteworthy that the nanostructured surface mediated more collagen secretion than the pure titanium surface after 14 d of osteogenic differentiation induction.The collagen network largely determines the degree of ECM mineralization, and it provides nucleation nodes for ECM mineralization [43].Assessment of ECM mineralization was consistent with trends in collagen secretion (figure 7(D)).The TiNA mediated the most abundant mineralized nodule formation and the degree of ECM mineralization mediated by the TiN sample was higher than Ti after 21 d of osteogenic differentiation induction.
The nanostructured surface observably promoted osteoblast spreading (figure 6(C)), which was the main reason for its enhancement of osteoblast proliferation and osteogenic differentiation.Numerous studies have shown that the hydrophilic nanostructured surfaces of titanium substrates after alkali treatment promote the adhesion, spreading, proliferation, and osteogenic differentiation of osteoblasticlineage cells [13,15,25].Cytoskeleton stretching and focal adhesion formation mediate the transmission of extracellular mechanical and chemical signals into the cells, thereby activating multiple signaling pathways and transducing signals into the nucleus to regulate cell proliferation and differentiation [44,45].Ascorbic acid acts as an essential cofactor to hydroxylate proline and lysine in pro-collagen, allowing collagen chains to form a helical structure [24].Furthermore, the efficacy of ascorbic acid contributes to the secretion of COL1 into the ECM, constructing an ECM network suitable for osteoblast binding and adhesion [27].The integrin α2β1 on the osteoblast membranes binds to the ECM, thus causing the cells to form focal adhesions, which activate the focal adhesion kinase (FAK)-mediated mitogen-activated protein kinase (MAPK) signaling cascades and transmit the signal to the nucleus [26,27].Activation of the MAPK signaling pathway leads to the accumulation of phosphorylated extracellular signal-regulated kinase (p-ERK) in the nucleus, which allows the activation of the transcription factor Runx2 and its binding to bone sialoprotein (BSP) gene promoters, inducing the expression of osteogenic differentiation-specific genes such as OCN and OPN [28,46].Additionally, ascorbic acid modulates the synthesis of matrix metalloproteinase to promote ECM maturation and osteogenic differentiation of osteoblasts [47].Therefore, surface decoration with ascorbic acid can dramatically enhance the induction of osteogenic differentiation of osteoblasts by titanium-based materials, as evidenced by the results in figure 7. The synergy of the nanostructure and ascorbic acid greatly ameliorated the osteogenic properties of titanium-based prostheses and is expected to effectively promote the chest wall repair and reconstruction.Notably, the role of ascorbic acid modification was most prominent.

Conclusion
In conclusion, the nanorod structured surface decorated with ascorbic acid was constructed on the titanium substrate through alkali treatment and spincoating process.The surface inhibited hBCC proliferation and its mediated cell microenvironment restrained hBCC migration, probably due to the nanorod structure suppressing hBCC spreading and the ascorbic acid disrupting hBCC metabolism by inducing oxidative stress.Furthermore, the ascorbic acid-decorated nanostructured surface improved the proliferation and osteogenic differentiation, which may be related to the nanostructure and ascorbic acid mediating the collagen secretion, focal adhesion, and cell spreading of osteoblasts.The ascorbic acid-decorated nanostructured surface is expected to improve the efficacy of titanium-based prostheses for BC recurrence inhibition and osteogenesis induction in clinical applications of chest wall reconstruction.

Figure 2 .
Figure 2. Surface elemental and chemical composition of the samples.(A) Surface element concentrations.(B) EDS mapping images of the surface element distribution.(C) DR-FTIR spectra of the sample surfaces.

Figure 3 .
Figure 3. Detection of ascorbic acid immobilized on the sample, analysis of ascorbic acid releasing performance, and evaluation of layer stability.(A) The amount of immobilized ascorbic acid.(B) The amount of ascorbic acid released into the environment.(C) Percentage of remained ascorbic acid on the sample.

Figure 5 .
Figure 5. Oxidative stress levels of the hBCCs mediated by the samples.(A) Fluorescent images of intracellular reactive oxygen species (ROS).(B) Quantification of ROS levels in a single cell.* * p < 0.01 and * * * p < 0.001.

Figure 6 .
Figure 6.Effects of the samples on osteoblast vitality and cytoskeleton assembly.(A) Live/Dead viability/cytotoxicity assay: (a) fluorescent images of the labeled osteoblasts, (b) quantitative analysis of cell viability.(B) Cell proliferation assessed by MTT assay.(C) Cytoskeleton assembly: (a) fluorescent images, (b) quantitative analysis of cell spreading area.* * p < 0.01 and * * * p < 0.001.

Figure 7 .
Figure 7. Effects of the samples on osteogenesis of osteoblasts.(A) Heatmap for osteogenic-related gene expression of the osteoblasts cultured on the samples detected by qRT-PCR.(B) Alkaline phosphatase (ALP) activity of the osteoblasts cultured on the samples: (a) qualitative optical images, (b) quantitative analysis.(C) Collagen secretion of the osteoblasts cultured on the samples: (a) qualitative optical images, (b) quantitative analysis.(D) Extracellular matrix (ECM) mineralization of the osteoblasts cultured on the samples: (a) qualitative optical images, (b) quantitative analysis.* * * p < 0.001.