Abstract

   Skin wound healing is a dynamic process, yet scaffolds enabling
   stage-specific modulation remain limited. We fabricated a nanofiber
   scaffold from FDA-approved materials, consisting of two outer layers of
   radially aligned and random poly(ε-caprolactone) fibers and a middle
   layer of electrosprayed phase-change microparticles loaded with
   platelet-derived growth factor–BB (PDGF-BB)/vascular endothelial growth
   factor (VEGF) in the periphery and PDGF-BB/epidermal growth factor
   (EGF) in the center. Near-infrared irradiation through a photomask
   enabled spatiotemporal control of growth factor release, aligning
   PDGF-BB, VEGF, and EGF delivery with specific phases of wound healing
   to promote vascularization, cell proliferation, and tissue remodeling.
   In a preclinical porcine model, it enhanced closure and modulated the
   microenvironment by activating PI3K-Akt, MAPK, and immune pathways,
   up-regulating genes for survival and repair while down-regulating those
   linked to apoptosis and inflammation. With scalable manufacturing and
   large-animal efficacy, this scaffold holds translational potential for
   skin wound healing.
     __________________________________________________________________

   Nanofiber scaffolds enable spatiotemporal release of bioactive
   effectors to promote skin wound healing.

INTRODUCTION

   Skin wounds represent a major clinical challenge, and rapid and
   effective healing is essential to reduce infection, minimize scarring,
   and restore function ([40]1, [41]2). Traditional treatments, such as
   autografts, face limitations including donor site morbidity and immune
   rejection ([42]3–[43]5). Synthetic scaffolds have emerged to address
   these issues ([44]6, [45]7). For example, Restrata, comprising
   electrospun poly(lactic-co-glycolic acid) and polydioxanone nanofibers,
   have shown efficacy in the clinical setting by providing a structural
   support ([46]8, [47]9). These scaffolds, however, primarily serve a
   passive protection role and lack the ability to dynamically regulate
   the complex biological processes involved in skin regeneration ([48]10,
   [49]11). Consequently, their use often results in suboptimal healing
   outcomes and substantial scar formation ([50]12).

   Addressing these limitations necessitates the development of
   multifunctional, bioactive scaffolds capable of dynamically adapting
   their functionality with the overlapping yet distinct phases of skin
   repair: hemostasis and inflammation, proliferation, and tissue
   remodeling ([51]13, [52]14). Each phase is characterized by specific
   cellular activities and extracellular matrix (ECM) interactions with
   growth factors, which collectively determine the success of skin
   regeneration ([53]15). In the early stage of wound healing, vascular
   regeneration plays a critical role, indirectly influencing neural
   network regeneration and modulating inflammatory responses ([54]16).
   Supplementing exogenous growth factors, such as platelet-derived growth
   factor (PDGF) and vascular endothelial growth factor (VEGF), is
   essential at this stage ([55]17, [56]18). During the proliferative
   phase, various types of cells, such as fibroblasts, epithelial cells,
   and keratinocytes, migrate and proliferate, as supported by ongoing
   vascular regeneration ([57]19). Growth factors like PDGF and epidermal
   growth factor (EGF), which enhance the migration and proliferation of
   these cells, are particularly important during this phase ([58]20,
   [59]21). In the tissue remodeling stage, the reorganization of collagen
   and deposition of ECM lead to the maturation of regenerated skin tissue
   ([60]22). Achieving a precise control over the timing and location of
   the release of these factors from the scaffolds is essential for
   effective skin tissue regeneration.

   As for the structure of the scaffold, electrospun nanofibers have
   gained considerable attention for their ECM-like properties, as
   demonstrated by the clinical products ([61]23–[62]26). Radially aligned
   fibers, in particular, can provide topographical cues that promote cell
   migration from the periphery to the center, holding great promise for
   wound healing ([63]27, [64]28). Now, bioactive substances are commonly
   loaded into the fibers using coaxial or multiple-axial electrospinning,
   with their release governed by diffusion and material degradation
   ([65]29, [66]30). This approach often results in an initial burst
   release, followed by minimal or no release in later stages, greatly
   undermining sustained tissue regeneration and raising concerns about
   potential side effects due to high initial dosages. Stimulus-responsive
   nanofiber scaffolds that respond to internal cues such as reactive
   oxygen species ([67]31, [68]32) or external triggers ([69]33–[70]35)
   offer a promising strategy for controlled release of bioactive
   molecules. For example, a chitosan-based nanofiber dressing responsive
   to pH and glucose levels promoted wound healing in diabetic rats
   ([71]32). Additionally, physical therapies such as electric fields
   ([72]34), ultrasound ([73]33), and light have been applied in
   combination with nanofiber scaffolds to trigger the release of growth
   factors. Typically, near-infrared (NIR) light, a light source used in
   clinical treatments, has been used in photothermal therapy, including
   both photothermally triggered release and mild hypothermal effect
   ([74]35–[75]37). In one study, the release of bone morphogenetic
   protein–4 from polydopamine nanoparticles was triggered using NIR light
   during the proliferative stage of skin wound healing to help inhibit
   scar formation ([76]38). In addition, photothermal therapy is commonly
   used to treat skin wounds infected with drug-resistant bacteria and
   diabetic ulcer treatment ([77]39). The mild hypothermal effect can
   effectively eliminate bacteria at a specific temperature, reduce
   inflammation, and promote wound healing ([78]40, [79]41). Among various
   exogenous stimuli, NIR light offers advantages for cutaneous
   applications, including noncontact operation and precise spatial
   activation, and deep tissue penetration without harming surrounding
   ([80]42). However, all these strategies are limited to partially
   controlled release for only one stage of skin repair, making it
   difficult to achieve stage-specific, targeted treatments.

   The integration of NIR light with thermosensitive materials provides a
   versatile platform for spatiotemporal drug delivery. Thermosensitive
   phase change materials (PCMs), such as fatty acid–based mixtures,
   undergo solid-to-liquid phase transitions upon photothermal activation,
   enabling the controlled release of encapsulated bioactive substances at
   specific temperatures ([81]43, [82]44). For instance, the eutectic
   mixture of lauric acid and stearic acid, with a melting point tailored
   to the physiological temperature (e.g., 39.5°C), offers a precise
   control of temperature-responsive release without causing thermal
   damage to the surrounding tissues ([83]45). Prior studies have used a
   combination of electrospraying and electrospinning techniques to
   fabricate nanofiber scaffolds embedded with PCM microparticles to
   efficiently encapsulate and release therapeutic agents ([84]46).
   NIR-induced photothermal activation, in conjunction with a photomask,
   triggered the localized melting of PCMs, enabling the targeted release
   of therapeutic agents and thereby enhancing cellular migration
   ([85]47). Despite these promising results, the in vivo efficacy of
   these systems remains to be explored. Additionally, the electrospraying
   process allows for a precise control of the spatial locations of
   various microparticles on specific regions of a scaffold using a
   physical mask. We hypothesize that such a system can dynamically
   regulate the release of different growth factors in response to NIR
   stimuli, thereby optimizing the microenvironment at various stages of
   the healing process.

   Here, we developed a nanofiber scaffold by incorporating multiple cues
   to spatiotemporally promote skin wound healing. These cues include an
   aligned topographical structure for directing cell migration, bioactive
   growth factors with defined concentrations to regulate cell behavior at
   specific stages, and mild photothermal effect to modulate the
   biological microenvironment. The scaffold was fabricated using a
   combination of electrospinning and masked coaxial electrospraying
   techniques ([86]Fig. 1). The resulting scaffold, with a diameter of 2.5
   cm, consists of three layers: an inner layer of radially aligned
   poly(ε-caprolactone) (PCL) nanofibers, a middle layer of growth
   factor–encapsulated PCM microparticles, and an outer layer of randomly
   oriented PCL nanofibers. The microparticles were deposited on the
   nanofibers by sequentially using circular and annular masks during
   electrospraying, creating two concentric regions within the middle
   layer: The inner circular region (1.875 cm in diameter) contained PCM
   microparticles loaded with PDGF-BB, EGF, and indocyanine green (ICG),
   while the outer annular region contained PCM microparticles loaded with
   PDGF-BB, VEGF, and ICG. ICG, a Food and Drug Administration
   (FDA)–approved photothermal agent, was included as the photothermal
   agent. This scaffold was denoted PCM-P/V/E to indicate the
   encapsulation of PDGF-BB, VEGF, and EGF. Spatiotemporally controlled
   release of these growth factors was achieved through periodic NIR
   irradiation using three size-adjustable photomasks, allowing for
   precise and localized delivery of the encapsulated effectors.
   Full-thickness skin injury models based on small animals (rabbit) were
   used to evaluate the scaffold’s in vivo efficacy, while large-animal
   (porcine) studies provided further insights into the mechanisms of skin
   tissue repair and critical data to support the promise for clinical
   translation. Notably, this study used FDA-approved, clinically
   validated raw materials, combined with a scalable one-step fabrication
   process, to develop a skin tissue regeneration scaffold optimized for
   clinical application. The used photothermal therapy features a
   lightweight, user-friendly device that can be self-administered by
   patients under medical guidance. The proposed scaffold offers a highly
   innovative therapeutic approach for patients with severe skin injuries,
   addressing the urgent need for advanced wound repair solutions.

Fig. 1. Schematic illustration of a bioactive nanofiber scaffold combined
with NIR irradiation for accelerated skin wound healing.

   [87]Fig. 1.
   [88]Open in a new tab

   The scaffold is fabricated using electrospinning and mask-assisted
   coaxial electrospraying. It consists of three layers: the inner layer,
   made of radially aligned nanofibers; the outer layer, composed of
   randomly oriented nanofibers; and a middle layer containing
   phase-change microparticles. These microparticles are loaded with
   different combinations of growth factors: The peripheral region
   contains PDGF-BB and VEGF, while the central region contains PDGF-BB
   and EGF. NIR light and a size-adjustable photomask are applied at
   specific time points, enabling spatiotemporally controlled release of
   growth factors, thereby creating a favorable environment for skin
   regeneration and functional recovery. D, the outer diameter of region
   B; d, the outer diameter of region A.

RESULTS

Characterizations of the nanofiber scaffolds

   The PCM-P/V/E scaffold composed of two layers of nanofibers sandwiched
   with PCM microparticles. The radially aligned nanofibers, with an
   average diameter of 489 ± 90 nm, extended from the center of the
   scaffold toward the periphery, providing topographical cues for cell
   migration and cytoskeletal alignment ([89]Fig. 2A and fig. S1A). The
   randomly oriented nanofibers, with an average diameter of 394 ± 72 nm,
   were added to enhance the mechanical integrity of the scaffold,
   allowing it to withstand the forces typically encountered during
   surgical suturing ([90]Fig. 2B and fig. S1B). The PCM microparticles
   were uniformly dispersed throughout the scaffold ([91]Fig. 2C). These
   microparticles, being larger than the pore size formed by the
   physically stacked fibers, were effectively entrapped between the two
   nanofiber layers. Following oxygen plasma treatment, the water contact
   angle of the scaffold was 35°, indicating a hydrophilic surface (fig.
   S2). In mechanical testing, the scaffold demonstrated a tensile
   strength of 11.4 MPa and an elongation at break of 105.3%, reflecting
   favorable mechanical properties ([92]Fig. 2D). During surgical
   handling, the scaffold maintained structural integrity and was able to
   endure the forces applied during suturing.

Fig. 2. In vitro physicochemical characterization of the scaffold.

   [93]Fig. 2.
   [94]Open in a new tab

   Scanning electron microscopy (SEM) images of the (A) radially aligned
   nanofibers and (B) randomly oriented nanofibers at low and high
   magnifications, respectively. Insets: Corresponding two-dimensional
   fast Fourier transform images derived from the SEM images. (C) SEM
   image of PCM microparticles distributed on the radially aligned
   nanofibers. (D) Stress-strain curve of the scaffold. (E) Temperature
   changes of the scaffold upon irradiation with an 808-nm NIR laser at
   varying power densities. (F) Temperature changes of the scaffold over
   five cycles of 2-min NIR irradiation. (G) Temperature changes of the
   scaffold over four cycles of 10-min NIR irradiation. (H) Schematic
   illustration of the photomask strategy for selective NIR irradiation of
   designated regions of the scaffold. (I to K) Schematic illustrations
   and corresponding temperature profiles of the scaffold in exposed and
   photomask-covered regions during (I) round 1, (J) round 2, and (K)
   round 3 of NIR irradiation, respectively. Insets: Infrared thermal
   images showing the temperature distribution of exposed and masked
   regions. (L) Total amounts and (M) cumulative release profiles of
   PDGF-BB, VEGF, and EGF from the scaffold following sequentially, masked
   NIR irradiation.

Stable photothermal effect of the nanofiber scaffold upon NIR irradiation

   The nanofiber scaffold exhibited a stable photothermal effect by
   encapsulating ICG within the PCM microparticles. Upon irradiation with
   an 808-nm NIR laser at varying power densities, we monitored the
   temperature changes using an infrared thermal imaging camera. As shown
   in [95]Fig. 2E, when irradiated at power densities of 0.6 and 0.7
   W/cm^2, the temperatures of the scaffold increased and eventually
   stabilized at 42.6° and 45.0°C, respectively, both above the melting
   point (39°C) of the PCM, which could induce a solid-to-liquid phase
   transition (fig. S3). Therefore, we selected a power density in the
   range of 0.6 to 0.7 W/cm^2 for subsequent experiments to ensure that
   the temperature exceeded the melting point of PCM without causing
   heat-induced damage to the surrounding tissue.

   To assess the stability of the photothermal effect, we irradiated the
   scaffold for multiple cycles with varied durations. As shown in
   [96]Fig. 2F, the temperature rapidly increased from room temperature to
   39.6° to 43.8°C upon irradiation by the NIR laser. After 2 min of
   irradiation, the laser was turned off, and the scaffold quickly
   returned to room temperature. This cycle was repeated for an additional
   four times, with the temperature response remaining consistent
   throughout. To better simulate the in vivo conditions, we extended the
   irradiation duration to 10 min per cycle and monitored the temperature
   changes over time. As shown in [97]Fig. 2G, upon laser irradiation, the
   temperature increased rapidly and stabilized within the range of 39.2°
   to 41.2°C. After 10 min of irradiation, the scaffold cooled down
   quickly once the laser was turned off. This process was repeated three
   additional cycles, during which the temperature remained within the
   desired range. These results demonstrate that the scaffold maintained a
   stable photothermal effect over multiple irradiation cycles. The good
   stability could be attributed to the temperature-regulating properties
   of PCM, which absorbed latent heat during the phase transition, thereby
   maintaining the scaffold temperature within the desired range.

NIR laser–triggered, spatiotemporally controlled release of specific growth
factors from the nanofiber scaffold

   We used a photomask strategy to selectively expose different regions of
   the scaffold to NIR laser irradiation, triggering the PCM
   microparticles to melt and thereby release the encapsulated growth
   factors in the irradiated regions ([98]Fig. 2H). During the first three
   rounds of irradiation, we used three circular photomasks with diameters
   of 2, 1.5, and 0.5 cm, respectively, to sequentially expose region I,
   regions I + II, and regions I + II + III to the laser. In the fourth
   round, the entire scaffold was exposed to the laser. During each round
   of irradiation, the temperature in the exposed region increased to 39°
   to 42°C, while the temperature in the photomask-covered regions only
   showed a slight increase ([99]Fig. 2, I to K). To validate the
   feasibility of spatiotemporally controlling the release of growth
   factors, we encapsulated two model substances, trypan blue and
   rhodamine B, within the PCM microparticles in the peripheral and
   central regions of the scaffold, respectively (fig. S4). As shown in
   fig. S4A, the microparticles were exactly deposited in the designated
   regions. Upon NIR irradiation with corresponding photomasks, both
   trypan blue and rhodamine B were sequentially released from the
   scaffold (figs. S5 to S7).

   We then used the masking strategy to irradiate different regions of the
   PCM-P/V/E scaffold for 10 min per week from week 0 to week 3 to
   evaluate the release profile of growth factors from the scaffold. In
   the first round of NIR irradiation, region I, which contained PCM
   microparticles loaded with PDGF-BB and VEGF, was exposed. The amounts
   of PDGF-BB and VEGF released were 36.1 ± 1.2 ng and 35.9 ± 1.1 ng,
   respectively ([100]Fig. 2L). In the second round, regions I + II
   (containing PDGF-BB, VEGF, and EGF) were exposed, resulting in the
   release of 29.0 ± 1.8 ng of PDGF-BB, 23.1 ± 1.5 ng of VEGF, and
   20.9 ± 1.1 ng of EGF. During the third round, regions I + II + III were
   exposed, leading to the release of 32.0 ± 1.3 ng of PDGF-BB and
   35.8 ± 1.5 ng of EGF. Last, when the entire scaffold was exposed to the
   laser, 10.4 ± 1.9 ng of PDGF-BB and 7.5 ± 0.4 ng of EGF were released.
   These results demonstrate that NIR laser irradiation combined with a
   photomask enabled spatiotemporal controls of growth factor release from
   specific regions of the scaffold, with sufficient amounts to induce a
   biological effect while avoiding excessive or uncontrolled delivery. We
   further quantified the cumulative release of the three growth factors
   over the course of the experiment. As shown in [101]Fig. 2M, the
   release of PDGF-BB showed a continuous upward trend, with a cumulative
   release amount of 107.5 ± 0.8 ng. VEGF was predominantly released from
   regions I + II during the initial first round and the first week, with
   a total release of 59.0 ± 1.8 ng. EGF began releasing from the first
   week and continued through week 3, with a cumulative release of
   64.2 ± 0.9 ng. These results confirm the ability to achieve NIR
   laser–triggered, spatiotemporally controlled release of specific growth
   factors from the scaffold.

Manipulation of cell migration and proliferation by the scaffolds

   To evaluate the impacts of the spatiotemporally delivered growth
   factors on cell behaviors, we fabricated additional control scaffolds:
   PCM-P (with only PDGF-BB and ICG), PCM-P/E (with PDGF-BB, EGF, and
   ICG), a pristine fiber scaffold (Pristine scaffold), and a scaffold
   only with PCM microparticles (PCM group). The biological activity of
   the released PDGF-BB and VEGF was evaluated through an in vitro
   angiogenesis assay by culturing human umbilical vein endothelial cells
   (HUVECs) on Matrigel matrix in the different culture medium. Growth
   factors released from the PCM-P, PCM-P/E, and PCM-P/V/E scaffolds were
   collected and added to the culture medium to induce angiogenesis in
   HUVECs. These groups were named Released-P, Released-P/E, and
   Released-P/V/E, respectively, with HUVECs cultured in plain medium on
   the tissue culture plate (TCP) serving as the Blank control group. For
   comparison, free PDGF-BB (100 ng/mL) or VEGF (60 ng/mL) was added at a
   concentration similar to the released amount to serve as controls,
   referred to as Free-P and Free-V, respectively. After culturing for 6
   hours, as shown in [102]Fig. 3A, the Blank group exhibited limited
   branching, with HUVECs predominantly adopting a flattened morphology
   and displaying filopodia. In contrast, the addition of free PDGF-BB or
   VEGF significantly promoted vascular network formation, with increased
   numbers of branch points and tubules, indicating angiogenic activity
   ([103]Fig. 3, B and C). The Released-P and Released-P/E groups showed
   similar angiogenesis to the Free-P group, with comparable branch points
   and tubule formation. The Released-P/V/E group, however, demonstrated
   more extensive vascular structures, including well-formed walls and
   lumens, and showed significantly increased numbers of branch points and
   tubules compared to the Released-P and Released-P/E groups. These
   results confirmed that the growth factors released from the scaffolds
   retained the biological activity.

Fig. 3. Manipulation of cell migration and proliferation by the scaffolds
under NIR irradiation.

   [104]Fig. 3.
   [105]Open in a new tab

   (A) Fluorescence micrographs showing angiogenesis of HUVECs cultured in
   different media: plain culture medium (Blank), medium supplemented with
   free PDGF-BB (Free-P) or free VEGF (Free-V), and medium containing
   growth factors released from scaffolds (Released-P, Released-P/E, and
   Released-P/V/E). Angiogenic structures were stained with calcein-AM
   (green). Quantification of (B) branch points and (C) number of tubules,
   respectively, formed by HUVECs in each group. (D) Optical density (OD)
   values of HUVECs cultured on different scaffolds at days 1, 3, 5, and
   7. *P < 0.05, the blank horizontal lines represent **P < 0.01. (E)
   Fluorescence micrographs of live/dead staining of HUVECs cultured on
   different scaffolds at days (d) 1 and 3. Live cells were stained with
   calcein-AM (green) and dead cells with propidium iodide (red),
   respectively. (F) Fluorescence micrographs showing the radial migration
   of L929, NIH-3T3, and HaCaT cells from the periphery toward the center
   of the scaffolds after 4 days. White dashed lines indicate the initial
   boundary at the periphery of the scaffold. F-actin was stained with
   Alexa Fluor 555 phalloidin (red) and nuclei with
   4′,6-diamidino-2-phenylindole (DAPI) (blue), respectively. (G to I)
   Quantitative analysis of the average and the farthest migration
   distances of (G) L929, (H) NIH-3T3, and (I) HaCaT cells, respectively,
   in different groups. **P < 0.01 versus all other groups; ^##P < 0.01
   versus other five groups.

   We further assessed the biocompatibility of the different scaffolds
   using the Cell Counting Kit-8 assay. NIR laser irradiation was applied
   to the PCM-P, PCM-P/E, and PCM-P/V/E scaffolds on days 0, 2, 4, and 6,
   using different photomasks to achieve spatially selective masked
   irradiation (MI). For comparison, a single irradiation (SI) was
   performed on the entire surface of the PCM and PCM-P scaffolds without
   masking. As shown in [106]Fig. 3D, the optical density (OD) values for
   all groups progressively increased from day 1 to day 7, confirming the
   good biocompatibility of all scaffolds. Notably, on day 7, cell
   proliferation in the PCM-P/V/E (MI) group was significantly higher than
   in the other groups, which can be attributed to the spatiotemporally
   delivered PDGF-BB, VEGF, and EGF. [107]Figure 3E presents the results
   of live/dead staining of HUVECs cultured on the scaffolds for 1 and 3
   days. A predominance of live cells (green fluorescence) with minimal
   dead cells (red fluorescence) was observed across all groups. Notably,
   the PCM-P/V/E (MI) group exhibited the highest density of live cells,
   indicating superior cytocompatibility and cell-supportive properties.

   Following skin injury, the recruitment and directed migration of
   fibroblasts and epidermal cells from the periphery of the wound toward
   the center are essential processes for effective wound healing. To
   replicate this behavior in vitro, the migratory response of L929 and
   NIH-3 T3 fibroblasts, along with human keratinocyte HaCaT cells, was
   assessed on the scaffolds over a 4-day period. A “wound” region was
   created by placing a polydimethylsiloxane disc (2 cm in diameter) at
   the center of the scaffold, and cells were then seeded
   circumferentially around the scaffold’s periphery. In the MI groups,
   NIR laser exposure was performed on days 0, 1, 2, and 3 using
   photomasks to spatially control the regions being irradiated. As shown
   in [108]Fig. 3F and fig. S8, cells cultured on TCP exhibited
   disorganized morphologies and minimal migration into the central
   region. In contrast, on the nanofiber scaffolds, L929 and NIH-3T3 cells
   displayed elongated morphologies aligned with the fiber orientation,
   while HaCaT cells appeared more densely clustered. All three cell types
   displayed greater migratory capacity on the scaffolds relative to TCP.
   Notably, the PCM-P (MI) group exhibited a larger migration area than
   the PCM-P (SI) group, indicating that multiple cycles of spatially
   controlled PDGF-BB release effectively sustained the promigratory
   effect. Among all groups, the PCM-P/V/E (MI) group induced the most
   extensive cell migration within the wound region. Quantitative analysis
   of L929 cell migration distances ([109]Fig. 3G) revealed a consistent
   trend, where both the average and the farthest migration distances
   followed the order: TCP < Pristine scaffold < PCM (SI) <  PCM-P
   (SI) < PCM-P (MI) < PCM-P/E (MI) < PCM-P/V/E (MI). Specifically, the
   average migration distances of the PCM-P (MI) and PCM-P/V/E (MI) groups
   were 379.2 ± 9.8 and 502.5 ± 4.0 μm, corresponding to 2.2- and 4.3-fold
   increases, respectively, relative to the Pristine scaffold group. The
   farthest migration distances in these groups were 883.3 ± 54.4 and
   1140.0 ± 35.6 μm, representing 1.6- and 2.4-fold increases,
   respectively. Similarly, both NIH-3T3 fibroblasts and HaCaT
   keratinocytes exhibited significantly enhanced migration in the
   PCM-P/V/E (MI) group. Specifically, the average migration distance
   increased by 3.3-fold for NIH-3T3 cells and 2.1-fold for HaCaT cells,
   respectively, compared to the Pristine scaffold group ([110]Fig. 3, H
   and I). Collectively, these results demonstrate that the integration of
   radially aligned nanofibers with spatiotemporally delivered PDGF-BB,
   VEGF, and EGF synergistically enhanced cell migration, a critical
   determinant for promoting efficient wound closure.

The scaffold promotes rabbit full-thickness skin wound healing upon NIR
irradiation

   In vivo experiments were conducted to evaluate the therapeutic efficacy
   of the scaffolds in promoting full-thickness skin wound healing.
   Circular wounds (2.5 cm in diameter) were created on the dorsal surface
   of New Zealand white rabbits, followed by suturing of the scaffolds
   into place (fig. S9). Six scaffold groups (i.e., those used in cell
   studies) were tested to evaluate the distinct functions of the cues:
   Pristine scaffold, PCM (SI), PCM-P (SI), PCM-P (MI), PCM-P/E (MI), and
   PCM-P/V/E (MI). An untreated defect group without scaffold implantation
   served as the Blank control. As shown in [111]Fig. 4A, for the MI
   groups, NIR irradiation was applied for 10 min once per week over a
   4-week period using three distinct photomasks. During irradiation, the
   temperature changes were monitored using an infrared thermal camera.
   The temperature in the irradiated region increased above 39°C within 90
   s and remained in the range of 39° to 42°C during all four rounds of
   irradiation (fig. S10). This controlled mild heating could also promote
   vascular regeneration and wound healing without causing thermal damage
   to the tissue, as supported by previous studies ([112]48, [113]49).
   Photographs of the wound site were taken at each time point to visually
   assess the wound closure, and, at week 4 postsurgery, skin tissue
   samples were collected ([114]Fig. 4B).

Fig. 4. In vivo evaluation of full-thickness skin wound healing in rabbits
under NIR irradiation.

   [115]Fig. 4.
   [116]Open in a new tab

   (A) Schematic illustration of the masked NIR irradiation protocol
   applied at weeks 0, 1, 2, and 3 postsurgery. The inset shows infrared
   thermal images of the scaffold, indicating a stable temperature range
   of 39° to 42°C during continuous irradiation. (B) Representative
   macroscopic images of wound healing in each group over a 4-week period.
   (C) Pie charts showing the percentage of wound closure in different
   groups. (D) Hematoxylin and eosin (H&E) staining and (E) Sirius red
   staining micrographs of regenerated skin tissues from different groups
   at 4 weeks postsurgery.

   At the early stage of healing (week 1), scab formation was observed in
   all groups. The Pristine scaffold group showed no significant
   improvement in wound closure compared to the Blank control, indicating
   limited bioactivity ([117]Fig. 4C and fig. S11). The PCM (SI) group
   exhibited a greater wound closure percentage of 33.6 ± 1.0%. In the
   PCM-P (SI) group, the percentage of wound closure was 1.3-fold higher
   than that of the PCM (SI) group, indicating the additional contribution
   of PDGF-BB delivery in promoting wound healing. The PCM-P (MI) group
   achieved a wound closure percentage of 63.6 ± 1.2%. However, no
   statistically significant differences in wound healing rates were
   observed between the PCM-P (MI), PCM-P/E (MI), and PCM-P/V/E (MI)
   groups at this early time point, suggesting that the effects of
   additional EGF and VEGF incorporation may require a longer period to
   manifest in measurable macroscopic wound closure.

   The second and third weeks postsurgery represented the peak phase of
   wound healing, characterized by rapid reduction in wound size.
   Specifically, at week 2, the Blank group exhibited minimal wound
   reduction, with signs of suppuration. The wound closure percentages in
   the Pristine scaffold, PCM (SI), PCM-P (SI), PCM-P (MI), PCM-P/E (MI),
   and PCM-P/V/E (MI) groups were 53.8 ± 1.0%, 84.3 ± 0.9%, 89.9 ± 1.1%,
   90.3 ± 1.8%, 92.6 ± 0.9%, and 95.4 ± 0.9%, respectively. Notably, the
   PCM-P/V/E (MI) scaffold obviously accelerated wound healing, achieving
   a 1.8-fold improvement compared to the Pristine scaffold group,
   indicating faster healing of a large wound. In addition, scaffolds and
   regenerated tissue from the PCM-P (SI) and PCM-P (MI) groups were
   collected for scanning electron microscopy (SEM) analysis (fig. S12).
   The fibrous morphology of the scaffolds was still visible, and the
   surrounding tissue was well integrated with the scaffold. By week 3
   postsurgery, notable wound size reduction was observed across all
   groups. The Blank, Pristine scaffold, and PCM (SI) groups still showed
   wound areas of 14.1 ± 1.3%, 9.8 ± 1.1%, and 3.7 ± 1.0%, respectively.
   For the PCM-P (SI) and PCM-P (MI) groups involving PDGF-BB, the SI
   group had a residual wound area of 2.8 ± 0.7%, while the MI group
   showed complete wound closure, indicating that multiple rounds of
   PDGF-BB delivery were more effective than a one-time delivery. Notably,
   the wounds in the three MI groups were fully healed, and the PCM-P/V/E
   (MI) group not only achieved a complete wound closure but also
   exhibited smooth, intact regenerated skin resembling healthy tissue,
   indicating a scarless healing process. By week 4, complete tissue
   coverage was achieved in all groups, with the multicue scaffold showing
   substantially enhanced regenerative outcomes.

   Hematoxylin and eosin (H&E) staining was performed on the regenerated
   skin tissue 4 weeks postsurgery to evaluate the morphology of the
   epidermis, dermis, and appendages ([118]Fig. 4D). In the Blank group,
   the epidermis was incomplete. The Pristine scaffold group demonstrated
   complete epidermal regeneration, although focal erosion was still
   evident. In the PCM (SI) group, an epidermal covering had formed, but
   partial separation between the epidermal and dermal layers was
   observed. In contrast, the four groups with growth factor
   delivery—PCM-P (SI), PCM-P (MI), PCM-P/E (MI), PCM-P/V/E (MI)—exhibited
   a four-layered epidermis, including stratum corneum, granular layer,
   spinous layer, and basal layer. Notably, in the MI groups, the
   epidermis was intact, and, beneath the epidermis, organized tissue
   structures were observed. The PCM-P/V/E (MI) group exhibited the most
   compact and well-ordered tissue arrangement. Dermal regeneration and
   appendage formation were next evaluated. In the Blank group,
   inflammatory cells were still present, and few appendages were
   observed. The Pristine scaffold group showed abundant proliferating
   fibrous connective tissue. The regenerated dermis in the PCM (SI) group
   was loosely arranged, reflecting incomplete tissue regeneration. In the
   PCM-P (SI) and PCM-P (MI) groups, the dermis was more organized, and a
   higher number of appendages were regenerated. The most prominent
   results were observed in the PCM-P/E (MI) and PCM-P/V/E (MI) groups,
   where the dermis was highly organized, and numerous skin appendages
   were present. These groups achieved not only structural integrity but
   also restored skin function.

   Collagen types I and III are essential for maintaining the structural
   integrity, strength, and elasticity of skin. In healthy tissue, the
   type I/III collagen ratio typically exceeds 3:1, but this ratio
   decreases following injury ([119]50). Type III collagen initially forms
   a flexible matrix for cell migration and proliferation, later
   transitioning to type I collagen during tissue remodeling to provide
   structural support ([120]51). Therefore, maintaining an optimal balance
   of collagen types is essential for effective wound healing. Sirius red
   staining was used to detect collagen types I (red) and III (green) in
   the regenerated tissue ([121]Fig. 4E). A quantitative analysis of the
   collagen I to III ratio is shown in fig. S13. For the PCM-P(SI) group,
   the ratio was 2.62 ± 0.15, 1.2 times higher than that in the Pristine
   scaffold group, suggesting that the delivery of PDGF-BB was beneficial
   to regulating collagen deposition and transition. The PCM-P/V/E (MI)
   group exhibited the largest collagen fibers with more extended
   morphology, and the ratio of collagen I to III was 3.89 ± 0.24, which
   was consistent with healthy skin, indicating well-organized collagen
   and accelerated tissue remodeling.

   To further investigate the regenerative mechanisms, we performed
   immunohistochemical and immunofluorescence staining ([122]Fig. 5).
   Because vascular regeneration is the first step in skin wound healing,
   VEGF-B and CD34 staining were used to assess the efficacy of the
   scaffolds on vascularization. As shown in [123]Fig. 5 (A and G), the
   relative expressions of VEGF-B in the PCM-P (SI), PCM-P (MI), PCM-P/E
   (MI), and PCM-P/V/E (MI) groups were 1.2, 1.5, 1.7, and 1.9 times
   higher, respectively, compared to the Pristine scaffold group. Similar
   results were obtained by quantifying neovessels through CD34
   immunofluorescence staining, where the numbers of vessels per unit area
   in the PCM-P (SI), PCM-P (MI), PCM-P/E (MI), and PCM-P/V/E (MI) groups
   were 2.0, 2.8, 2.8, and 4.0 times, respectively, greater than that in
   the Pristine scaffold group ([124]Fig. 5, B and H, and fig. S14).
   Additionally, the PCM-P/V/E (MI) group exhibited a 1.4-fold increase in
   neovascularization compared to the PCM-P/E (MI) group, demonstrating
   that the incorporation and phase-specific release of VEGF notably
   enhanced angiogenesis. To evaluate cell proliferation, KI-67 staining
   was performed. The relative expressions of KI-67 in the PCM-P (SI),
   PCM-P (MI), PCM-P/E (MI), and PCM-P/V/E (MI) groups were 1.3, 1.6, 1.9,
   and 2.2 times, respectively, higher than that in the Pristine scaffold
   group ([125]Fig. 5, C and I). Enhanced proliferation in these groups
   likely contributed to improved tissue regeneration. Epidermal
   regeneration was further assessed by CK-4 immunohistochemical analysis.
   As shown in the immunohistochemical images ([126]Fig. 5D), the group
   receiving spatiotemporally controlled delivery of multiple growth
   factors exhibited a marked increase in epidermal thickness. The
   relative expressions of CK-4 in the PCM-P (SI), PCM-P (MI), PCM-P/E
   (MI), and PCM-P/V/E (MI) groups were 1.4, 2.3, 2.8, and 3.5 times,
   respectively, higher than that in the Pristine scaffold group
   ([127]Fig. 5J). Notably, the relative expression of CK-4 in the PCM-P/E
   (MI) group was 1.2 times higher than that in the PCM-P (MI) group,
   indicating that the sustained release of EGF provided a targeted
   enhancement of epidermal regeneration. Last, we assessed skin tissue
   remodeling by evaluating collagen deposition and transformation. The
   relative expressions of collagen III in the PCM-P (SI), PCM-P (MI),
   PCM-P/E (MI), and PCM-P/V/E (MI) groups were 87.2, 79.5, 74.4, and
   56.4% of that in the Pristine scaffold group, respectively, while the
   relative expressions of collagen I were 1.3, 1.4, 1.5, and 1.6 times
   higher in these groups ([128]Fig. 5, E, F, K, and L, and fig. S15).
   These findings are consistent with the Sirius Red staining results.
   Upon masked NIR irradiation to trigger the on-demand release of
   multiple growth factors, the nanofiber scaffold promoted vascular
   regeneration, cell proliferation, collagen deposition, and skin tissue
   remodeling.

Fig. 5. Immunohistochemical and immunofluorescence staining analysis of
regenerated rabbit skin tissues in different treatment groups.

   [129]Fig. 5.
   [130]Open in a new tab

   (A) Immunohistochemical staining of VEGF-B. (B) Immunofluorescence
   staining of CD34. Immunohistochemical staining of (C) KI-67, (D) CK-4,
   (E) collagen III, and (F) collagen I, respectively. (G) Quantification
   of relative average OD of VEGF-B. (H) Quantification of blood vessel
   density. Quantification of relative average OD of (I) KI-67, (J) CK-4,
   (K) collagen III, and (L) collagen I, respectively. *P < 0.05,
   **P < 0.01, ***P < 0.001, and ****P < 0.0001.

The scaffold promotes porcine full-thickness skin wound healing upon NIR
irradiation

   To further evaluate the therapeutic efficacy of the scaffolds, a
   large-animal model in porcine were established on the basis of the
   findings from small animal experiments. This allowed for a more
   accurate representation of the human physiological environment, thus
   increasing the translational relevance of the results. Full-thickness
   skin wounds, 2.5 cm in diameter and 5 mm in thickness, were created on
   the backs of pigs, and various scaffolds were applied (fig. S16). NIR
   irradiation was performed for 10 min every 2 weeks using photomasks
   ([131]Fig. 6A). To reduce the anesthesia duration, several NIR lasers
   were administered simultaneously to multiple wounds on each pig, with
   temperature changes monitored using an infrared thermal camera
   ([132]Fig. 6B). In all groups subjected to NIR irradiation, the
   temperature of the scaffolds increased from body temperature to the
   phase transition temperature of PCM within 1.5 min. During the
   subsequent 10 min of continuous irradiation, the temperature remained
   stable around 39° to 43°C ([133]Fig. 6C and fig. S17), and this
   elevated temperature was maintained across all four rounds of
   irradiation, demonstrating the long-term photothermal stability of the
   scaffold ([134]Fig. 6C and fig. S18).

Fig. 6. In vivo evaluation of porcine full-thickness skin wound healing under
NIR irradiation.

   [135]Fig. 6.
   [136]Open in a new tab

   (A) Schematic illustration of the masked NIR irradiation protocol
   applied at weeks 0, 2, 4, and 6 postsurgery. (B) Photograph showing
   simultaneous NIR laser application to multiple wound sites. (C)
   Infrared thermal images showing the temperature of the scaffold
   increasing from baseline and stabilizing within the range of 39° to
   43°C during continuous irradiation. (D) Representative macroscopic
   images of wound healing in different groups over 8 weeks. (E) Pie
   charts depicting the percentage of remaining wound area in each group.
   (F) H&E staining and (G) Masson’s trichrome staining micrographs of
   regenerated skin tissue in different groups at 8 weeks postsurgery,
   shown at varying magnifications. I and II indicate the epidermal and
   dermal regions, respectively.

   Wound images were captured at 0, 2, 4, 6, and 8 weeks postsurgery, and
   the wound areas were quantitatively measured to assess healing rates.
   As shown in [137]Fig. 6 (D and E) and fig. S19, by week 2, wounds in
   the Blank, Pristine scaffold, and PCM (SI) groups exhibited limited
   contraction. In contrast, the PCM-P (SI), PCM-P (MI), PCM-P/E (MI), and
   PCM-P/V/E (MI) groups showed significant wound contraction, with
   remaining wound areas of 69.0 ± 3.6%, 73.9 ± 1.7%, 70.7 ± 2.2%, and
   66.1 ± 1.4%, respectively. These groups demonstrated healing rates of
   3.6, 3.0, 3.4, and 3.9 times higher than that of the Pristine scaffold
   group, due to the delivery of growth factor at this stage. By the
   fourth week, significant reductions in wound surface area were observed
   in the growth factor delivery groups. Specifically, the PCM-P/V/E(MI)
   group had a wound area of 52.1 ± 3.2%, reflecting the synergistic
   action of PDGF-BB, VEGF, and EGF in promoting skin repair. Weeks 4 to 6
   marked a critical phase of wound healing, coinciding with the third
   round of NIR-triggered growth factor release. At the sixth week, wounds
   in all groups had significantly decreased in size, especially in the MI
   groups. Compared to the PCM-P (SI) group, which had a wound area of
   31.2%, the wound area in the PCM-P (MI) group was reduced to 22.2%,
   demonstrating the efficacy of multiple cycles of PDGF-BB delivery. The
   wound healing rates for the PCM-P (MI), PCM-P/E (MI), and PCM-P/V/E
   (MI) groups were 1.4, 1.5, and 1.6 times higher than the Pristine
   scaffold group, respectively. Notably, in the PCM-P/V/E (MI) group, the
   remaining wound area was just 8.8 ± 3.2%, with no open wound bed
   observed, indicating sustained, phase-specific delivery of growth
   factors during this peak regenerative window played a pivotal role in
   accelerating wound closure and contributing to the improved final
   repair outcomes. As a comparison, in a recent study, a
   polythiourea-based polyurethane foam scaffold was compared with two
   commercial products, NovoSorb BTM and Integra, for the repair of
   porcine full-thickness skin wounds (2 cm by 1 cm) ([138]52). By day 31,
   the remaining unhealed wound area was ~50% (~1 cm^2) in the NovoSorb
   BTM group and about 43% (~0.9 cm^2) in both the polyurethane foam
   scaffold and Integra groups. In our porcine model, the surface area of
   the skin defects was 2.45 times larger than what was used in that
   study. By weeks 4 and 6, the wounds treated with PCM-P/V/E (MI) reached
   47.9% (~2.4 cm^2) and 91.2% (~4.5 cm^2) of repair, respectively.
   Compared to NovoSorb BTM, the polyurethane foam scaffold, and Integra
   at 31 days, the repair efficiency of PCM-P/V/E (MI) was approximately
   increased by 1.4-, 1.2-, and 1.2-fold at week 4 and by 3.5-, 3.1- and
   3.1-fold at week 6, respectively. These results underscore the
   effectiveness of spatiotemporal regulation of the injury
   microenvironment in accelerating wound healing. By the eighth week,
   wounds in the PCM-P (MI), PCM-P/E (MI), and PCM-P/V/E (MI) groups were
   reduced to less than 10%, while wounds in the Blank and Pristine
   scaffold groups remained at 32.9 ± 5.1% and 26.3 ± 4.4%, respectively.
   Notably, the PCM-P/V/E (MI) group exhibited complete wound healing,
   demonstrating that the sustained, on-demand delivery of multiple growth
   factors at different stages notably accelerated wound healing.

   Pathological analysis of regenerated skin tissue at 8 weeks postsurgery
   was performed using H&E staining ([139]Fig. 6F). In the Blank and
   Pristine scaffold groups, epidermal coverage was incomplete, with signs
   of crusting and erosion. The PCM (SI) group showed a better healing,
   but the epidermal structure remained irregular. Complete epidermal
   regeneration with notable thickness was observed in the PCM-P (SI)
   group, supported by PDGF-BB delivery. Upon masked NIR irradiation, the
   PCM-P (MI), PCM-P/E (MI), and PCM-P/V/E (MI) groups exhibited complete,
   well-organized epidermal structures. Among them, the PCM-P/V/E (MI)
   group showed the most organized dermal fibrous tissue, closely
   resembling the morphology of normal pig skin.

   Dermal regeneration was also evaluated. In the Blank group, the dermis
   was loosely arranged with immature collagen. The Pristine scaffold, PCM
   (SI), and PCM-P (SI) groups showed denser dermis, although less
   organized. In the groups subjected to MI, the presence of skin
   appendages indicated not only structural regeneration but also
   functional recovery. Notably, the PCM-P/E (MI) and PCM-P/V/E (MI)
   groups exhibited well-organized, compact collagen fibers in the dermis,
   with the PCM-P/V/E (MI) group closely resembling normal tissue. Masson
   staining was used to assess overall collagen deposition in the
   regenerated skin tissue ([140]Fig. 6G). The Blank group showed minimal
   collagen deposition, while the Pristine scaffold and PCM (SI) groups
   had increased collagen content, although the fibers were loosely and
   irregularly arranged. The PCM-P (SI) group exhibited higher collagen
   content with more orderly fiber arrangement. Upon MI, collagen
   secretion was significantly increased, and the newly formed collagen
   fibers became more organized and thicker. In the PCM-P/V/E (MI) group,
   collagen content, fiber organization, and thickness were comparable to
   those found in normal tissue.

   To further evaluate key aspects of skin wound healing,
   immunofluorescence staining of various markers was performed ([141]Fig.
   7). CD31 immunofluorescence staining was used to visualize the number
   of newly formed blood vessels. As shown in [142]Fig. 7A and fig. S20,
   the numbers of vessels per unit area in the PCM-P (SI), PCM-P (MI),
   PCM-P/E (MI), and PCM-P/V/E (MI) groups were 2.5, 3.4, 3.3, and 5.4
   times, respectively, higher than that in the Pristine scaffold group
   ([143]Fig. 7E). Cytokeratin-pan immunofluorescence staining was
   performed to assess epidermal regeneration ([144]Fig. 7B and fig. S21).
   The epidermal thickness and fluorescence intensity were progressively
   greater in the following order: Blank group, Pristine scaffold, PCM
   (SI), PCM-P (SI), PCM-P (MI), PCM-P/E (MI), and PCM-P/V/E (MI). The
   relative expression of cytokeratin-pan in the PCM-P/V/E (MI) group was
   3.3 times higher than that in the Pristine scaffold group ([145]Fig.
   7F). In evaluating the dermal regeneration, vimentin immunofluorescence
   staining was used ([146]Fig. 7C and fig. S22). The relative expressions
   of vimentin in the PCM-P (SI), PCM-P (MI), PCM-P/E (MI), and PCM-P/V/E
   (MI) groups were 1.2, 1.9, 2.0, and 3.4 times, respectively, higher
   than that in the Pristine scaffold group ([147]Fig. 7G). Dual
   immunofluorescence staining of collagen I and collagen III was
   performed to examine tissue remodeling ([148]Fig. 7D). The ratio of
   collagen type I to III in the PCM-P/V/E (MI) group was 3.17 ± 0.16,
   which was within the typical range found in healthy skin tissue and 1.4
   times higher than that in the Pristine scaffold group ([149]Fig. 7H)
   ([150]53). These results collectively demonstrate that the nanofiber
   scaffold, combined with masked NIR irradiation therapy, could
   spatiotemporally trigger the on-demand delivery of PDGF-BB, VEGF, and
   EGF. This sequential and synergistic effect promoted vascular
   regeneration, epidermal formation, dermal regeneration, and,
   ultimately, tissue remodeling, emphasizing the potential of this
   approach to accelerate skin wound healing.

Fig. 7. Immunofluorescence staining analysis of regenerated porcine skin
tissues in different treatment groups.

   [151]Fig. 7.
   [152]Open in a new tab

   Representative images showing the expression of (A) CD31, (B)
   cytokeratin-pan, (C) vimentin, and (D) collagen I and collagen III,
   respectively. (E) Quantification of blood vessel density. Relative
   average OD of (F) cytokeratin-pan and (G) vimentin, respectively. (H)
   Ratio of collagen I to collagen III. *P < 0.05, **P < 0.01,
   ***P < 0.001, and ****P < 0.0001.

The scaffold promotes porcine skin wound healing by comprehensively
modulating the microenvironment

   To investigate the activation of signaling pathways and gene
   regulation, transcriptomic sequencing was conducted on regenerated
   porcine skin tissues from the Blank and PCM-P/V/E (MI) groups.
   Principal components analysis ([153]Fig. 8A) and hierarchical
   clustering heatmap ([154]Fig. 8B) revealed distinct transcriptional
   profiles between the two groups. The volcano plot ([155]Fig. 8C)
   demonstrated a large number of differentially expressed genes, while
   the Venn diagram ([156]Fig. 8D) showed 1881 coexpressed genes shared
   between the Blank and PCM-P/V/E (MI) groups. As show in [157]Fig. 8E,
   Gene Ontology enrichment analysis revealed significant alterations in
   genes associated with biological processes including cell adhesion,
   migration, ECM remodeling, and positive regulation of inflammatory
   response. In the cellular component category, notable changes were
   observed in extracellular space, cell membrane, extracellular region,
   exosome, cell surface components, and ECM, indicating the role of the
   scaffold in modulating cellular behaviors and promoting tissue
   remodeling. [158]Figure 8F shows the Kyoto Encyclopedia of Genes and
   Genomes pathway enrichment analysis, highlighting that the scaffold
   primarily promoted skin wound healing by regulating key signaling
   pathways, including phosphatidylinositol 3-kinase (PI3K)–Akt,
   cytokine-cytokine receptor interaction, and mitogen-activated protein
   kinase (MAPK). These pathways are crucial for tissue regeneration and
   work together to enhance the healing process. The activation of the
   cytokine-cytokine receptor interaction pathway indicates that the
   delivery of growth factors such as PDGF-BB, VEGF, and EGF enhanced cell
   migration, blood vessel formation, and epidermal regeneration, all of
   which are critical for effective wound healing. Additionally, pathways
   related to chemokine signaling, cell adhesion, and hematopoietic cell
   lineage were activated. Moreover, the incorporation of PCM in the
   scaffold activated fatty acid metabolism, further aiding tissue
   regeneration.

Fig. 8. Transcriptome sequencing analysis of regenerated porcine skin tissues
from the Blank and PCM-P/V/E (MI) groups.

   [159]Fig. 8.
   [160]Open in a new tab

   (A) Principal components analysis (PCA) revealing distinct gene
   expression patterns between the two groups. (B) Hierarchical clustering
   heatmap showing significantly up-regulated and down-regulated genes
   after PCM-P/V/E (MI) treatment (fold change > 1.5, P < 0.05). (C)
   Volcano plot indicating the distribution of differentially expressed
   genes. (D) Venn diagram displaying the overlap of expressed genes
   between the two groups. (E) Gene Ontology term enrichment analysis for
   differentially expressed genes. BP, biological process; CC, cellular
   component; MF, molecular function. (F) Kyoto Encyclopedia of Genes and
   Genomes pathway enrichment analysis showing the top 10 significantly
   enriched signaling pathways. (G) Heatmap and (H) protein-protein
   interaction (PPI) network of key genes involved in the PI3K-Akt
   signaling pathway (fold change ≥ 2, P < 0.05). (I) Heatmap and (J) PPI
   network of differentially expressed genes involved in the MAPK
   signaling pathway (fold change ≥ 2, P < 0.05). Heatmaps of genes in the
   (K) nuclear factor κB (NF-κB) and (L) AMPK and Hedgehog signaling
   pathways, respectively (fold change > 1.5, P < 0.05). (M) Schematic
   illustration summarizing the proposed mechanism of scaffold-assisted
   therapeutic modulation to promote coordinated skin wound healing.

   A detailed analysis of the PI3K-Akt and MAPK signaling pathways
   revealed specific changes in gene expression. As show in [161]Fig. 8G,
   in the PI3K-Akt pathway, inflammation-related genes such as NFKB1, IL6,
   OSMR, and JAK3 were significantly down-regulated, suggesting that the
   sustained delivery of growth factors by the scaffold promoted tissue
   regeneration while suppressing the inflammatory response. Conversely,
   genes associated with metabolism (e.g., GYS2, CREB3L4, and FGF21) and
   cell survival, migration, and proliferation (e.g., AKT2, BCL2, FGF10,
   EFNA4, EGFR, and ERBB3) were significantly up-regulated. Additionally,
   genes involved in angiogenesis (e.g., EFNA1) and nerve regeneration
   (e.g., NGFR, NTF4, and EFNA5) were also up-regulated. These results
   indicate that the scaffold facilitated vascular regeneration and the
   formation of neural networks, contributing to the restoration of skin
   function. Protein-protein interaction (PPI) networks were constructed
   on the basis of the differentially expressed genes ([162]Fig. 8H),
   revealing that key proteins such as AKT2, EGFR, and ERBB3—central
   regulators of cell survival, migration, and proliferation—were highly
   interconnected, underscoring their critical roles in coordinating skin
   wound healing. In the MAPK signaling pathway ([163]Fig. 8I), genes
   related to immune response and apoptosis—such as TAOK2, MAX, STK4, and
   RELB—were down-regulated, indicating that the scaffold helped suppress
   inflammation and inhibit apoptosis. Conversely, genes involved in cell
   migration, proliferation, and tissue morphogenesis (RAC3, AKT2, EFNA3,
   FGF5, and DUSP8), as well as lipid metabolism (PLA2G4B), were
   up-regulated. Notably, nerve regeneration-associated genes including
   NGRF, PPP3CC, and EFNA4, were significantly up-regulated, indicating
   active neurogenesis alongside skin tissue regeneration, an essential
   aspect for restoring the sensory function of skin. The PPI network of
   MAPK-related genes ([164]Fig. 8J) further confirmed their central roles
   in orchestrating skin wound healing.

   In addition to the PI3K-Akt and MAPK signaling pathways, several other
   pathways were activated by the scaffold. Inflammation-related pathways,
   including nuclear factor κB (NF-κB), tumor necrosis factor (TNF), and
   chemokine signaling, exhibited widespread down-regulation of gene
   expression, particularly in proinflammatory genes, indicating an
   anti-inflammatory effect ([165]Fig. 8K and figs. S23 to S25). In
   contrast, the Hedgehog signaling pathway was significantly
   up-regulated, supporting tissue patterning and scarless skin
   regeneration, including hair follicle neogenesis (fig. S26) ([166]54).
   Several metabolism-related pathways were also activated, such as the
   adenosine 5′-monophosphate-activated protein kinase (AMPK) pathway and
   the alanine, aspartate, and glutamate metabolism pathway (figs. S27 and
   S28). Notably, the peroxisome signaling pathway, which is involved in
   cellular detoxification and lipid metabolism, was up-regulated,
   suggesting enhanced fatty acid metabolism, reduced oxidative stress,
   and promotion of skin cell proliferation (fig. S29). Collectively, gene
   expression changes in the Hedgehog and AMPK pathways suggest that
   modulation of cellular metabolism enhanced keratinocyte proliferation,
   fibroblast activity, and overall tissue remodeling, thereby
   facilitating effective skin wound healing ([167]Fig. 8L).

   Together, the scaffold, integrating a radially aligned topology, NIR
   light irradiation, and spatiotemporally delivered growth factors,
   promoted skin wound healing and functional recovery by establishing a
   favorable regenerative microenvironment ([168]Fig. 8M). It
   synergistically down-regulated genes associated with inflammation and
   apoptosis while up-regulating those involved in angiogenesis, nerve
   regeneration, cell survival, migration, proliferation, and ECM
   remodeling. These effects were mediated through the activation and
   cross-regulation of key signaling pathways, including PI3K-Akt, MAPK,
   NF-κB, TNF, Hedgehog, and AMPK, thereby comprehensively enhancing the
   wound healing process.

DISCUSSION

   The skin, as the body’s largest organ and first line of defense, plays
   a crucial role in various physiological functions. Although the skin
   has innate regenerative capabilities, extensive wounds often require
   artificial therapeutic interventions to accelerate healing. Clinical
   products like allografts, xenografts, and synthetic matrices have
   advanced wound care ([169]8). Allografts, such as Apligraf and
   Dermagraft, offer bioactivity that reduces healing time but face
   challenges with production and storage due to the presence of live
   cells. Xenografts, like Integra and Oasis Wound Matrix, address these
   issues but may cause immune rejection. Synthetic polymer–based
   scaffolds, such as Restrata, Hyalomatrix, and NovoSorb BTM, offer
   nonimmunogenicity and biocompatibility but lack bioactivity, limiting
   their ability to modulate the healing process. With the improved
   understanding of the mechanisms underlying skin wound healing,
   scaffolds that precisely control biophysical and biochemical cues are
   increasingly recognized, which requires the dynamic manipulation of key
   healing stages, such as inflammation, cell proliferation, and tissue
   remodeling, to create a microenvironment that supports effective skin
   wound healing ([170]55).

   Our study presents an innovative bioactive scaffold that incorporates
   multiple guidance cues and leverages NIR irradiation for the
   spatiotemporally controlled delivery of specific growth factors,
   thereby actively regulating each stage of the wound healing process.
   The scaffold features a hierarchical structure, with radially aligned
   nanofibers facilitating directed cell migration and a randomly oriented
   outer layer providing mechanical support. Constructed from FDA-approved
   materials through scalable electrospinning and electrospraying
   techniques, the scaffold holds strong potential for clinical
   translation. This approach directly addresses the limitations of
   current therapeutic options. In contrast to existing synthetic
   matrices, the scaffold integrates topographical cues, a tunable
   photothermal effect, and sustained biochemical signaling to establish a
   dynamic microenvironment tailored to the distinct phases of wound
   healing. A key innovation is its capacity for on-demand, localized
   delivery of growth factors in response to NIR irradiation. The
   photothermal response, enabled by ICG-encapsulated phase-change
   microparticles, maintains temperatures above 40°C under clinical
   conditions. A photomask allows for spatially and temporally precise
   release of PDGF-BB, VEGF, and EGF, synchronized with specific phases of
   the healing process, which is a great challenge. In vitro, the released
   factors (107.45 ± 0.75 ng of PDGF-BB, 59.02 ± 1.81 ng of VEGF, and
   64.15 ± 0.85 ng of EGF) significantly enhance angiogenesis and cell
   migration and proliferation. Furthermore, the synergistic effect of
   topographical and biochemical cues facilitates radial migration of
   L929, NIH-3T3, and HaCaT cells, improving cellular distribution across
   the wound bed.

   In a rabbit full-thickness wound model with a diameter of 2.5 cm, our
   scaffold combined with photomask-assisted NIR irradiation achieves an
   impressive healing rate of 95% (~4.7 cm^2) within 2 weeks, notably
   outperforming conventional scaffolds. Notably, the wounds heal without
   visible scarring, and the newly regenerated skin is indistinguishable
   from native tissue. Histological analysis confirms organized collagen
   deposition and a collagen I/III ratio comparable to that of healthy
   skin, reflecting the capacity of the scaffold to dynamically modulate
   the regenerative microenvironment. This modulation effectively supports
   vascular regeneration, epidermal proliferation, and dermal remodeling.
   Compared with studies using ECM-mimetic coatings from natural sources,
   our scaffold demonstrates a 5.7-fold improvement in healing efficiency,
   highlighting its transformative potential in wound care ([171]56).
   Moreover, the controlled release of growth factors mitigates risks
   associated with unregulated delivery, such as aberrant tissue formation
   and potential carcinogenesis.

   Porcine models, which closely resemble human skin in terms of
   physiology and wound healing dynamics, further validate the
   translational potential of our scaffold. In full-thickness wounds (2.5
   cm in diameter, 5 mm in depth), the scaffold achieves 47.9% healing by
   the fourth week, corresponding to a wound closure area of 2.4 cm^2. In
   comparison, a recent study using a reactive oxygen species-responsive
   polyurethane foam for similar wounds (2-cm length by 1-cm width),
   equivalent to 40.8% of the wound surface area in our model, reports 57%
   closure after 31 days, with a surface area of ~1.1 cm^2 ([172]52). In
   addition to outperforming this experimental material, our scaffold
   exhibits faster and more complete healing than commercially available
   products such as NovoSorb BTM and Restrata ([173]12). By the eighth
   week, wounds treated with our scaffold achieve near-complete closure.
   Unlike conventional treatments that often result in fibrotic scarring,
   our scaffold supports regeneration characterized by restored skin
   architecture and function. Histological analysis of regenerated tissue
   reveals a higher density of neovasculature, increased epidermal
   thickness, enhanced dermal matrix formation, and a collagen type I/III
   ratio within the typical range observed in healthy skin. To elucidate
   the underlying mechanisms, we conducted in vivo transcriptomic
   sequencing, which reveals that the scaffold activates key regenerative
   signaling pathways, including PI3K-Akt and MAPK, while also modulating
   immune responses and metabolic reprogramming. Genes involved in
   angiogenesis, neurogenesis, cell migration, proliferation, and cellular
   metabolism are significantly up-regulated, whereas those associated
   with apoptosis and inflammation are down-regulated. Notably, this
   coordinated molecular response enables the simultaneous initiation of
   angiogenesis and neurogenesis within the same ECM microenvironment.
   Neovascularization not only enhances nutrient and oxygen delivery but
   also creates a permissive niche for neural network development, further
   supporting functional restoration. Therefore, the favorable outcome
   stems from the scaffold’s capacity to dynamically modulate the wound
   microenvironment. The early release of PDGF-BB and VEGF effectively
   suppresses acute inflammation and initiates robust angiogenesis, laying
   the groundwork for regeneration. As healing progresses, the sustained
   release of PDGF-BB and EGF, in conjunction with the scaffold’s radially
   aligned fibers and fatty acid–rich matrix, promotes cell migration,
   proliferation, and tissue remodeling. Concurrently, fatty acid
   metabolism is regulated to further support cellular energy demands
   during tissue repair. By establishing a sequentially optimized and
   spatially structured microenvironment, the scaffold orchestrates a
   comprehensive series of regenerative events, ranging from
   vascularization and immune regulation to neural integration and
   collagen remodeling. As a result, it not only accelerates wound closure
   but also ensures the regeneration of functionally and structurally
   complete skin tissue, with minimal risk of adverse effects.

   The proposed scaffold provides a comprehensive solution to several
   persistent challenges in wound care by integrating
   photothermal-responsive technology with an advanced, adaptable scaffold
   design. Constructed from FDA-approved and clinically validated
   materials, the scaffold is compatible with scalable manufacturing
   techniques, which lowers regulatory hurdles and reduces production
   costs, enhancing its potential for broad clinical adoption. Its
   activation via a handheld NIR light emitter allows for convenient,
   noninvasive treatment under medical guidance, making it especially
   well-suited for outpatient care and resource-limited settings. By
   dynamically responding to the wound healing process, the scaffold
   minimizes risks such as chronic inflammation and excessive scarring,
   thereby promoting sustained tissue regeneration and long-term recovery.
   Its customizable structure supports use across a diverse range of wound
   types and sizes, from acute injuries to complex chronic wounds. The
   scaffold’s capacity to sustain and precisely regulate growth factor
   delivery addresses a critical need in managing chronic wounds,
   including diabetic ulcers and burn injuries.

   While the proposed scaffold shows great potential, future modifications
   may further enhance its suitability for clinical applications. One area
   of improvement lies in the precise regulation of NIR-triggered growth
   factor release, which depends on accurately monitoring the wound
   microenvironment to identify the factors needed at specific healing
   stages, enabling timely and targeted activation. Another important
   consideration is thermal control, as excessive local heating may
   compromise tissue viability and therapeutic consistency. To address
   this, future strategies could focus on developing smart,
   feedback-controlled NIR systems that integrate real-time thermal
   sensors to automatically adjust irradiation parameters. In parallel,
   PCMs could be further engineered with thermally responsive molecular
   regulators that become optically inactive upon reaching their melting
   point, thereby limiting further NIR absorption and preventing
   overheating. Collectively, these advances would contribute to a
   next-generation therapeutic paradigm for skin wound repair, one that
   enables precise, spatiotemporal regulation of regenerative cues across
   all stages of healing.

MATERIALS AND METHODS

Study design

   This study aimed to develop a nanofiber scaffold that enables the
   spatiotemporally controlled delivery of multiple growth factors,
   specifically designed to address the unique requirements of each phase
   in skin tissue regeneration for effective wound repair. The scaffold’s
   morphology was examined via SEM. Key properties such as chemical
   composition, mechanical strength, and photothermal performance were
   thoroughly evaluated. To validate the NIR-triggered controlled release
   system, an in vitro release simulation was set up to observe color
   changes corresponding to the release of differently colored dyes. The
   quantities of released growth factors were quantified using
   enzyme-linked immunosorbent assay kits, and their bioactivity was
   assessed. Various cell types relevant to skin tissue regeneration,
   including HUVECs, L929 fibroblasts, NIH-3T3 fibroblasts, and HaCaT
   cells, were used as in vitro models to assess the scaffold’s ability to
   regulate vascularization, promote cell proliferation, and enhance cell
   migration. For in vivo validation, full-thickness skin wound models
   were established in both small animals (rabbits) and large animals
   (pigs) to evaluate the efficacy of the scaffold combined with NIR
   irradiation in promoting skin wound healing. The rabbit skin wound
   repair experiment was approved by the Animal Ethics and Welfare
   Committee of Institute of Radiation Medicine, Chinese Academy of
   Medical Sciences and followed ethical guidelines (IRM-DWLL-2021210).
   The porcine skin wound repair experiment was approved by the Animal
   Ethics and Welfare Committee of Peking University People’s Hospital and
   followed ethical guidelines (2025PHE013). During these experiments,
   photomasks of adjustable sizes and regular NIR light interventions were
   applied, and photographs were taken to track wound healing progress.
   After the wound healing process, regenerated skin tissues were
   harvested and subjected to H&E staining, Masson’s trichrome staining,
   Sirius red staining, and a series of immunohistochemistry and
   immunofluorescence assays to assess key regenerative factors, including
   vascularization, cell proliferation, epidermal regeneration, dermal
   regeneration, and collagen deposition. To investigate the underlying
   molecular mechanisms, transcriptomic gene sequencing was performed on
   the regenerated skin tissue from pigs to identify pathway changes and
   gene expression regulation during skin repair.

Statistical analysis

   All experiments were performed with at least three replicates, and data
   were expressed as means ± SD. Statistical analysis was conducted using
   Origin software and GraphPad Prism 8.0 software. Differences between
   two groups were analyzed using a t test. Comparisons among more than
   two groups were performed using one-way analysis of variance (ANOVA). A
   P value of less than 0.05 was considered statistically significant
   (*P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001).

Acknowledgments