Multiscale Hybrid Surface Topographies Orchestrate Immune Regulation, Antibacterial Defense, and Tissue Regeneration

TL;DR

A new surface design helps implants integrate better by reducing infection, controlling immune responses, and promoting tissue healing.

Contribution

A modular nano–micro hybrid surface platform is introduced to simultaneously regulate bacteria, immune cells, and tissue regeneration.

Findings

Nanoscale features reduce bacterial biofilm by over 50% compared to flat surfaces.

Microscale features increase M2 macrophage markers by 3-fold and ALP activity by 8-fold.

Macrophages show context-dependent behavior, balancing inflammation and repair at the implant site.

Abstract

Implant-associated complications—including infection, adverse immune responses, and poor tissue integration—pose significant risks to patients, often leading to implant failure, revision surgeries, or chronic disease. Current chemical-based strategies, such as antibiotic or drug-releasing systems, are limited by short-term efficacy, narrow therapeutic windows, and potential toxicity. Surface topography offers a promising alternative, but most designs target single cell types and overlook the complex, multicellular dynamics at the implant–host interface. Here, a new multifunctional platform is introduced based on nano–micro hybrid wrinkled topographies fabricated via a custom nanofabrication method that combines layer-by-layer (LbL) self-assembly with mechanical nanomanufacturing. This system simultaneously modulates bacteria, immune cells, and tissue progenitors to enable antibacterial activity, immune regulation, and tissue regeneration. On hybrid surfaces, nanoscale features disrupt bacterial adhesion (>50% biofilm reduction vs. flat controls), while microscale features enhance macrophage polarization (≈3-fold increase in M2 markers) and osteogenic differentiation (>8-fold increase in ALP activity), indicating strong pro-healing responses. Notably, macrophages exhibit context-dependent behavior—driving inflammation during bacterial infection and repair in its absence—creating an immune-balanced microenvironment for implant integration. The modular nature of this platform allows expansion to other cell types and disease contexts, offering a broadly applicable strategy for next-generation biomaterials.

Figures

d](#F2),[e](#F2)). As the number of bilayers (*n*) increased from 1 to 20, wavelength (*λ*) increased from 423 to 2710 nm ([Figure 2d](#F2)), and amplitude (*A*) increased from 272 to 1370 nm ([Figure 2e](#F2)). This finding is in line with classical buckling theory, which predicts a wavelength of buckle, $[eqn]$ , as a function of coating thickness *h* ( $[eqn]$ is the elastic modulus of the coating (c) and substrate (s)).^[[24](#R24),[25](#R25)]^ This model qualitatively explains the formation of isotropic wrinkled structures via biaxial deformation. Our data confirm that topography can be p

f](#F3)), the pro-inflammatory markers TNF-*α* and IL1*β* were most highly expressed on the nano, suggesting a stronger inflammatory response. In contrast, macrophages on hybrid surfaces exhibited significantly lower expression, indicating reduced inflammation. For M2 polarization ([Figure 3g](#F3)), anti-inflammatory markers IL-10 and IL-4 were significantly upregulated on micro and hybrid surfaces, confirming that these topographies effectively promote M2 activation. Beyond their immunomodulatory effects, macrophages also play an important role in bone tissue regeneration by secreting osteoi

a](#F4),[b](#F4)). Fluorescence staining of cytoskeleton actin (green) and nuclei (blue) revealed robust cell adhesion across all surfaces, with significant differences in cell spreading ([Figure 4c](#F4)). Cells on coated planar controls adopted a round-to-polygonal shape with smooth boundaries and flat lamellipodia, increasing surface coverage 2.63-fold compared to uncoated controls. On nano surfaces, cells exhibited slightly elongated shapes with enhanced filopodia, leading to a 2.29-fold increase, similar to uncoated controls. In contrast, micro and hybrid surfaces significantly enhanced c

e](#F5)). Runx2 regulates early osteoblast differentiation and mineralization, and OCN acts as a late-stage marker involved in mitochondrial gene regulation and matrix mineralization. At both day 7 and 14, cells cultured on micro and hybrid surfaces exhibited significantly higher expression levels of ALP, Runx2, and OCN compared to other groups. By day 14, hybrid surfaces showed the highest gene expression, with a 5.46 fold increase in ALP, a 2.62 fold increase in Runx2, and a 2.61 fold increase in OCN relative to uncoated controls. Micro surfaces also induced strong expression, though slightl

e](#F6),[f](#F6)). For *S. aureus*, biofilm formation was lowest on the hybrid surface (46.8 ± 11.2% relative to uncoated controls), with bacterial viability reduced to 59.0 ± 8.86%. The nano and micro surfaces also reduced biofilm formation to 59.7 ± 8.49% and 85.7 ± 11.8%, with viabilities standing at 87.1 ± 10.2% and 102 ± 13.8%, respectively. Similarly, for *E. coli*, biofilm formation on the hybrid surface was reduced to 37.0 ± 12.2%, with viability reduced to 54.1 ± 17.8%. To assess the release of film components, we conducted the Kirby-Bauer disk diffusion assay using wrinkled MXene LbL

c](#F7),[d](#F7)). After co-culture, samples were washed to remove non-adherent bacteria, and stained with fluorescent dyes: DAPI to label both bacteria and macrophage nuclei, and F-actin (GFP) to stain F-actin in the macrophage cytoplasm. IF images showed more pseudopod extensions of macrophages toward bacteria on patterned surfaces, confirming enhanced macrophage activation on these surfaces. The number of phagocytosed bacterial cells was quantified manually after deconvolution of the microscopic images ([Figure 7d](#F7)). Macrophages on hybrid and nano surfaces exhibited the highest levels

d](#F8)) in which hybrid nano-micro topographies promote macrophage adhesion and mechanosensing through integrin clustering and focal adhesion formation, which in turn activate downstream RhoA/ROCK signaling. This signaling cascade regulates cytoskeletal dynamics and gene expression, ultimately influencing macrophage polarization. This mechanism is further reinforced by the upregulation of Arp3 and Arp4, which promote actin polymerization and podosome formation, essential processes for macrophage motility, adhesion, and polarization.^[[43](#R43),[48](#R48),[50](#R50),[51](#R51)]^ Collectively,

Full text

Introduction

Implant-associated complications—such as infections, adverse immune responses, and poor tissue integration—remain major barriers to long-term success of biomedical implants, from orthopedic implants to cardiovascular and neural prosthetics.^[1,2]^ Among these, infections are one of the most frequent and severe complications, accounting for >25% of all healthcare-associated infections in the U.S.^[3]^ They are driven by bacterial adhesion and biofilm formation, significantly increasing the risk of implant failure (e.g., >25% for artificial heart valves, up to 44% for post-mastectomy breast implants, 55% for revision hip arthroplasties).^[4–6]^ Left untreated, infections can lead to chronic disease, complex removal/replacement surgeries, sepsis, and even early death. Beyond infections, adverse immune responses and inadequate tissue integration also pose additional challenges.^[7–9]^ Biomaterials can trigger immune cascades, disrupting the host immune system and leading to fibrosis and collagen encapsulation—a phenomenon known as the foreign-body response (FBR).^[10–12]^ These complications not only compromise patient outcomes but also impose billions of dollars of healthcare costs.

Traditional preventive strategies to mitigate infections and FBR rely on chemical cues—antibiotics, cytokines, growth factors, and host defense peptides—to provide antibacterial or immunomodulatory functionality.^[13–15]^ However, these strategies have fundamental limitations, including short-term efficacy due to the depletion of therapeutic agents, narrow therapeutic windows, and potential systemic toxicity (i.e., off-target effects). Furthermore, rising antibiotic resistance, including methicillin-resistant Staphylococcus aureus (MRSA) and extended-spectrum beta-lactamase (ESBL)-producing bacteria, further complicates infection control.

A promising alternative is the use of physical cues, particularly surface topography. Nanoscale and microscale topographies have been shown to modulate bacterial adhesion and host cell responses.^[16–19]^ For example, nanoscale topographies have bactericidal effects by mechanically disrupting bacterial membranes, while microscale topographies can inhibit bacterial adhesion by altering cell-material interactions. These topography-based approaches provide long-term antibacterial performance without relying on antimicrobial agents, thus reducing the risk of resistance. Beyond antibacterial effects, topographical cues can regulate host immune and tissue cell behavior, fostering a favorable immune microenvironment for tissue healing and integration (when required).^[20,21]^

Despite these advances, the pathophysiology of implant-associated complications remains poorly understood, particularly in the context of host-pathogen interactions on implant surfaces. Current research has primarily focused on two separate tracks: (i) topography–bacteria interactions, and (ii) topography–host interactions (Table S1, Supporting Information).^[1]^ However, these approaches fail to capture the dynamic crosstalk between materials, host cells, and pathogens, which is essential for a comprehensive understanding of implant complications. Surprisingly, topography-based surface modifications such as micro-roughening are widely used in medical devices (e.g., orthopedic/breast implants, stents, catheters) to improve performance, yet their potential impact on immune protection and infection susceptibility remains poorly understood.

To address these gaps, we developed a generalizable strategy to explore the multifunctionality of topographical cues in simultaneously guiding host cell behavior and suppressing bacterial colonization. To target multiple cell types across scales, we fabricated multiscale hybrid topographies—integrating nano- and microscale wrinkled features into a single system—using an upgraded version of our bottom-up nanofabrication methods.^[16,17]^ In this study, we used orthopedic implants as a model application and focused on examining interactions among three key cell types: i) bacteria, ii) immune cells, and iii) tissue cells. We hypothesized that nano-features in hybrid topographies would disrupt bacterial adhesion via mechanical interactions, while micro-topographies would enhance immune and tissue cell adhesion and spreading. We further expected context-dependent immune responses depending on bacterial presence: pro-inflammatory (M1) activation under bacterial challenge and anti-inflammatory (M2) polarization in sterile conditions.

This study presents a novel hybrid platform that integrates nano- and micro-wrinkled topographies into a single system that can induce mechanotransductive signaling across multiple cell types within a single system (Figure 1). By integrating both nano- and microscales, this approach uniquely bridges antibacterial, immunomodulatory, and osteogenic functions—unlike prior studies that examine these responses separately. To our knowledge, this is the first demonstration of a topographically engineered system that simultaneously achieves antibacterial, immunomodulatory, and tissue-regenerative effects to address infection and FBR for improved host–implant integration. We systematically evaluated the impact of multiscale and monoscale wrinkled surfaces on cell behaviors and intercellular interactions between i) bacteria, ii) immune cells (macrophages), and iii) tissue cells (preosteoblasts) using co-culture models. Additionally, we performed transcriptomic analysis to dissect immune cell (macrophage) gene expression and signaling pathways in response to different topographies. Our data demonstrate that the hybrid wrinkled topographies can i) significantly reduce bacterial adhesion and biofilm formation, ii) induce context-dependent immune modulation depending on bacterial presence, and iii) promote osteogenic proliferation and differentiation. Notably, the modular and scalable nature of these approaches could enable broad applicability across healthcare, agriculture, food safety, environmental protection, and other fields.

Results

System Development

2.1.

Fabrication of Monoscale and Multiscale Wrinkled Topographies

2.1.1.

We developed a customized bottom-up nanofabrication approach to create precisely controlled wrinkled topographies across multiple length scales (nm to μm). This involves creating wrinkled structures via controlled mechanical shrinkage of a stiff coating on a soft, compliant substrate. A key aspect of this approach is the use of 2D sheet-like nanomaterials, specifically MXene—a family of transition metal carbides and nitrides—to create ultrathin, mechanically stiff coatings suitable for precise nanopatterning. To achieve nanoscale precision, we used electrostatic layer-by-layer (LbL) self-assembly, which enables subnanoscale control, scalability, and cost-efficient processing. As a complementary species of MXene, we selected gentamicin sulfate (GS)—a positively charged antibiotic—due to its small molecular size (477 Da), strong electrostatic interactions with MXene, and broad-spectrum bactericidal activity (Figure S1, Supporting Information).

Figure 2a illustrates the processes of monoscale and multiscale wrinkled topography fabrication. The fabrication process for monoscale topographies involved two main steps: i) LbL coating of a MXene multilayer film on a biaxially prestrained polystyrene (PS) substrate, and ii) mechanical nanomanufacturing via thermal annealing (130 °C), which biaxially shrinks (ϵ2D, x, y-direction) the PS substrate, forming isotropically wrinkled MXene LbL films due to modulus mismatch between the coating and the substrate.^[22,23]^ To create multiscale hybrid topographies, we introduced an additional set of steps: i) fabricate a 1st-generation (G_1_) wrinkled coating on a prestrained PS substrate, ii) remove of the PS substrate via dissolving in dichloromethane (DCM), iii) transfer of the freestanding G1 wrinkled coating onto another MXene LbL-coated PS substrate, and iv) apply a second mechanical nanomanufacturing step (ϵ2D) to achieve multiscale hierarchical structures (G_2_) with independently tunable primary and secondary pattern sizes.

Characterization of Wrinkled Topographies

2.1.2.

Our LbL multilayers, denoted as (MXene/GS)n, exhibited linear growth at high resolution (≈6 nm per bilayer or ≈3 nm per layer) via the orderly stacking of atomically thin MXene nanosheets (Figure 2c). As the number of bilayers (n) increased from 1 to 20, the coatings progressively darkened due to increasing MXene content (Figure 2c). This trend confirms the precise control of coating thickness through LbL assembly, which is critical for defining wrinkle feature sizes.

We characterized wrinkled topographies (both monoscale and multiscale) using SEM and AFM (Figure 2b, f). SEM images revealed sharp edges of the MXene wrinkles and variations in feature size and density. For hybrid structures, we first generated nanoscale wrinkled G_1_ coatings (with (MXene/GS)3 films), transferred them onto (MXene/GS)20 films, and applied a second mechanical nanomanufacturing step to create nano-micro hybrid topographies, as confirmed by SEM imaging (Figure 2b).

Wrinkled structures were defined by wavelength (λ, distance between peaks) and amplitude (A, peak-to-valley height difference). To quantify amplitude, we modeled ridge-like structures as triangular waveforms, using A = 2√3×RMS (root-mean-square). A linear relationship was observed between the thickness of the as-prepared LbL coating and the feature sizes of G_1_ wrinkled structures (Figure 2d,e). As the number of bilayers (n) increased from 1 to 20, wavelength (λ) increased from 423 to 2710 nm (Figure 2d), and amplitude (A) increased from 272 to 1370 nm (Figure 2e). This finding is in line with classical buckling theory, which predicts a wavelength of buckle, $[eqn]$ , as a function of coating thickness h ( $[eqn]$ is the elastic modulus of the coating (c) and substrate (s)).^[24,25]^ This model qualitatively explains the formation of isotropic wrinkled structures via biaxial deformation. Our data confirm that topography can be precisely controlled by adjusting coating thickness or the number of LbL layers, offering a scalable strategy for engineering surface patterns.

For biological studies, we categorized the wrinkled multilayers into three distinct groups based on characteristic wavelengths: nano (≈400 nm), micro (≈2700 nm), and hybrid (nano-micro combination). For hybrid structures (G_2_ generation), we observed both small (λ1 = 225 ± 30.7 nm) and large (λ2 = 3450 ± 823 nm) wavelengths, along with a corresponding amplitude of A1 = 1200 ± 222 nm, reflecting the integration of nano- and microscale features. These results demonstrate the tunability of structural parameters via film thickness and strain control, allowing precise modulation of topographical features for biomaterial applications.

Regarding the mechanical properties of our engineered coatings, since the platform is implemented as an ultra-thin coating on a bulk material, it is not expected to significantly alter the intrinsic mechanical properties of the underlying material, which primarily determine load-bearing capacity in bone repair. Instead, the critical requirement is that the coating itself remain mechanically robust and stably adhered to the substrate. To assess this, we tested resistance to interfacial failure under ultrasonication: planar MXene multilayer coatings delaminated after just 10 min in water, whereas wrinkled coatings (both monoscale and multiscale) remained intact with negligible morphological distortion (Figure S2, Supporting Information). We further examined degradation stability by incubating the coatings in cell culture media for 14 days; SEM analysis revealed no significant morphological changes (Figure S3a, Supporting Information). In addition, the coatings withstood repeated autoclaving cycles without delamination or morphological degradation (Figure S3b, Supporting Information), further demonstrating stability under sterilization conditions, which is an essential requirement for future preclinical and clinical evaluation. Adhesion strength, quantified by lap-shear testing, reached 21.5 ± 0.82 MPa for hybrid coatings on PS substrates (Figure S4, Supporting Information), confirming robust bonding that ensures the coating remains intact during handling and under physiological stresses. This value lies at the upper range of typical polymer–polymer joints (≈10–25 MPa) and substantially exceeds many reported biomedical coatings (≈2–10 MPa), underscoring the robust interfacial stability of our system.

Topography–Biology Interactions

2.2.

Topography–Immune Cell Interactions (Macrophage Polarization)

2.2.1.

The immune microenvironment is crucial for host defense and tissue healing, regulating immune and tissue cell behavior.^[26,27]^ Upon implantation, the innate immune system reacts immediately, creating a localized immune niche that ultimately shapes implant-to-bone integration.^[28–30]^ Among the first immune cells to interact with the implant, macrophages orchestrate the immune response, regulate osteogenesis, and prevent infection. These adaptable cells polarize into two distinct phenotypes: M1 (pro-inflammatory, antibacterial) and M2 (anti-inflammatory, pro-healing).^[31]^ Their polarization is influenced by environmental cues, with M1 macrophages typically exhibiting a rounded shape, while M2 macrophages adopt an elongated morphology.^[32]^ Building on the link between macrophage morphology and function, as well as our previous findings,^[16]^ we hypothesized that microtopographical cues present on micro and hybrid surfaces, which promote enhanced cell adhesion and spreading, would more effectively drive M2 polarization compared to nanotopographies or flat surfaces. To test this, we analyzed how macrophage morphology and activation were influenced by different surface structures.

To assess immunomodulatory effects, RAW 264.7 macrophages were cultured on various test surfaces for 24 h (Figure 3a). The test groups included wrinkled MXene multilayers (“nano,” “micro,” and “hybrid”) and two controls: uncoated planar (denoted “control”) and coated planar (denoted “planar”), as summarized in Table S2 (Supporting Information). The samples were analyzed using i) SEM to assess cell morphology, ii) immunofluorescence (IF) staining to evaluate macrophage polarization via iNOS (M1 marker) and CD206 (M2 marker) expression, and iii) qPCR to analyze gene expression associated with macrophage activation and osteoinductive potential.

As shown in SEM images (Figure 3b), macrophages on micro and hybrid surfaces displayed significantly greater elongation, with higher aspect ratios (5.07 and 5.79, respectively) compared to nano, uncoated, and coated planar controls, which ranged from 1.60 to 2.13 (Figure 3c). These findings suggest that micro and hybrid topographies promote cell elongation, favoring M2 polarization. IF staining (Figure 3d,e) further confirmed these observations, showing that iNOS expression was highest on uncoated controls, decreased on wrinkled topographies, and was lowest on hybrid surfaces (0.79 relative to uncoated controls), indicating reduced M1 activation. Conversely, CD206 expression was highest on micro (1.36) and hybrid (1.27) surfaces, while flat uncoated controls exhibited the lowest levels, confirming that micro and hybrid topographies drive M2 polarization, while unpatterned surfaces favor M1 activation.

To further investigate macrophage polarization, we performed per-cell image cytometry analysis of iNOS and CD206 expression (Figure S5, Supporting Information). The proportion of M1-like cells was highest on nano (25.0%) and planar (18.6%) surfaces, lower on control (9.82%), and lowest on micro (7.33%) and hybrid (5.24%) topographies. Conversely, M2-like cells were most abundant on micro (60.0%) and hybrid (47.1%) surfaces, compared to lower levels on planar (32.0%), nano (23.0%), and controls (7.98%). Accordingly, the M1/M2 ratio was highest on nano and planar surfaces and lowest on micro and hybrid surfaces, confirming that micro and hybrid topographies strongly promote M2 polarization while suppressing M1 polarization.

We next performed qPCR analysis to evaluate M1- and M2-associated gene expression. The results revealed topography-dependent macrophage activation. For M1 polarization (Figure 3f), the pro-inflammatory markers TNF-α and IL1β were most highly expressed on the nano, suggesting a stronger inflammatory response. In contrast, macrophages on hybrid surfaces exhibited significantly lower expression, indicating reduced inflammation. For M2 polarization (Figure 3g), anti-inflammatory markers IL-10 and IL-4 were significantly upregulated on micro and hybrid surfaces, confirming that these topographies effectively promote M2 activation. Beyond their immunomodulatory effects, macrophages also play an important role in bone tissue regeneration by secreting osteoinductive factors such as bone morphogenetic protein-2 (BMP-2) and oncostatin M (OSM). The expression of OSM, a bone damage marker, was highest on nano and planar surfaces, suggesting a more inflammatory environment, whereas OSM expression was significantly lower on micro and hybrid surfaces, indicating a more favorable bone-healing environment. Interestingly, BMP-2, a key osteoinductive marker, was significantly upregulated in macrophages on micro and hybrid surfaces (Figure 3h). This suggests that these topographies not only promote M2 polarization but also enhance the secretion of osteoinductive factors, fostering an environment conducive to bone regeneration.

In addition to evaluating macrophage polarization, we examined the biosafety assessment of our engineered topographies. No detectable cytotoxicity was observed on any of the surfaces up to 5 days of culture, and viability was higher on micro and hybrid structures, suggesting enhanced adhesion and proliferation compared to flat controls (Figure S6a, Supporting Information). To further evaluate the effects of degradation products, we conducted extract cytocompatibility assays, using extracts generated by incubating topographical samples in culture medium for 14 days (Figure S6b, Supporting Information), confirming the biosafety of the scaffold degradation products under in vitro conditions.

Topography–Tissue Cell Interactions (Osteogenesis)

2.2.2.

Successful osseointegration depends on the migration and attachment of osteoprogenitor cells, guided by chemoattractive signals.^[33,34]^ Surface topographical cues provide mechanotransductive signals that regulate cell morphology, proliferation, and differentiation into osteoblasts.^[35,36]^ Based on our previous findings,^[16]^ we hypothesized that microscale topographies (micro and hybrid) would enhance osteogenic behavior more effectively than nanoscale topographies. To validate this, we evaluated the morphology, proliferation, and differentiation of murine preosteoblast MC3T3-E1 cells cultured on nano, micro, and hybrid surfaces, with uncoated and (MXene LbL)-coated planar substrates as controls.

We evaluated the morphology and spreading of preosteoblast cells after 24 h using fluorescence imaging and SEM (Figure 4a,b). Fluorescence staining of cytoskeleton actin (green) and nuclei (blue) revealed robust cell adhesion across all surfaces, with significant differences in cell spreading (Figure 4c). Cells on coated planar controls adopted a round-to-polygonal shape with smooth boundaries and flat lamellipodia, increasing surface coverage 2.63-fold compared to uncoated controls. On nano surfaces, cells exhibited slightly elongated shapes with enhanced filopodia, leading to a 2.29-fold increase, similar to uncoated controls. In contrast, micro and hybrid surfaces significantly enhanced cell adhesion and spreading, with cells adopting osteoblast-like morphologies characterized by stellate shapes and elongated cytoplasmic processes that extended lamellipodia to anchor onto protrusions. This resulted in 4.00-fold and 7.13-fold increases in surface coverage for micro and hybrid groups, respectively. Quantification of individual cell spreading confirmed that hybrid surfaces promoted the most extensive spreading (2.41 fold vs uncoated flat controls), followed by micro-patterned surfaces (2.36 fold).

To assess cell elongation, we quantified cell aspect ratios from SEM images (Figure 4d). Cells on hybrid surfaces exhibited the highest aspect ratios (5.79 fold vs nano/uncoated). Similarly, microsurfaces showed a 5.07 fold increase, suggesting that both micro and hybrid topographies promote osteoblast-like phenotypes conducive to bone formation.

Preosteoblast proliferation was assessed over 5 days (Figure 4e). All coated surfaces promoted higher proliferation than uncoated controls, with micro and hybrid surfaces demonstrating the highest proliferation rates. On Day 1, cell counts on hybrid (65.8 ± 30.5 × 10^3^) and micro (37.8 ± 13.8 × 10^3^) exceeded those of uncoated controls (22.9 ± 10.9 × 10^3^). By Day 3, proliferation on micro (362 ± 86.0 × 10^3^) and hybrid (312 ± 77.0 ×10^3^) significantly outpaced uncoated controls (99.0 ± 40.0 × 10^3^). On Day 5, micro (593 ± 49.0 × 10^3^) and hybrid (604 ± 110 × 10^3^) continued to outperform the uncoated control (159 ± 114 × 10^3^).

To further evaluate osteogenic potential, alkaline phosphatase (ALP) activity was measured after culturing preosteoblasts on all surfaces in osteogenic induction media (Figure 5a–d). Due to the nontransparent nature of MXene coatings, cells were transferred to 24-well plates for accurate colorimetric assessment under transmitted light microscopy.

By day 7, hybrid-patterned surfaces exhibited the highest ALP activity (2.47 fold vs uncoated control), followed closely by micro-patterned surfaces (2.14), consistent with our hypothesis (Figure 5c). Coated planar surfaces showed moderate ALP activity (0.97), indicating that MXene multilayers alone contribute minimally to early osteogenic differentiation. Nano surfaces showed ALP activity comparable to uncoated controls, suggesting limited osteogenic influence from nanoscale topographies.

By day 14, hybrid surfaces maintained the highest ALP activity, reaching 8.15 times the uncoated control at day 7 and showing a 3.39-fold increase over the uncoated control at day 14 (Figure 5c). Microsurfaces also sustained robust osteogenic activity, with ALP levels reaching 7.02 and 2.93 times the uncoated control at day 7 and day 14, respectively. The coated planar group showed only a modest increase in ALP activity (2.92), while nano-patterned surfaces were comparable to planar surfaces, confirming the superior osteogenic efficacy of microscale topographies.

To explore osteogenic differentiation at the gene level, we performed qPCR analysis of key osteogenic markers—ALP, runt-related transcription factor 2 (Runx2), and osteocalcin (OCN)—after 7 and 14 days of culture in groups (Figure 5e). Runx2 regulates early osteoblast differentiation and mineralization, and OCN acts as a late-stage marker involved in mitochondrial gene regulation and matrix mineralization. At both day 7 and 14, cells cultured on micro and hybrid surfaces exhibited significantly higher expression levels of ALP, Runx2, and OCN compared to other groups. By day 14, hybrid surfaces showed the highest gene expression, with a 5.46 fold increase in ALP, a 2.62 fold increase in Runx2, and a 2.61 fold increase in OCN relative to uncoated controls. Micro surfaces also induced strong expression, though slightly lower than hybrid surfaces. These data confirm that micro and hybrid topographies directly enhance osteogenic gene expression in preosteoblasts.

Topography–Macrophage–Preosteoblast Crosstalk (Paracrine effects)

2.2.3.

To investigate whether surface topography influences osteogenesis indirectly through immune signaling, RAW 264.7 macrophages were cultured on each surface for 48 h. After macrophage culture and polarization, cytokine-containing supernatant was collected and applied to preosteoblast cultures to evaluate osteogenic response (Figure 5b). ALP activity increased significantly on micro and hybrid surfaces when preosteoblasts were exposed to macrophage-conditioned media (Figure 5d). By day 14, ALP activity peaked for hybrid surfaces (11.4 fold vs uncoated control), followed by micro surfaces (8.97). These results indicate that macrophages polarized on micro and hybrid surfaces secrete cytokines that strongly enhance osteogenic differentiation, further amplifying the already robust direct topography-driven osteogenesis.

Topography–Bacteria Interactions (Antibacterial Activity)

2.2.4.

In addition to promoting osteogenesis and immunomodulation, surface topographies have been shown to be effective in antibacterial defense—a key factor in preventing infections associated with implantable materials.^[37,38]^ Based on our previous findings,^[17]^ we hypothesized that nanoscale topographies (nano and hybrid) would show more effective antibacterial activity than microscale topographies. To test this, we evaluated bacterial survival rates and bacterial coverage on five different types of surfaces, as described above. We used two model bacterial species: i) a Gram-positive S. aureusus USA300, a leading cause of prosthetic valve endocarditis, catheter-associated bloodstream infection, and about one third of surgical-site infections; and ii) a Gram-negative E. coli UTI89, the primary causative agent of catheter-associated urinary tract infections (CAUTI).

First, to evaluate the antibacterial properties, we used the spread plate method (SPM) to quantify bacterial survival rates after 24 and 48 h of incubation with bacteria. Bacteria from each surface were harvested and cultured on agar plates for quantification (Figure 6a,b). Compared to uncoated controls, the coated planar, nano, micro, and hybrid surfaces all significantly reduced bacterial survival. For S. aureus, survival rates were 72.9 ± 15.7% (coated planar), 25.6 ± 16.1% (nano), 63.6 ± 26.7% (micro), and 20.8 ± 11.9% (hybrid). For E. coli, survival rates were 63.2 ± 20.3% (coated planar), 44.7 ± 9.00% (nano), 91.7 ± 9.38% (micro), and 23.2 ± 11.4% (hybrid).

Next, SEM imaging provided further insights into bacterial adhesion and morphology on each surface (Figure 6c). On uncoated and coated planar surfaces, S. aureus formed dense clusters with extensive biofilm formation. On microsurfaces, bacterial clusters were fewer and more dispersed, indicating a moderate reduction in adhesion. On nano surfaces, bacterial adhesion was further reduced, with sparse and less dense clusters. The hybrid surface showed the most pronounced antibacterial effect, with bacterial cells sparsely distributed and large areas completely free of bacteria, indicating minimal biofilm formation. A similar trend was observed for E. coli, with the hybrid and nano surfaces significantly reducing bacterial adhesion and disrupting bacterial morphology compared to the other groups.

To further assess antibacterial activity, we performed immunofluorescence imaging to visualize live (green) and dead (red) bacterial cells (Figure 6d). On uncoated and coated planar surfaces, S. aureus biofilms were predominantly composed of live cells, indicating minimal antibacterial activity. The micro surface showed an increase in dead cells, indicating moderate antibacterial effects, while the nano surface showed an even greater proportion of dead cells. The hybrid surface exhibited the strongest antibacterial effect, with the highest proportion of dead cells and minimal live cells. A similar trend was observed for E. coli, where the hybrid surface demonstrated the greatest reduction in live bacteria among all groups.

Statistical analysis further confirmed these trends (Figure 6e,f). For S. aureus, biofilm formation was lowest on the hybrid surface (46.8 ± 11.2% relative to uncoated controls), with bacterial viability reduced to 59.0 ± 8.86%. The nano and micro surfaces also reduced biofilm formation to 59.7 ± 8.49% and 85.7 ± 11.8%, with viabilities standing at 87.1 ± 10.2% and 102 ± 13.8%, respectively. Similarly, for E. coli, biofilm formation on the hybrid surface was reduced to 37.0 ± 12.2%, with viability reduced to 54.1 ± 17.8%. To assess the release of film components, we conducted the Kirby-Bauer disk diffusion assay using wrinkled MXene LbL films, which provides qualitative information on the diffusion of GS or MXene through agar by measuring the clear zone of inhibition (ZOI). No ZOI was observed against S. aureus, indicating negligible or no release of the film components (Figure S7, Supporting Information).

These findings highlight the superior antibacterial performance of hybrid and nano surfaces in reducing bacterial adhesion, biofilm formation, and viability for both bacterial species.

Topography–Macrophage–Bacteria Crosstalk (Phagocytosis)

2.2.5.

In addition to directly influencing bacterial adhesion and biofilm formation, topographies can also reshape the defensive role of immune cells, particularly macrophages, in responding to bacterial colonization.^[39]^ We hypothesized that, under bacterial challenge, nanoscale topographies (nano and hybrid), in combination with bacterial chemical cues (e.g., toxins and extracellular polymeric substances), would enhance macrophage activation, promoting inflammatory responses and antibacterial defense mechanisms. To examine how different topographies influence macrophage–bacteria interactions, we performed co-culture experiments in which macrophages were introduced onto pre-colonized surfaces. Specifically, S. aureus and E. coli were first allowed to adhere to the test surfaces for 24 and 48 h, respectively, before macrophages were seeded onto the samples and incubated for an additional 2 h. To confirm that the DMEM used for macrophage culture did not affect bacterial viability, control samples were incubated with DMEM alone. SEM and immunofluorescence (IF) imaging were used for quantitative analysis of antibacterial activity.

SEM imaging provided detailed insights into macrophage–bacteria interactions, showing the positioning of macrophage membranes and pseudopods. Macrophages on patterned surfaces exhibited greater extension of pseudopods and higher activity compared to those on uncoated and coated controls (Figure 7a). Biofilm coverage was quantified and normalized to bacterial coverage on each surface without macrophages. Across all surface types, biofilm coverage decreased for both strains in the presence of macrophages, indicating active phagocytosis (c.f. Figures 7b and 6e). For S. aureus, biofilm coverage (normalized to the uncoated control without macrophages) was reduced to 59.4 ± 18.9% (uncoated), 74.2 ± 18.6% (coated planar), 22.2 ± 10.7% (nano), 40.2 ± 14.8% (micro), and 8.05 ± 5.82% (hybrid). For E. coli, the hybrid group exhibited the lowest biofilm coverage at 13.3 ± 8.51%, while the uncoated control showed the highest at 60.6 ± 19.0%.

Phagocytosis was visualized and quantified as the number of engulfed bacteria per immune cell (Figure 7c,d). After co-culture, samples were washed to remove non-adherent bacteria, and stained with fluorescent dyes: DAPI to label both bacteria and macrophage nuclei, and F-actin (GFP) to stain F-actin in the macrophage cytoplasm. IF images showed more pseudopod extensions of macrophages toward bacteria on patterned surfaces, confirming enhanced macrophage activation on these surfaces. The number of phagocytosed bacterial cells was quantified manually after deconvolution of the microscopic images (Figure 7d). Macrophages on hybrid and nano surfaces exhibited the highest levels of bacterial phagocytosis, with an average of 8.95 ± 3.44 and 8.45 ± 3.73 E. coli cells, and 6.95 ± 2.04 and 6.35 ± 3.10 S. aureus cells phagocytosed per macrophage cell, respectively. These results demonstrate that topographical features enhance macrophage activity and bacterial phagocytosis compared to planar controls, consistent with our hypothesis.

Unlike previous studies showing that certain biomaterials can impair immune cell function and reduce antibacterial efficacy,^[40,41]^ our findings demonstrate that topographical cues not only preserve macrophage function but also enhance their phagocytic activity, significantly reducing bacterial biofilm formation and viability.

Transcriptomic Analysis of Macrophage Polarization on Wrinkled Topographies

2.3.

To better understand how surface topographies influence immune responses at the implant interface, we conducted transcriptomic analysis of macrophages cultured on different topographies. We focused on macrophages because they play a central regulatory role in the immune-tissue microenvironment, acting as key mediators between the implant surface and surrounding tissues. Given their pivotal role in orchestrating immune surveillance, inflammation, and tissue remodeling, we reasoned that understanding how macrophages respond at the transcriptional level to different topographies could provide key mechanistic insights into how surface design shapes implant outcomes.

To isolate the effects of chemical and physical cues, we compared gene expression between coated control versus uncoated control (assessing the chemical effects of MXene coatings) and wrinkled topographies versus coated control (assessing topography-driven effects). As shown in Table S3 (Supporting Information), significant differences in gene expression were observed across these comparisons: i) 302 differentially expressed genes (DEGs) in coated control versus uncoated control, ii) 172 DEGs in nano versus coated control, iii) 379 DEGs in micro versus coated control, and iv) 411 DEGs in hybrid versus coated control, with 16 genes shared across all groups compared to the uncoated control. Volcano plots (Figure S8 and Table S3, Supporting Information) highlighted the up/downregulated genes in each comparison, with: i) 132 upregulated and 170 downregulated genes (coated control vs uncoated control), ii) 81 upregulated and 91 downregulated genes (nano vs coated-control), iii) 68 upregulated and 311 down-regulated genes (micro vs coated-control), and (iv) 83 upregulated and 328 downregulated genes (hybrid vs coated-control). These results indicate that surface chemistry and topography both drive substantial gene expression changes in macrophages, with the hybrid nano-micro topography eliciting particularly extensive transcriptional responses, consistent with enhanced mechanotransduction.

To classify these DEGs, we performed Gene Ontology (GO) analysis, categorizing them into biological process (BP), molecular function (MF), and cellular component (CC) classes. The top enriched GO terms for coated control versus uncoated control (Figure S9, Supporting Information) and hybrid versus coated control (Figure 8a) included processes related to cell adhesion, cytoskeletal organization, cellular localization, and tissue development, aligning closely with the morphological changes observed in macrophages cultured on MXene-coated and wrinkled surfaces. These results indicate that cytoskeletal remodeling and surface adhesion dynamics are central to the macrophage transcriptional response to topographical features.

To further probe how these transcriptional changes map onto biological pathways, we performed KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway analysis. For control versus planar (Figure S9, Supporting Information), enriched pathways included PI3K-AKT signaling, MAPK signaling, and rheumatoid arthritis, reflecting the immunomodulatory role of MXene coatings.^[42,43]^ In contrast, for hybrid versus coated control (Figure 8b), pathways associated with the regulation of actin cytoskeleton and focal adhesion were upregulated, while IL-17 signaling, osteoclast differentiation, rheumatoid arthritis, and TNF signaling were downregulated. This pathway shift is consistent with a pro-healing, M2 polarization profile, supporting a tissue-repair phenotype in the absence of infection.^[44–47]^

To better capture the role of hybrid topography-driven mechanotransduction, we profiled the expression of key genes associated with focal adhesion assembly, cytoskeletal remodeling, and cell-surface sensing (Figure 8c). Compared to planar and nano surfaces, hybrid surfaces induced consistently higher expression of integrin subunits (α3, α5, β1, β7), focal adhesion-associated proteins (vinculin, talin, Src), and Rho GTPases (RhoA, Rock1, Rock2, Rac2, Rac3). These molecules are critical regulators of cell spreading, formation of membrane protrusions (e.g., filopodia), and dynamic actin remodeling—key processes that underlie macrophage mechanosensing and are strongly associated with M2 polarization.^[44,48,49]^

Together, these findings support a working model (Figure 8d) in which hybrid nano-micro topographies promote macrophage adhesion and mechanosensing through integrin clustering and focal adhesion formation, which in turn activate downstream RhoA/ROCK signaling. This signaling cascade regulates cytoskeletal dynamics and gene expression, ultimately influencing macrophage polarization. This mechanism is further reinforced by the upregulation of Arp3 and Arp4, which promote actin polymerization and podosome formation, essential processes for macrophage motility, adhesion, and polarization.^[43,48,50,51]^ Collectively, these data suggest that the hybrid topography acts as a topographical niche, integrating nanoscale features that enhance mechanosensitive signaling with microscale structures that promote stable focal adhesion formation, resulting in a pro-healing M2-like phenotype conducive to improved host–implant integration.

Discussion

Implant integration requires coordination among immune, tissue, and bacterial responses, yet most studies focus on single-cell models.^[1]^ While significant progress has been made in understanding how surface topography regulates individual cell behavior, this narrow focus fails to capture the multidirectional crosstalk at the implant interface. Our work sought to address this critical gap by introducing a new multifunctional topographical platform that integrates immune modulation, antibacterial defense, and tissue regeneration within a single system. By combining nano- and microscale wrinkled features, our hybrid surfaces orchestrate biological responses across key cell types—macrophages, bacteria, and osteoprogenitors—via mechanotransductive signaling.

Macrophages are among the first responders to implants and serve as central regulators of both antibacterial defense and tissue repair, making them a natural starting point for our analysis. On hybrid surfaces, macrophages adopted context-dependent phenotypes: under sterile conditions, they polarized toward an M2-like, pro-healing state with elongated morphology and elevated anti-inflammatory and osteoinductive markers (e.g., CD206, IL-10, BMP-2), while under bacterial challenge, they shifted toward an M1-like phenotype, enhancing phagocytic activity and pathogen clearance.

This immune plasticity was tightly coupled with osteogenic and antibacterial outcomes. Through paracrine signaling, macrophages promoted osteogenic differentiation of preosteoblasts, with ALP activity (up to 11.4 fold vs uncoated controls). In addition to these immune-mediated effects, hybrid surfaces directly enhanced preosteoblast spreading, ALP activity, and osteogenic gene expression. In parallel, hybrid surfaces suppressed bacterial adhesion and biofilm formation, reducing biofilm coverage by ≈55% for S. aureus and ≈65% for E. coli. In macrophage–bacteria co-cultures, bacterial challenge further amplified the antibacterial response, as macrophages exhibited enhanced phagocytosis and synergistically decreased biofilm burden. To our knowledge, this level of coordinated, multifunctional integration is rarely—if ever—achieved in current biomaterial strategies, which typically address immune regulation, antibacterial defense, and osteogenesis separately.

This multifunctionality arises from the synergistic design of nano-micro hybrid wrinkled topographies. Nanoscale features enhance focal adhesion density and mechanosensitive signaling, while microscale structures provide anchoring points that support broader cell spreading and cytoskeletal organization. This hybrid design creates a topographical niche that is more effective at modulating mechanotransduction than either scale alone, highlighting that hybrid surfaces represent a strategic design innovation rather than an incremental variation.

Transcriptomic analysis further revealed how hybrid surfaces regulate macrophage behavior. Macrophages on hybrid surfaces upregulated genes related to focal adhesion, cytoskeletal remodeling, and actin polymerization, with enrichment of Rho/ROCK and actin-regulating pathways. Increased expression of integrins, focal adhesion proteins, Rho GTPases, and actin-nucleating factors highlights the central role of mechanotransduction in shaping macrophage phenotype and immune function.

This study had some limitations. First, we used simplified co-culture models that may not fully capture the complexity of in vivo environments. However, in vitro systems were intentionally used to enable mechanistic exploration of cell–material interactions under controlled conditions, allowing us to dissect early-stage biological responses with high resolution. Future animal studies will be important for validating our findings under physiological conditions and assessing long-term implant performance, including integration, remodeling, and host tolerance. Second, while we demonstrated multifunctionality, further enhancements could be achieved by incorporating additional features, such as i) on-demand actuation mechanisms to introduce dynamic topography or ii) incorporating chemical cues into the coatings to further boost antibacterial, immunomodulatory, or osteogenic activity. These additions are straightforward to implement without altering the overall system design. Lastly, we focused on early-stage outcomes such as polarization, adhesion, and gene expression—key determinants of downstream responses—but future work will explore longer-term effects, including adaptive immune responses and matrix deposition.

In summary, this work demonstrates the novelty of a hybrid nano–micro topographical system that uniquely integrates antibacterial, immunomodulatory, and osteogenic functions into a single platform. Unlike prior approaches that address these challenges separately, our design shows that uniquely combining nano- and microscale wrinkled features enables coordinated, multifunctional regulation of host and bacterial responses. By aligning physical design with biological complexity, we offer a blueprint for next-generation biomaterials that can coordinate balanced immune responses, prevent infection, and promote tissue integration. While this study uses bone tissue as a model, the underlying design principles are generalizable and adaptable to a range of implantable materials and devices. Future in vivo studies will be essential to validate long-term efficacy and advance clinical translation.

Experimental Section

Materials and Cell Lines:

Ti_3_AlC_2_ MAX phase powders (300 mesh, ≥99%) were obtained from Laizhou Kai Kai Ceramic Materials Co. Shrinkable polystyrene (PS) films were obtained from Grafix. Ultrapure deionized (DI) (18.2 MΩ, Mill-Q pore) water was used for most aqueous preparations. Unless specified otherwise, chemicals and reagents were purchased from Sigma-Aldrich or Thermo Fisher Scientific and used without further purification.

Cell lines used in this study, including MC3T3-E1 preosteoblasts and RAW 264.7 macrophages (RRID: CVCL_0493), were purchased from ATCC. MC3T3-E1 cells (RRID: CVCL_5440) were cultured in minimum essential medium α (αMEM), while RAW 264.7 macrophages were maintained in Dulbecco’s modified Eagle’s medium (DMEM), with both media supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin (Corning). Cells were incubated at 37 °C in a humidified environment with 5% CO_2_ and passaged upon reaching 80% confluence using 0.05% Trypsin-EDTA. All cell lines were routinely tested and confirmed to be free of mycoplasma contamination and were used at low passage numbers to ensure experimental consistency.

Synthesis of Ti3C2Tx:

Ti_3_C_2_T_x_ MXene nanosheets were synthesized using a modified in situ hydrofluoric acid etching approach. (Figure S10, Supporting Information) The synthesis involved etching 1.0 g of Ti_3_C_2_T_x_ MAX phase powder in a solution containing 7.5 g of lithium fluoride (LiF) and 40 mL of 9 m hydrochloric acid (HCl). The reaction mixture was continuously stirred at 35 °C for 24 h, facilitating the selective removal of aluminum layers. The obtained solid residue was then separated by centrifugation and washed multiple times with deionized water until the pH stabilized ≈6.0.

To achieve exfoliation, 100 mL of DI water was added to the collected residue, followed by bath sonication at 4 °C for 1 h. The resulting dispersion was centrifuged at 3500 rpm for 30 min to remove unexfoliated particles, yielding a stable suspension of Ti_3_C_2_T_x_ MXene nanosheets in the supernatant.

Fabrication of Wrinkled MXene Multilayers:

Both Gentamycin (GS) and Ti_3_C_2_T_x_ MXene sheets were diluted to a concentration of 1.0 mg mL^−1^ in deionized (DI) water, respectively, without any further adjustment. PS substrates were cleaned by sequential sonication in ethanol and water for 15 min each, dried with nitrogen, and treated with oxygen plasma (Tergeo-Plus, PIE Scientific) at 50 W under vacuum for 10 min. For the layer-by-layer (LbL) multilayer assembly, an automated dip coater (KSV NIMA, Biolin Scientific) was used with Ti_3_C_2_T_x_ MXene and GS solutions.

The plasma-treated substrates were first immersed in a poly(diallyldimethylammonium chloride) (PDAC) solution for 30 min to deposit a base layer, followed by two rinses in DI water for 2 min each. Bilayer multilayers were fabricated by alternately dipping the substrate into the cationic MXene solution for 5 min, followed by two rinse steps in DI water, and then into the anionic GS solution for 5 min, with the same rinse process. This cycle created one bilayer of (MXene/GS)1, with additional bilayers denoted as (MXene/GS)n, where n is the number of bilayers. The prepared multilayer films were air-dried and stored in a sealed container for future analysis. The MXene multilayers on PS substrates were then cut to the desired size and shrunk by heating at 130 °C for 5 min to form wrinkled films. The strain (ϵ) of the PS film was calculated using the formula ϵ = (Li – Lf)/Li, where Li and Lf represent the film’s initial and final lengths, respectively, and the wrinkled films were stored in a sealed container for subsequent analysis.

Fabrication of MXene Hybrid Structures:

After forming wrinkled nanostructures of (MXene/GS)3 coatings, the PS substrate was dissolved in dichloromethane (DCM) for 2 h, followed by rinsing in acetone for 15 min and ethanol for 5 min. The free-standing nanostructured MXene film was then carefully transferred onto a planar (MXene/GS)20 film. The combined film was air-dried at room temperature for a few minutes, followed by incubation at 130 °C for 1 h. During this process, the planar film shrank, forming wrinkled (MXene/GS)20 microstructures beneath while retaining the nanoscale features on the surface.

Materials Characterization:

Film thickness and surface roughness were measured using atomic force microscopy (AFM, Dimension Icon, Bruker Corp.) in tapping mode. Film deposition was measured using a UV–vis spectrophotometer (Agilent 8453). For AFM analysis, MXene multilayers deposited on glass substrates were scratched with a plastic razor blade, and thickness measurements were taken at three predefined locations. The surface morphology of the wrinkled structures was further examined using scanning electron microscopy (SEM, Tescan Mira3).

Immunostaining Assay for Macrophages:

RAW 264.7 macrophages were seeded onto the test samples at a density of 2 × 10^4^ cells and incubated for 24 h. After incubation, the cells were fixed and treated with a permeabilization/blocking buffer (PBS containing 5% normal goat serum and 0.3% Triton X-100) for 60 min. Once the buffer was removed, a cocktail of CD206 monoclonal antibody (MR5D3, rat/IgG2a, Invitrogen, 1:200) and iNOS monoclonal antibody (D6B6S, rabbit/IgG, Cell Signaling Technology, 1:200) in staining buffer (PBS with 1% BSA and 0.3% Triton X-100) was added. The cells were incubated overnight at 4 °C in the dark. The following day, cells were washed with PBS and stained with secondary antibodies—anti-rabbit IgG Alexa Fluor 488 (1:500) and anti-rat IgG Alexa Fluor 647 (1:500)—at room temperature for 1 h, and then stained with DAPI and examined using an upright fluorescence microscope (Axio Imager 2, Carl Zeiss Inc.). Macrophages stained immediately after attachment to tissue culture substrates were used to define the baseline population (M0 macrophages, serving as negative controls). For image cytometry analysis, positivity thresholds for iNOS and CD206 were set as the mean fluorescence intensity plus two standard deviations (mean + 2 × SD) of these negative controls.

Biosafety Assessment:

The cytocompatibility of topographical samples was evaluated using RAW 264.7 macrophages. For direct cytotoxicity tests, macrophages were seeded onto the samples and cultured for 1 or 5 days. Cell viability was quantified using the Cell Counting Kit-8 (CCK-8, Abcam) according to the manufacturer’s instructions. For extract cytocompatibility tests, topographical samples were incubated in complete culture medium at 37 °C for 14 days to generate extracts. The extracts were then collected, filtered, and applied to macrophages cultured under the same conditions as the direct tests. After 24 h of exposure, cell viability was measured using the CCK-8 assay.

Cell Morphological Observation:

Cells were seeded onto the films (uncoated and coated controls, nano, micro, and hybrid structures) at a density of 2–4 × 10^4^ cells for 24 h. Following incubation, the samples were washed with PBS and fixed in 4% paraformaldehyde for 15 min. After another wash with PBS, the samples were treated with 1% osmium tetroxide in PBS for 30 min. The fixed samples were then dehydrated using a series of ethanol concentrations (30%, 50%, 70%, 100%, and 100%), with each step lasting 5 min. Subsequently, the samples were dried using a critical point dryer (Leica EM CPD300) and sputter-coated with gold. Imaging was performed using a scanning electron microscope (SEM; Tescan MIRA3).

To analyze the cell cytoskeleton, cells were fixed with 4% paraformaldehyde for 15 min after 1 day of culture and subsequently permeabilized with 0.1% Triton X-100 in PBS for an additional 15 min, with multiple PBS washing steps in between. Following permeabilization, the cells were incubated with Alexa Fluor 488 Phalloidin and 1% bovine serum albumin (BSA) for 30 min at room temperature in the dark. After staining with DAPI, fluorescence microscopy was used to visualize both the cell nuclei and cytoskeleton. CellProfiler was then employed for further analysis.

Cell Proliferation Test for Preosteoblast Cells:

MC3T3-E1 cells were cultured on the test samples, which included uncoated and coated planar films, as well as nano-, microscale topographies, and hybrid structures, at a density of 2 × 10^4^ cells per sample for 1, 3, and 5 days. After incubation, the cells were fixed and stained with DAPI for fluorescence imaging. CellProfiler was then utilized to count the cells based on the number of stained nuclei.

ALP Assay:

To evaluate the osteogenic efficacy of the test samples, the ALP activity assay was conducted. MC3T3-E1 cells were seeded at a density of 2 × 10^4^ cells per group and incubated for 7 and 14 days in an osteogenic induction medium comprising growth medium supplemented with 50 μm L-ascorbic acid, 10 mm glycerol phosphate, and 50 nm dexamethasone. After the incubation periods, the cells were washed with PBS, trypsinized, and transferred to a new 24-well plate. Following a brief incubation to allow for cell adhesion, the cells were fixed. The ALP activity was subsequently measured using an ALP Colorimetric Assay kit (Abcam), following the manufacturer’s instructions for quantifying the enzyme activity.

Immunoregulation of Osteogenesis:

RAW 264.7 macrophages were cultured on various samples for 48 h to collect their supernatants. To investigate the regulatory effects of these macrophages on the osteogenesis of MC3T3 preosteoblasts, MC3T3 cells were subsequently cultured on the same samples. The MC3T3 cells were maintained in the cell culture medium, supplemented with macrophage supernatants collected from the corresponding sample at a 1:1 ratio. ALP assays were then performed to assess osteogenic activity.

Quantitative Polymerase Chain Reaction (qPCR) Analysis:

Total RNA was extracted from the cells using TRIzol (Invitrogen) and subsequently reverse-transcribed into cDNA with the PrimeScript RT Master Mix (Takara Bio), using the manufacturer’s instructions. Real-time quantitative PCR was conducted on a Quant Studio 3 RT-PCR system, using glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as the internal reference. The target genes included osteogenic factors such as Runx2, ALP, and OCN for preosteoblast cells, as well as M1 markers (IL-1β and TNF-α), M2 markers (IL-4 and IL-10), and immune osteogenesis-related markers (BMP-2 and OSM). A summary of the primers used is provided in Table S4 (Supporting Information).

Antibacterial Assay:

Bacterial cells were first grown in Tryptic Soy medium (for S. aureus) or LB medium (for E.coli) overnight at 37 °C with shaking at 200 rpm. The overnight cultures were then diluted to approximately 10^4^ CFU mL^−1^ and inoculated onto the surfaces to promote biofilm growth. The surfaces were incubated at 37 °C for 24 h for MRSA and 48 h for E. coli.

To quantify the attached cells, the biofilm surfaces were washed three times with deionized (DI) water and then stained with a mixture of SYTO9 (5 μm) and propidium iodide (PI) (20 μm) for 15 min at room temperature in the dark. The surfaces were subsequently imaged using fluorescence microscopy. Image analyses were conducted using ImageJ and CellProfiler, using the imaging channels to count live and dead cells.

For the spread plate method (SPM), cells were seeded onto various samples and incubated to form biofilms. Following this, samples were washed 3 times with PBS, and 1 mL of PBS was added to each sample, and the samples were vortexed and briefly sonicated to detach the cells from the samples. The resulting solutions were then diluted and plated onto agar plates to quantify the number of colonies formed.

Macrophages Phagocytosis of Adhered Bacterial Cells:

S. aureus and E. coli were cultured and seeded onto different samples in a 24-well plate. Subsequently, 2 × 10^4^ macrophages were added to each well containing DMEM supplemented with 5% fetal bovine serum (FBS) and incubated at 37 °C with 5% CO_2_ for 2 h. Control samples were treated with DMEM containing 5% FBS and incubated under identical conditions. Following co-culture, extracellular bacteria were removed by treating the samples with gentamicin antibiotics (200 μg mL^−1^) in PBS for 1 h at 37 °C, allowing for selective quantification of phagocytosed bacteria. After antibiotic treatment, DAPI staining was performed, IF imaged, and phagocytosis was assessed using CellProfiler. Samples were also processed for SEM using previously described protocols in the “Cell Morphological Observation” experimental section.

Transcriptome Sequencing and Data Analysis:

Macrophage cell suspensions (2 × 10^5^ cells) were cultured with different samples in a well plate for 48 h. Cells were then lysed using Zymo columns (Zymo Research), and the lysates were stored at −80 °C for subsequent Poly-A-enriched RNA-seq library construction via TruSeq kit. Each library was sequenced on NovaSeqXPlus (Illumina, USA) for 151 bp paired-end sequencing of 48 million to 78 million reads per sample (Table S5, Supporting Information). Reads were mapped to the reference genome GRCm38 (ENSEMBL 102), using STAR v2.7.8a^[52]^ and assigned count estimates to genes with RSEM v1.3.3.^[53]^ Differential genes were identified via DESeq2 v1.42.1^[54]^ library using R in v4.3.1. Performing the DESeqDataSetFromMatrix function resulted in a p-value assigned to each gene. For any pairs of samples, upregulated genes were identified as those with log_2_fold change > 1 and adjusted p-value < 0.1, while down-regulated genes had log_2_fold-change < −1 and adjusted p-value < 0.1. Gene expression values were transformed to log_10_ [cTPM (Transcripts Per Million reads) + 1]. All analyses were done using R programming and free online platforms (http://bioinformatics.sdstate.edu/go77/ and https://biit.cs.ut.ee/gprofiler/gost).

Statistics:

All statistical analyses and data visualization were conducted using GraphPad Prism 10. Correlation assessments were performed using linear least-squares regression at a 95% confidence level, with the Pearson correlation coefficient (r) quantifying relationships between variables. Group comparisons were carried out using a nonparametric t-test for two-group analyses, while analysis of variance (ANOVA) followed by post hoc testing was applied for comparisons involving more than two groups. All statistical tests were two-tailed, and significance was defined as p < 0.05.

Supplementary Material

SI

Supporting Information is available from the Wiley Online Library or from the author.