Hydroxyapatite microspheres encapsulated within hybrid hydrogel promote skin regeneration through the activation of Calcium Signaling and Motor Protein pathway

Abstract

Hydroxyapatite (HAp), traditionally recognized for its efficacy in bone regeneration, has rarely been explored for skin regeneration applications. This investigation explored HAp microspheres with distinct physicochemical properties tailored away from conventional bone regeneration parameters, and the capacity promoting skin regeneration and mitigating the aging process were investigated when encapsulated in hyaluronate hydrogels. By benchmarking against well-established dermal fillers like PMMA and PLLA, it was revealed the specific attributes of HAp that were conducive to skin regeneration, providing initial insights into the underlying mechanism. HAp enhanced the fibroblast functionality by triggering minimal adaptive immune responses and enhancing the Calcium Signaling and Motor Protein Signaling pathways. This modulation supported the production of normal collagen fibers, essential for ECM maturation and skin structural integrity. The significant ECM regeneration and remodeling capabilities exhibited by the HAp-encapsulated hybrid hydrogels suggested promising application in facial rejuvenation procedures, potentially making a breakthrough in aesthetic and reconstructive surgery.

Graphical abstract

Highlights

• •Explores distinct HAp microspheres for skin regeneration, beyond bone healing.
• •Compares HAp with dermal fillers like PMMA and PLLA for skin rejuvenation.
• •HAp promotes fibroblast function by triggering calcium and motor protein signaling.
• •Demonstrates HAp's potential in ECM regeneration for facial rejuvenation.

1. Introduction

Facial aging is an inevitable biological process that impacts an individual's self-esteem and social interactions significantly [1,2]. External stimuli, alterations in the microenvironment, and changes in the immune system contribute to the progressive aging of skin cells. This results in decreased cell proliferation, increased resistance to apoptosis, secretion of inflammatory and tissue-damaging factors, and accumulation of senescent cells [[3], [4], [5]]. Gradually, these changes impair the facial skin extracellular matrix and imped the regeneration process through a complex, multi-dimensional process involving alternations in several compositions and structures, such as collagen fibers and fat pads. The visible signs of facial aging include skin thinning, elasticity reduction, moisture loss, and hollow formation. Notably, specific indicators of facial aging, such as temporal and tear trough depressions, and deepening of nasolabial folds, can have a significant impact on an individual's appearance and psychological well-being in modern society [6,7].

Consequently, there is a growing interest in improving facial depressions to achieve smoother facial lines and rejuvenation. According to data from the International Society of Aesthetic Plastic Surgery (ISAPS), non-surgical procedures have increased significantly by 57.8 % over the past four years. Injectable products have diversified to meet clinical needs, with hyaluronic acid (HA) and polymer-based fillers dominating the market. HA is a physical filler widely used for its excellent biocompatibility and can be modified for various clinical applications since its initial introduction in the early 21st century [8]. In 2023, HA injections ranked second among the world's top 10 non-surgical aesthetic treatments, with a 29 % increase in usage compared to 2022, solidifying its position as the leading filler material. However, the bioactivity of HA is limited, and over-injection in recent years has led to poor aesthetic outcomes in a significant number of cases [9,10]. With the enhanced knowledge of implantable biomaterials and increasing demand for natural and long-lasting beauty, the safety and efficacy of microsphere-encapsulated fillers such as poly (L-lactic acid) (PLLA), poly (methyl methacrylate) (PMMA), poly (caprolactone) (PCL), and hydroxyapatite (HAp) have been verified in minimally invasive approaches, and multiple products have been approved and launched globally [2,8,[11], [12], [13], [14], [15]]. Nonetheless, the increasing application of these microsphere-encapsulated fillers has also led to more reported clinical complications, such as nodules, granulomas, and recurrent painful swelling, which may require surgical excision and cause significant stress to patients [[16], [17], [18], [19], [20], [21], [22], [23]].

Among the emerging biomaterials, hydroxyapatite (HAp) is a major inorganic component of human bones and teeth, known for its excellent biocompatibility and bioactivity [24,25]. While HAp has been widely applied in bone regeneration for decades, its use in facial rejuvenation remains relatively limited. Currently, a filler named Radiesse® (Merz Aesthetics), composed of hydroxyapatite and carboxymethyl cellulose (CMC) has been approved and applied in the USA and Europe [14,26]. According to the Business Research Company (TBRC), the HAp fillers market is expected to grow from $620 million in 2023 to $710 million in 2024, with a Compound Annual Growth Rate (CAGR) of 13.8 %. By 2028, the market size is projected to reach $1.19 billion, maintaining a CAGR of 14 %. Despite representing only 1.8 % of global non-surgical treatments in 2023, HAp fillers have experienced a 62 % growth since 2019, ranking eighth among non-surgical procedures and second among filler materials. Prior research on HAp as a dermal filler primarily focused on biocompatibility evaluation, with limited exploration of the intrinsic connection between material properties and biological effects, as well as the tissue regeneration mechanisms [[27], [28], [29], [30], [31], [32]].

In this study, the material factors of HAp with controlled preparation parameters were emphasized, and the causality of its in vivo performance was investigated when implanted using HA hydrogel as the carrier. Also, the role and mechanism of HAp in skin regeneration were discussed, compared with representative degradable and non-degradable polymer microspheres. Through the identification of key determinants of material performance and an appropriate understanding of the in vivo molecular mechanisms, HAp-based soft tissue filler can be anticipated as a novel therapeutic approach for facial rejuvenation treatments.

2. Materials and methods

2.1. Preparation of hydroxyapatite microspheres

Hydroxyapatites were synthesized through wet chemical precipitation using a reactive system of Ca(NO₃)₂⋅4H₂O and (NH₄)₂HPO₄ (Sinopharm Chemical Reagent Co., Ltd., Shanghai, China) solutions according to the molar ratio of Ca/P at 1.67, while maintaining the pH around 10.0 by adding NH₄OH. Subsequently, the precipitate was aged, washed, spray-dried, and sintered at three different temperatures (300 °C, 550 °C, and 800 °C). The resulting microspheres were labeled as HAp300, HAp550, and HAp800, respectively. Afterward, they were sieved to isolate particles within the specific size range of 25–40 μm for subsequent studies.

2.2. Characterization of hydroxyapatite microspheres

Granularity analysis was conducted using a laser particle sizer (Mastersizer 3000, Malvern, Worcestershire, UK). The chemical identification of HAp was performed using attenuated total reflection Fourier transform infrared spectroscopy (ATR-FTIR, Nicolet 6700, Thermo, USA) within the wavenumber range of 4000–400 cm⁻¹. The surface morphology of HAp was visualized through a scanning electron microscope (SEM, S-4800N, Hitachi, Japan). Crystallinity analysis was conducted using X-ray diffraction (XRD, Shimazu XRD-6100, Kyoto, Japan), with scans performed from 20° to 60° at a rate of 5°/min. The following equation was used to calculate the crystallinity (Xc) of HAps:

$${\mathrm{X}}_{\mathrm{c}}=1-\left({\mathrm{V}}_{112/300}/{\mathrm{I}}_{300}\right)$$ (1)

where, V_112/300 represented the intensity of the dip between (112) and (300) reflections, and I₃₀₀ represented the intensity of the (300) diffraction peak.

2.3. In vitro degradation and Ca²⁺ Release Analysis

The degradation HAp was characterized by soaking in the buffer solution and measuring the concentration of released calcium. Briefly, the HAp was dispersed in Tris-HCl solution (0.1M, pH = 7.4, Solarbio Science Technology Co., Ltd., Beijing, China) at a 0.1 g/mL concentration. The mixture was then shaken at 37 °C for different time durations and 300 μL of the centrifuged supernatants was sampled for calcium ion concentration measurement using a Calcium Assay Kit (manufacturer). After sampling, 300 μL of flesh Tris-HCl was added to restore the solution volume to 1 mL. The Ca²⁺ concentration in the supernatant was measured at given time intervals.

2.4. In vitro cell culture

Human adult skin fibroblasts (HSF, iCell Bioscience Inc., Shanghai, China) and macrophages (Raw 264.7, iCell Bioscience Inc., Shanghai, China) were cultured in DMEM medium (Servicebio Technology Co., Ltd., Wuhan, China) supplemented with 10 % Fetal bovine serum (FBS, Thermo Fisher Scientific, Waltham, MA, USA) and 1 % penicillin and streptomycin (Thermo Fisher Scientific, Waltham, MA, USA). Cultures were maintained at 37 °C in a humidified incubator with a 5 % CO₂ atmosphere. For cell viability and proliferation analysis, HSFs were seeded at a density of 10⁴ cells per well in 24-well plates. After cell adhesion, HAp extraction was added and co-cultured with cells for 1, 2, and 3 days according to the standard protocol. The HAp extraction was obtained at a concentration of 0.2 g/mL (ISO 10993-12-2021) by incubating for 24 h at 37 °C. To observe the fibroblast morphology on HAps directly, the microspheres were immobilized on agar (Agar/HAps) with the HAp surface exposed. In addition, RAW264.7 cells were seeded at a density of 106 cells per well on the Agar/HAps surface in 24-well plates. After 24 h of co-culture, macrophage phenotypes were assessed using flow cytometry and immunofluorescence staining.

Cell viability and proliferation of fibroblasts were evaluated by Cell Counting Kit-8 (CCK-8, Dojindo Laboratories, Kumamoto, Japan) assay and a live/dead staining method. For the CCK-8 assay, the medium was first removed after HSF cells were cultured with HAp extractions overnight. At each designated time point, the cells were incubated with a fresh DMEM containing water-soluble tetrazolium-8 (WST-8) at a volume ratio of 1:9 (WST-8 to DMEM). After incubation for 2 h at 37 °C in a humidified 5 % CO₂ atmosphere, the optical density (OD) values in all wells were measured at 450 nm by a microplate reader (EON; BioTek, Winooski, VT, USA). For the Live/Dead staining, a dural-staining procedure was performed using Di-O-acetyl fluorescein (FDA, Sigma, USA) and Propidium iodide (PI, Sigma, USA). Cells were incubated with the staining solution for 10 min in the dark at room temperature. The 24-well glass-bottom plates (Nest, China) were then washed twice with phosphate-buffered saline (PBS, Gibco, USA) for 10 min each. Subsequently, images were captured using a Confocal Laser Scanning Microscope (CLSM, LSM880, Carl Zeiss, Germany).

Fibroblasts were seeded and co-cultured on these surfaces for 1 day and 14 days. F-actin fluorescence staining was employed to evaluate the morphology of fibroblasts. After in vitro culture for 1 and 3d, the samples were washed with PBS to remove nonadherent cells and then fixed in 4 % paraformaldehyde for 10 min at 25 °C. Subsequently, the samples were permeabilized with 0.5 % triton X-100 (Servicebio Technology Co., Ltd., Wuhan, China) solution, followed by cytoskeleton fluorescent staining with phalloidin-trite (Sigma-Aldrich, USA) and 4’, 6-Diamidino-2-phenylindole dihydrochloride (DAPI, Sigma-Aldrich, USA), respectively. Finally, cell morphology was observed and recorded using a CLSM.

Flow cytometry was quantified to analyze the macrophage phenotype after 24 h of co-culture with HAp. After removing the original medium, pre-cooled PBS was added to rinse the cells gently, and those adhered to the material's surface were carefully detached and transferred to a centrifuge tube. The cells were centrifuged at 1000 rpm for 5 min, and the supernatant was discarded. CD16/CD32 blocking reagent (BD, USA) was then added to the cell pellet, followed by incubation at room temperature for 25 min. Afterward, CD86 and CD206 antibodies (BD, USA) were added, and the cells were incubated in the dark at room temperature for 30 min. Following incubation, the cells were washed by centrifugation at 1000 rpm for 5 min, resuspended in PBS, and the washing step was repeated. Finally, the cells were resuspended in a flow buffer, and a 200 μl aliquot of the suspension was transferred to a flow cytometry tube for analysis. The prepared samples were then analyzed to determine macrophage phenotypes based on their surface markers.

Immunofluorescence staining was performed to qualitatively evaluate the material's effect on macrophage phenotype. After 24 h, the medium was discarded, and the samples were washed three times with PBS. RAW 264.7 cells were fixed with 4 % paraformaldehyde for 25 min at room temperature, followed by three additional PBS washes. Permeabilization was achieved using 0.5 % Triton X-100 solution (Servicebio Technology Co., Ltd., Wuhan, China) for 25 min at room temperature, and the cells were washed three more times with PBS. 10 % goat serum was applied to submerge the material, incubating it at 37 °C for 30 min. After removing the blocking solution, diluted CD206 and iNOS primary antibodies (Abcam, UK) were added, and the samples were incubated overnight at 4 °C. The next day, a fluorescent secondary antibody was applied in the dark, followed by a 1-h incubation at room temperature. After three PBS washes, DAPI was used to stain the nuclei for 3 min at room temperature. Finally, the samples were washed three more times with PBS, PBS was added, and the stained samples were observed using a CLSM.

2.5. In vivo animal experiment

All animal experiments were approved by the Animal Care and Use Committee of Sichuan University (Approval No. WCHSIRB-D-2024-530) and adhered to the animal welfare guidelines established by the Chinese Society for Laboratory Animals. 6 to 8-week-old male SD rats with an average weight of 180g were purchased from Byrness Weil biotech. Ltd. (China). The rats were randomly divided into 7 groups and provided with a standard diet for one week.

Besides the three HAp groups (HAp300, HAp550, and HAp800), commercially available implant-grade polymethylmethacrylate microspheres (PMMA, particle size 30 ± 2 μm) and poly-l-lactic acid microspheres (PLLA, particle size 20–50 μm) were chosen as control materials. Before the implantation, microspheres in all groups were homogeneously encapsulated in BDDE-crosslinked HA filler using two syringes connected by a medical 3-way mixer. The syringes were alternately pushed 50 times to obtain the hybrid hydrogels containing microspheres at a volume ratio of 10 %, and labeled as HA/PMMA, HA/PLLA, HA/HAp300, HA/HAp550, and HA/HAp800, respectively. Normal saline (NS) and pure HA hydrogel (HA) served as blank and negative controls.

2.6. Characterization of hybrid hydrogels

The so-obtained hybrid hydrogels were physically characterized to evaluate their stability and operability before implantation. The internal structure and microsphere distribution within the hybrid hydrogel were visualized through SEM images of the freezing-dried sample. Besides, the composite hydrogels were loaded into standard 1 mL syringes equipped with 27G needles, and the extrusion force was determined using a Universal Testing Machine (AGS-X, Shimadzu Inc., Japan) after expelling the air bubbles. The loading force-displacement curves were obtained by extruding the hydrogels at a constant rate of 30 mm/min. The rheological properties of the hybrid hydrogels were assessed using a rotational rheometer (MCR302, Anton Paar, Inc., Austria). Specifically, equal volumes of each hydrogel were deposited on the sample stage and subjected to a linearly increasing shear rate from 0 to 100 s⁻¹. Fifty data points were recorded at various shear rates to derive viscosity values at 25 °C.

2.7. Implantation of composite hydrogels

Following anesthesia induced by isoflurane inhalation, the rats had their back hair removed and underwent iodophor disinfection. The composite hydrogels were injected into the subcutaneous tissue layer of the rats with 25G needles at a dose of 200 μL per injection site. A total of six injection sites were established on the back of each rat for subsequent analysis at 3 days, 1 month, 3 months, 6 months, and 9 months.

2.8. In vivo biocompatibility and immune response

Infrared thermography and digital imaging were employed to observe the surrounding tissues post-material implantation. At specified time intervals, the rats were euthanized to harvest the remaining hydrogel and surrounding skin samples, which were fixed in 10 % neutral buffered formalin solution (NFB, Servicebio Technology Co., Ltd., Wuhan, China) for histological and immunohistochemical (IHC) examination. Hematoxylin and Eosin (H&E) staining, Myeloperoxidase (MPO) staining, inducible Nitric Oxide Synthase (iNOS), and Mannose receptor (CD206) were conducted on 4-μm-thick tissue sections. Subcutaneous tissues collected one month after material implantation were analyzed for protein expression using ELISA kits. Proteins related to inflammation, including IL-1β, IL-4, and IL-10, were quantified to evaluate the tissue inflammatory status.

2.9. Determination of in vivo degradation

Skin ultrasound testing enabled the comparison of changes in the skin, subcutaneous tissue, and implants across different time points in the same injected area of live rats. Ultrasound images of composite fillers in vivo were captured by Vevo 3100 Imaging System (Fujifilm VisualSonics, Toronto, Canada) equipped with a 40-MHz linear array ultrasound transducer (MX550D, VisualSonics Inc.), which were performed at the 1m and 3m post-injection. The major scanning parameters included: slice thickness = 0.1 mm, depth off = 1.00 mm, and gain = 22 dB. The attained data were utilized to reconstruct and calculate the remaining implants using the Ultrasound Workstation Dongle. A computed tomographic scanner (Quantum GX II, PerkinElmer, Inc. Chicago, USA) was applied to quantify the remaining volume of HAp in vivo at specific conditions: slice thickness = 0.1 mm, Rows ∗ Columns = 512∗512, voltage = 90 KV, current = 88 mA, and field of view (FOV) = 37 ∗37 mm∗mm. The occupancy of HAp was further reconstructed using Mimics Research 20.0 based on the CT data, and a degradation pattern graph was generated to illustrate the in vivo degradation of HAp microspheres at 1m, 3m, and 6m.

2.10. Regenerative effect evaluation

Special tissue staining methods such as Masson's trichrome and Sirius Red were used for evaluating collagen regeneration. Briefly, samples were fixed at 4 °C in 2 % paraformaldehyde-2.5 % glutaraldehyde, dehydrated, sliced, and stained according to the route protocol. The morphology of collagen fibers and fibroblasts was assessed with SEM and transmission electron microscopy (TEM, JEM-1400 Plus, JEOL Ltd., Japan) after staining with 2 % uranyl acetate and Reynolds lead citrate. Furthermore, ultrasonography (Vevo 3100 Imaging System) was used as a non-invasive technique to evaluate dermal thickness.

2.11. Exploration of regenerative mechanisms

Tissue RNA sequencing, immunofluorescence, and Enzyme-linked immunosorbent assay (ELISA) were employed to elucidate the mechanism by which hydroxyapatite particles (HAps) enhanced tissue regeneration.

Three months after material implantation, skin tissues were collected post-euthanasia for RNA sequencing analysis. Total RNA was extracted using TRIzol reagent (Invitrogen, USA), and quality control was performed to ensure integrity before sequencing. Differential gene expression and KEGG pathway enrichment analyses were conducted to preliminarily explore the underlying mechanisms of tissue response.

Samples collected 3 days and 3 months post-implantation into subcutaneous tissue were analyzed via immunofluorescence staining to detect the expression of fibroblast growth factor 2 (FGF2) and S100 calcium-binding protein A4 (S100A4).

Additionally, subcutaneous tissues collected one month after implantation were assessed for protein expression using ELISA kits. Proteins related to cellular process-d, including Myosin Heavy Chain 9 (MYH9), S100A4, integrin αV (Itg αⅤ), integrin β1 (Itgβ1), and Vinculin, were quantified to evaluate the biological response to the implanted material.

2.12. Statistical analysis

All experimental data were presented as the average value with standard deviation (SD), while at least three parallel samples were investigated for each group. Statistical analyses were performed using one-way analysis of variance (ANOVA) via SPSS 20.0 software. Statistical significance was defined at P < 0.05 and P < 0.01.

3. Results

3.1. Characterization of HAps

As shown in Fig. 1A, HAp microspheres presented as regular spheres with unsmooth surfaces, which was significantly different compared with the compact and smooth surfaces of both PMMA and PLLA microspheres. Nano-size grains and pores could be observed on HAp surfaces, which increased with the sintering temperature of the samples. Meanwhile, the microsphere size of HAp shown in Fig. S1A ranged from 20 μm to 60 μm with an average diameter of 29 ± 2 μm, which could be considered consistent with the sizes of polymer microspheres. ATR-FTIR results shown in Fig. S1B suggested that HAps prepared at different sintering temperatures exhibited similar major characteristic absorption peaks, including the vibrational peak of OH⁻ at 3560 cm⁻¹ and the telescopic vibrational peaks of PO₄³⁻ at 550 cm⁻¹, 650 cm⁻¹, 940 cm⁻¹, and 1120 cm⁻¹. Based on the XRD results and standard curve of HAp given in Fig. 1C, all as-prepared HAp showed characteristic peaks and the crystallinity of HAp obtained at different sintering temperatures was 43.75 % for HAp300, 47.03 % for HAp550, and 81.95 % for HAp800, respectively. The results suggested that the crystallinity of HAp could be modulated by the preparation parameters, such as the sintering temperature in this case, and the higher temperature would obtain the higher crystallinity of HAp which was consistent with the previous reports. The in vitro degradation and calcium ion release results (Fig. 1D) showed that HAp sintered at low temperatures could release more calcium ions in a short period, and HAp800 still released a lower concentration of calcium ions at 120 h.

Fig. 1.

Characterization of microspheres (HAps, PMMA and PLLA) and hybrid fillers. A, Surface morphology of HAps, PMMA, and PLLA: Sintering at 800 °C resulted in denser spheres with more distinct grain boundaries compared to sintering at 300 °C and 550 °C. The PMMA and PLLA microspheres exhibited smooth surfaces lacking cell adhesion sites. B, Surface morphology of hybrid hydrogels: HA presented a loose mesh structure incorporating dispersed HAp, PMMA, and PLLA microspheres. PMMA and HAp particles were uniformly sized, while PLLA microspheres varied in size. C, Crystallinity analysis: The crystallinity of HAp increased with higher sintering temperatures. D,In Vitro Ca²⁺ Release Analysis: HAp sintered at low temperatures released more calcium ions. E, Pushing force of the hydrogel materials using 27G needles: The pushing forces of the HA/HAp groups fell between those of HA/PMMA and HA/PLLA, with minor fluctuations but still allowing for smooth injection. F, Viscosity properties of the hybridized hydrogels: As the shear rate increased, the hybrid hydrogels showed a decrease in viscosity, exhibiting shear-thinning behavior. n = 7, ∗∗∗∗p < 0.0001.

3.2. In vitro cytocompatibility

Since HSFs are the dominant cells in the dermis and play a crucial role in tissue regeneration, they were chosen to investigate the influence of materials on cell behaviors, including cell viability and proliferation. In Fig. S1C, The Live/Dead staining results demonstrated a notable increase in live cell numbers over time across all groups. In the blank group, cells maintained a spherical morphology, whereas cells in the three HAp groups displayed enhanced spreading and star or fusiform shapes. These findings underscored the favorable cytocompatibility of HAp, promoting cytoskeletal spreading.

The CCK-8 results shown in Fig. S1D revealed that HAps in all groups were noncytotoxic. Notably, cell proliferation in the HAp800 group was significantly higher than those in HAp300 and HAp 500, although no significant difference was observed relative to the control group after 1d of culture. With the extension of culture time, cells proliferated well in all groups with the HAp800 exhibiting the highest absorbance values. However, the difference among the groups became insignificant at 3d owing to high cell confluence in the culture wells. These results suggested that HAp synthesized and applied in this study was cell-compatible and could promote HSF proliferation under suitable conditions.

3.3. In vitro fibroblasts morphology on hydroxyapatites surface

Fig. S2A demonstrated that hydroxyapatite microspheres were uniformly immobilized on agarose, with their surfaces exposed to facilitate cell adhesion. Cytoskeletal staining in Fig. S2E revealed that fibroblasts spread effectively on the hydroxyapatite surface, with no significant differences observed in the spreading areas of individual microspheres across the three HAp groups.

3.4. In vitro macrophage phenotype on hydroxyapatites surface

Flow cytometry and immunofluorescence analysis (Figs. S2B–D) indicated that a higher proportion of M2-type RAW 264.7 macrophages adhered to the microsphere surfaces at this time point.

3.5. Characterization of composite hydrogel

SEM analysis reveals that HA exhibits a loose mesh structure, with dispersed HAp, PMMA, and PLLA microspheres, as shown in Fig. 1B. PMMA and HAp display uniform-sized particles, whereas PLLA microspheres exhibited a broader size range. Also, both PMMA and PLLA demonstrated an agglomeration phenomenon, which was particularly noticeable in PMMA and possibly due to their hydrophobic interaction in a hydrophilic surrounding. The extruding forces of the hydrogels through a 27G needle were attained and the curves were given in Fig. 1E, revealing the loading force to initial and maintain the extrusion of different composite hydrogels, as well as the influence of possible microsphere aggregation. It could be found that the pure HA hydrogel needed the lowest force to initialize and maintain the extrusion, while the microsphere encapsulated hydrogels required similar forces to extrude through the needle, and the maintaining force are around 20N for HA/PMMA and HA/HAp. In contrast, the extrusion force of the HA/PLLA group experienced a gradual to rapid increase, which might be caused by microsphere aggregation and needle obstruction at the later injection stage. Meanwhile, it could be noticed that the loading force curve in HA/HAp was smoother than that in the HA/PMMA group and the slight fluctuations could be restored quickly, which indicated a better particle distribution and tendency not to aggregate in HA/HAp. Rheological testing further revealed that the viscosity of HA/HAp at the same shear rate was highest among the groups, and it decreased dramatically to an equivalent level when the shear rate increased (Fig. 1E). It could be explained by the coarse surface of HAp microsphere which would increase the friction between the materials and raise the threshold force to inject. The good hydrophilic property and even but not agglomerating distribution HAp microspheres were beneficial for the reduction of viscosity at higher shear rates, which meant a lower and stable force to main the injection. These shear-thinning behaviors were consistent with the results acquired from the extrusion force curves.

3.6. Degradation of HAps and composite hydrogel in vivo

The Micro-CT reconstruction images shown in Fig. 2A and the quantitative analysis results shown in Fig. 2C revealed that the degradation rates of HAp300, HAp550, and HAp800 were 91.5 %, 89.3 %, and 2.2 % at 1-month post-implantation, respectively. At 3 months, the corresponding degradation rates increased to 99.5 %, 96.6 %, and 85.7 %, while the degradation rates further progressed to reach 99.8 % for HAp300, 98.9 % for HAp550, and 97.0 % for HAp800 at 6-months. Therefore, it could be inferred that HAp prepared at controlled conditions could be fully degraded in vivo during a limited period, and the higher sintering temperature in a specific range would lead to a longer maintenance time after implantation.

Fig. 2.

In vivo degradation of HAps and hybrid hydrogels and their impact on dermal thickness. A and C, Micro-CT reconstruction images and quantitative results of HAps degradation: Hydroxyapatites exhibited varying degradation patterns attributed to differences in crystallinity. B and D,in vivo ultrasound imaging reconstructions of the subcutaneous occupancy volume of hybrid fillers. E, The increase in dermal thickness of SD rats. F, Schematic illustration of the volume maintenance and tissue regeneration of HAp microspheres over time. Fig. 2C and D: ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. Fig. 2C: n = 3, Fig. 2D: n = 4. Fig. 2E: n = 3, ∗∗p < 0.01 (vs HA), ∗∗∗p < 0.001 (vs HA), ##p < 0.01 (vs HA/PMMA), ###p < 0.001 (vs HA/PMMA), ####p < 0.0001 (vs HA/PMMA), &&p < 0.01 (vs HA/PLLA), &&&p < 0.001 (vs HA/PLLA), &&&& p < 0.0001 (vs HA/PLLA).

The reconstruction images and statistical volume of the composite hydrogels obtained by 3D ultrasound at 1-month post-implantation were given in Fig. 2B and D. It indicated that the average volumes of HA, HA/PMMA, HA/PLLA, HA/HAp300, HA/HAp550, and HA/HAp800 were 164.77 mm³, 181.43 mm³, 196.83 mm³, 224.10 mm³, 224.84 mm³, and 182.81 mm³, respectively. At 3 months, the occupancy volumes of HA, HA/PMMA, HA/PLLA, and HA/HAp800 were slightly decreased, while those in HA/HAp300, and HA/HAp550 were sustained or increased. The dermis thickness could be further statistically analyzed using Image J according to the ultrasound images. The results (Fig. 2E & S10 Video1-18) indicated a more significant increase in skin thickness in the HA/HAp groups compared to other groups, and the increase was particularly pronounced in the HA/HAp300 at both 1 month and 3 months post-implantation.

3.7. Stability and biocompatibility of HAps in vivo

No evident signs of inflammation, redness, or swelling during the observation period following implantation of various material groups, according to the temperature records (Fig. S3) and the appearance of subcutaneous tissue after the exposure of the implants and surrounding tissues (Fig. S4A). H&E staining (Fig. 3A&S4B) showed the appearance of inflammatory cell aggregation bands at the interface of HA/microspheres and tissues on 3rd day, particularly pronounced in the HA/PMMA group. Tissue responses were comparable among the HA/HAp300, HA/HAp550, and HAp800/HA. The distribution of microspheres within the tissue was also discernible, while HAp300 and HAp550 were nearly absent at 3 months compared with those in the same group at 1 month. Moreover, multinucleated giant cells were observed in the HA/HAp800 group and they were absent at 6 months because of the degradation of HAp800. In addition, multiple aggregates of inflammatory cells were visible in the HA/PMMA.

Fig. 3.

The in vivo stability and biocompatibility of implants. A, H&E staining. B, Immunohistochemical staining for CD206, iNOS, and MPO at 3d post-implantation. C, Ratio of M2-type to M1-type macrophages at 3d post-implantation. D, Quantification of neutrophil numbers at 3d post-implantation. E-G, IL-1β, IL-4, and IL-10 protein expression were detected by ELISA at 1M post-implantation, respectively. Yellow arrow: Inflammatory cell aggregation bands at the interface between HA/microspheres and tissues. Green arrow: Inflammatory cell aggregation in the HA/PMMA group. Black arrow: microspheres. Fig. 3C: n = 3, ∗∗p < 0.01 (vs HA), ∗∗∗p < 0.001 (vs HA), ####p < 0.0001 (vs HA/PMMA), &&& p < 0.001 (vs HA/PLLA), &&&& p < 0.0001 (vs HA/PLLA). Fig. 3D–G: n = 3, ∗∗∗∗p < 0.0001, ns: p > 0.05.

Myeloperoxidase (MPO) is a heme-containing peroxidase expressed mainly in neutrophils, which could serve as a vital indicator of early inflammatory response and a marker for neutrophil hyperactivation. IHC staining of MPO in tissue samples harvested on day 3 revealed a higher number of positive neutrophils in the HA/PMMA group compared with other groups (Fig. 3B and D). Following the IHC staining of CD206 and iNOS as shown in Fig. 3B, the ratio of CD206 to iNOS-positive cells was quantified and provided in Fig. 3C. The HA/HAp groups exhibited higher values of CD206/iNOS-positive cells, statistically differing from the other groups, with no significant difference observed between the HA/HAp groups.

One month after the implantation, the IL-1β expression measured by ELISA was significantly higher in the HA/PLLA group compared to the other groups as shown in Fig. 3E, while no significant difference was observed between the HA/HAp300 group and HA group. Meanwhile, the expression of IL-4 was the highest in the PMMA group among the groups, and the HA/HAp300 group also showed significantly higher expression of IL-4 than the other HAp-loading groups in Fig. 3F. In Fig. 3G, IL-10 exhibited comparable high expression in both HA/PMMA and HA/PLLA groups, significantly differing from those in HA/HAp groups, among which HA/HAp800 showed the lowest expression.

3.8. Quality evaluation of regenerative tissue

Regenerative tissues were observed in all groups with material implanted according to the Masson staining results given in Fig. 4A&S5, while a high proportion of fibrous tissues with the characteristic orderly arrangement was presented in the HA/HAp groups. At 9 months after the implantation, as shown in Fig. S7, it could be noticed that abundant PMMA microspheres were aggregated and encapsulated in adjacent fiber tissues in the HA/PMMA group. Moreover, the HA/PLLA exhibited a greater presence of myofibrils within the collagen fibers generated, which showed a more dispersed distribution than that in the HA/PMMA group. Since there was no introduced collagen inside the implants, the positively stained fibers in Masson staining images could be considered as newly secreted and deposited extracellular matrix. Subsequently, the quantitative results acquired using Image J according to the area ratios of positive stained nascent fibers within the Masson staining images were presented in Fig. 4E. It could be found that a high ratio of regenerated fibers after 1-month implantation of HA/HAp300. After 3 months of implantation, the proportions of regenerated tissues generally increased but varied in different groups. The new fibers in HA were rare and significantly lower than that in other groups, while the proportions of fibers were pronounced increased in HA/HAp300 and HA/HAp550 and significantly higher than that in other groups. At 6 months post-implantation, a significant increase of regenerated tissue amount was noted in HA/HAp800, while the regenerated tissue in HA, HA/HAp300, and HA/HAp550 decreased compared to those at 3 months post-implantation. Meanwhile, the new fiber proportions in both HA/PMMA and HA/PLLA increased at 3M and maintained at 6M compared to the previous time point. Fig. 2F illustrates a schematic representation of HAp microsphere volume maintenance and tissue regeneration over time.

Fig. 4.

HAp significantly and sustainably promoted the secretion and deposition of matured ECM (3M). A, Masson's trichrome staining. B, Sirius red staining. C, Scanning electron microscope images. D, Transmission electron microscopy images. E, Quantification of the positive area ratio of nascent collagen fibers within the materials based on Masson's trichrome staining results. 1, HA; 2, PMMA; 3, PLLA; 4, HAp300; 5, HAp550; 6, HAp800. n: fibroblast nuclei; Arrow: collagen fibers; white rectangular outlines: cytoplasmic vacuolization. n = 3, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Sirius red staining was applied to observe the production of type I and type III collagen fibers within the implanted material shown in Fig. 4B and S6, which revealed a similar tendency of fiber production with the findings from Masson's staining. In the HA/HAp groups, collagen fibers were observed at 1 month post-implantation, and the positive staining increased at 3 and 6 months. Also, the collagen fibers exhibited a spatial arrangement similar to native ECM and remodeled to a longer and thicker condition in HA/HAp groups as time prolonged. In contrast, the HA group displayed few amounts of new collagen fibers at all time points. In the HA/PMMA group, abundant fibrous bundles were generated and characterized by surrounding the implant and disorganized fibers. Meanwhile, the change of collagen fibers in the HA/PLLA group experienced a moderate process during the observation course, in which the collagen amount was smaller but the fibers were better aligned, especially at 6 months post-implantation.

SEM images were provided to examine the distribution and morphology of collagen and elastic fibers around the material (Fig. 4C and S7). In the newly implanted composites (1D after the injection), all the microspheres were separated from the surrounding and the surface was clear and intact. After 1 month, the ECM was deposited and embedded HAp microspheres in three HA/HAp groups, and the collagen fiber bundles were composed of multiple fibrils and arranged in a structured manner, while the finer elastic fibers interwove into a network around the collagen fibers. While the fiber bundles in HA/HAp300 were thicker than those in the other groups. With the increase in implantation time, the secretions in HA/HAp550 and HA/HAp800 were thicker after 3 and 6 months, while HAp microspheres were occasionally visible in HA/HAp300. In the control group, a continuous HA sheet could be observed and little regenerative tissue was formed on the surface and around the HA material throughout the period. The PMMA and PLLA microspheres with smooth surfaces could be easily found at different stages, with fibers distributed around the microspheres. In addition, the fibers surrounding PMMA microspheres were disorganized and not thick bundles as observed in HA/HAp groups.

TEM results in Fig. 4D revealed that fibroblasts extensively surrounded the HAp in the HA/HAp group. These fibroblasts exhibited a spread cytoskeleton, robust pseudopods, and a nucleus with uniformly distributed chromatin. The cytoplasm was abundant with normal organelles, including mitochondria and rough endoplasmic reticulum. It was important to notice that the collagen fibers embedded in the microspheres displayed a distinct pattern of stripes (as the arrow indicated), which was formed by a sequence of more dense and less dense fiber regions and was a characteristic of native collagen with advanced structures. In contrast, the fibroblasts in the HA group had a limited spread-out area, thin pseudopods, and a scarcity of surrounding collagen fibers. The fibroblasts in the HA/PMMA and HA/PLLA groups showed significant cytoplasmic vacuolization and a lack of pericellular collagen fibers. A few amounts of collagen fibers could be found in distant areas of PMMA microspheres, and the fibers did not display a similar pattern of stripes observed in HA/HAp groups.

3.9. Regenerative mechanisms

Comparing the tissue transcriptome data between the HA/HAp group and HA group, the differential analysis revealed 884 genes were differentially expressed, with 476 upregulated and 408 downregulated as the volcano map shown in Fig. 5A. KEGG pathway analysis (Fig. 5B) indicated that differentially expressed genes related to HA/HAp primarily enriched the calcium signaling pathway and motor protein pathway among others. According to Gene Set Enrichment Analysis (GSEA, Fig. 5C and D), these pathways' gene datasets are significantly clustered at the top of the total gene expression dataset. This analysis preliminarily suggested the calcium ion signaling and motor protein pathways were upregulated in the HA/HAp group.

Fig. 5.

RNA sequencing results of tissues harvested at 3M post-implantation (HA/HAps vs. HA groups). A, Volcano plot illustrated gene expression differences. B, KEGG pathway analysis. C, Gene Set Enrichment Analysis (GSEA) of the calcium signaling pathway. D, GSEA of motor proteins. E-G, GO enrichment analysis for biological process (BP), cellular component (CC), and molecular function (MF).

Further, GO enrichment analyses were performed for biological processes (BP, Fig. 5E), cellular components (CC, Fig. 5F), and molecular function (MF, Fig. 5G). The results revealed that HA/HAp-related differentially expressed genes predominantly contributed to BP such as tissue regeneration. Differential genes related to CC upregulated by HA/HAps were primarily found in the actin cytoskeleton, actin filament, calcium channel complex, and voltage-gated calcium channel complex. To MF, the differentially expressed genes exhibited the HAp accelerated expression of calmodulin binding and actin binding.

The comparison of RNA sequencing results between the HA/HAp group and HA/PMMA group (Fig. S8), and the HA/HAp group and HA/PLLA group (Fig. S9) illustrated consistently that both calcium ion signaling and motor protein pathways were up-regulated, and immune rejection-related signaling pathways (such as Natural killer cell-mediated cytotoxicity, Allograft rejection, and Graft-versus-host disease) were down-regulated in the HA/HAp group.

The immunofluorescence staining results of the implants and surrounding subcutaneous tissues after 3 days of implantation was shown in Fig. 6A. The mean fluorescence intensity (MFI) quantified by Image J revealed elevated expression of basic FGF2 and S100A4, as given in Fig. 6B and C. Both FGF2 and S100A4 were notably higher expressed in the HA/HAp300 and HA/HAp550 groups, significantly differing from other groups. After 3 months of implantation, according to the immunofluorescence staining and quantitative results given in Fig. 6A–D, and E, the FGF2 and S100A4 expression inside of the implant in HA/HAp550 significantly surpassed those of other groups.

Fig. 6.

Tissue immunofluorescence results demonstrated the expression of FGF2 and S100A4 proteins in both the exterior and interior regions of the implantations. A, The expression of FGF2 and S100A4 proteins on the exterior and interior of implants at 3d post-implantation. B-C, Mean fluorescence intensity (MFI) of FGF2 and S100A4 protein expression on the exterior of implants at 3d post-implantation, respectively. D-E, The MFI of FGF2 and S100A4 protein expression on the interior of implants at 3M post-implantation, respectively. Dotted lines: The interface between implants and tissues (IIT). n = 3,∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

The ELISA assay results of tissue samples harvested 1M after implantation are shown in Fig. 7. S100A4 protein was significantly higher in HA/HAp300 compared to all other groups (Fig. 7A). MYH9 expression was notably elevated in both HA/HAp300 and HA/PLLA, showing significant differences from other groups (Fig. 7B). Adhesion-associated proteins Itg αⅤ, Itg β1, and Vinculin were highly expressed in HA/HAps, particularly in the HA/HAp800 and HA/HAp300 groups. Notably, Vinculin expression did not significantly differ between HA/PLLA and HA/PMMA (Fig. 7C, D, E).

Fig. 7.

Expression of proteins related to pro-tissue regeneration at 1M post-implantation. A-E, S100A4, MYH9, Itg αⅤ, Itg β1, and vinculin protein expression were detected by ELISA, respectively. n = 3,∗p < 0.05, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, ns: p > 0.05.

4. Discussion

HAp has been applied for bone regeneration for many decades and the safety and efficacy have been adequately validated, but the application history for facial rejuvenation is finite, and the slow or even resistant degradation is suspected to be suitable as the composition of dermal filler [[33], [34], [35], [36]]. In this research, three HAps with different crystallinity were synthesized by manipulating the sintering temperature and implanted subcutaneously to understand the performance in vivo. The so-prepared HAp300, HAp550, and HAp800 were sieved to attain a particle size ranging from 20 μm to 50 μm, to avoid the quick phagocytosis by immune cells which might be caused if the particle size was smaller than 20 μm, and to reduce the immune-inflammatory reactions which tended to occur when they were larger than 50 μm [37]. To load the HAp and other microspheres into subcutaneous safely and effectively, BDDE cross-linked hyaluronic acid (HA) hydrogel which was one of the most widely applied dermal fillers at present was chosen as a carrier to prepare the microsphere-encapsulated composite hydrogels. These composite hydrogels were characterized and further implanted in vivo to study the tissue regeneration and the possible mechanism underlying.

4.1. Variations in degradation patterns of HAps at different sintering temperatures

An ideal dermal filler to achieve facial rejuvenation should possess balanced durability and biodegradability, which would maintain the face contours effectively and not cause nodules after a long period of implantation [2,38]. The microstructures of HAp, such as crystallinity, density, grain size, and microporosity, were determined by prepare technology and parameters including sintering temperature, and in turn affected other physical and chemical properties [36,39]. In this study case, the HAps were sintered at a temperature lower than 1100 °C which was considered as the formation temperature of calcium phosphate ceramic crystals. Below the crystal forming temperature, the crystallinity of HAp would increase as the sintering temperature raised, while the density and grain size would increase, and the specific surface area and contact angle would decrease [40]. As a result, the HAp sintered at a higher temperature below the crystal forming temperature had lower solubility and thus released lower Ca²⁺ concentration at the same condition than HAp prepared at a lower sintering temperature [[40], [41], [42]]. The results of this study showed that HAp800 microspheres had the largest grain size (Fig. 1A), the highest crystallinity (81.95 %), and the lowest calcium ion concentration (Fig. 1D) compared to those of HAp300 and HAp550. The high crystallinity and slow degradation of HAp800 led to the extended retention time to 6 months in vivo, as the Micro-CT results illustrated (Fig. 2A–C). Therefore, the durability and biodegradation of HAp-based hydrogel could be modulated by the control of preparation parameters. It was also noticed that the released Ca²⁺ concentrations of HAp300 and HAp550 at 120h were decreased than those at the previous time points, which could be explained that calcium ions were redeposited on the surface of HAp to form a calcium-rich layer after the saturated dissolution of calcium ions.

4.2. Biocompatibility evaluation and inflammation modulation effect of HAps

The in vitro cell culture results followed standard sample preparation and the cytocompatibility assay indicated that HAp-encapsulated hydrogels were cytocompatible and could promote HSF spreading and proliferation. The in vivo tissue compatibility was evaluated since it was inversely proportional to the magnitude and duration of the reaction with the host produced by the material implantation. In this study, the inflammatory cell aggregation bands at the interface of implants and tissues (yellow arrows in Fig. 3A) suggested that acute inflammation had been activated in all groups 3 days after the implantation, which could be considered a normal and necessary function of the innate immune system. The retention of neutrophils in the tissue for 3 days or more was considered one of the most important features of the acute inflammatory response. Generally, the implants would be infiltrated by neutrophils and cause the deposition of plasma proteins rapidly after the recognition of foreign materials [[43], [44], [45], [46]]. The significantly bigger number of neutrophils in HA/PMMA groups than those in other groups meant a stronger acute inflammatory response with PMMA microspheres, while PLLA and HAps generated comparable acute inflammation. Then, neutrophils would remove the foreign material by phagocytosis if they could, otherwise, they would secrete various inflammatory factors, such as IL-1β, IL-4, and IL-10, which could stimulate the migratory differentiation of monocytes into macrophages and the activation of pro-inflammatory state M1 macrophages. M1 macrophages would feedback to increase neutrophil recruitment and inhibit neutrophil apoptosis by releasing pro-inflammatory factors, which further prolong and increase inflammatory severity [43,46]. The ELISA results verified that the inflammatory factors were maintained at a relatively high level in groups with polymer microspheres, indicating the long-lasting inflammation and increase of M1 macrophages in these groups. In contrast, the inflammatory factor expressions in HAp groups were generally lower, suggesting a moderate inflammation as that in the HA group. After existing in and around the foreign material at the early stage of the inflammatory response, neutrophils would undergo apoptosis and release signals to up-regulate macrophage phagocytosis-related gene expression, and further promote the transformation of M1 macrophages to anti-inflammatory M2 macrophages [43,46,47]. The ratio of M2 to M1 macrophages which was referred to as the ratio of characteristic protein CD206 to iNOS signified that the implantation of HAps induced more M2 macrophages to anti-inflammation and then recruit fibroblasts to gather and promote tissue regeneration. Therefore, early neutrophil apoptosis and M2 macrophage conversion could be modulated by the implantation and were important for tissue regenerative repair [45].

In addition, the macrophages usually underwent a fusion to form multinucleated foreign-body giant cells if the implant was too large to be phagocytosed by macrophages. These multinucleated giant cells could lead to chronic inflammation and cause adverse reactions such as nodular granulomas in the long term [43,[48], [49], [50]]. In this study, a few foreign body giant cells appeared near the aggregated HAp microspheres at the early implantation stage, but they disappeared and no granuloma was produced because of the degradation of HAps under 6M long observation. In comparison, substantial aggregates of inflammatory cells were observed surrounding the undegraded PMMA microspheres, suggesting the potential development of nodular granulomas. This observation indicated that non-degradable microspheres represented inferior biocompatibility, which bore a higher risk of inducing prolonged inflammation and fibrous capsule formation. Currently, numerous cases of chronic inflammation attributed to PMMA have been documented in the literature [16,20,21].

On the other hand, the inflammation in HA/PLLA was moderate and most of the indexes fell in between HA/HAps and HA/PMMA. It was acceptable since mild chronic inflammation was generated owing to the gradual degradation of PLLA, although this synthesized polymer caused acute inflammation at the early stage of implantation.

Surface hydrophilicity is another critical property of biomaterials that significantly influences the immune-inflammatory response. Studies have demonstrated that hydrophilic surfaces can resist protein adsorption, thereby preventing the formation of a protein corona that may trigger pro-inflammatory reactions [51,52]. This promotes a microenvironment conducive to anti-inflammatory macrophages. Hotchkiss et al. further verified that increased surface roughness enhanced both M1 and M2 macrophage activation. However, when roughness was combined with hydrophilicity, the expression of pro-inflammatory markers was markedly suppressed, while anti-inflammatory markers were notably upregulated [53]. In this study, the incorporation of HAp modulated the hydrogel's hydrophilicity and roughness compared to control microspheres, thereby mitigating excessive immune responses. In addition, the degradation properties of HAp microspheres help prevent excessive immunoinflammation.

4.3. Promotion of tissue regeneration and matrix remodeling effect modulated by HAps

The promotion of collagen secretion, deposition, and remodeling should be considered an important aspect of tissue regeneration and efficacy of implantable medical devices [54]. As the carrier of different microspheres in this study, BDDE crosslinked HA hydrogel showed slight acute and chronic inflammation, but the cell migration into the hydrogel and the new collagen secretion in HA hydrogel was deficient owing to an insufficient supply of bioactive sites for cell adhesion [[55], [56], [57], [58], [59]]. Consequently, the distinctive ECM forming patterns and tissue regenerative results observed among the groups in this study were attributed to the different biocompatibility and degradation behaviors of encapsulated microspheres. The non-degradable and bioinert PMMA microspheres stably remained in vivo during the experimental period. Although newly deposited collagen was observed in the HA/PMMA group, the matrix was not well aligned and mainly constructed the fibrous capsule which would lead to a high risk of granuloma. As a biocompatible and biodegradable polymer, PLLA had been widely applied as the starting material in tissue engineering, but its degradation rate was low and the regenerative dermal matrix was limited after subcutaneously implanted for 6M.

In comparison, hydroxyapatite microspheres (HAp300, HAp550, HAp800) exhibited various dissolution and degradation patterns due to their distinct degrees of crystallinity, leading to inconsistent calcium ion release concentrations at the same time as shown in Fig. 1D. Based on the observed correlation between the retention of HAp microspheres and the volume of regenerative tissue, it was hypothesized that the tissue regeneration was closely associated with the degradation of HAp. Specifically, HAp300 exhibited rapid degradation within the first month post-implantation, coinciding with the highest newly secreted collagen among all groups (Fig. 4E). HAp550 showed a slightly delayed degradation in vivo, achieving the peak collagen production at 6M post-injection. HAp800, characterized by the highest degree of crystallinity in this study, initialized the degradation process after 1M and fully degraded at 6M, leading to a gradual increase in new collagen production up to 6M. Consequently, it could be deduced from the degradation rules and collagen secretion patterns that the ECM regeneration proceeded in tandem with HAp degradation, albeit with a slight hysteresis.

Furthermore, as the local distribution and advanced structure of newly formed collagen illustrated in Fig. 4D, it was rational to infer that HAp and its degradation established a suitable microenvironment for fibroblasts for two reasons. Firstly, the hydrophilic and biomimetic components of HAp provided adequate adhesion sites for cells, which also improved the cell migration inside the implant and further even the distribution of newly secreted collagen. Secondly, the timely degradation of HAp and proper concentrations of released calcium and phosphate activated the function expression of fibroblasts and made spaces for the deposition of new ECM. Comparatively, the collagen fibers generated in HA/PMMA and HA/PLLA were not as thick and evenly distributed as those in HAp groups. Also, there was almost no fiber with characteristic stripes observed in a distant area from both PMMA and PLLA microspheres. The primary cause could be the non-degradability of PMMA and the unbefitting degraded products from PLLA. Therefore, the quantity increase and the quality improvement of the new ECM could be mainly considered as the consequence of different implanted microspheres, which were dependent on the intrinsic properties of the materials.

4.4. Tissue regeneration facilitated by activation of calcium signaling and motor protein pathways through calcium ion release from HAps

It has been investigated that biological molecules and proteins could be adsorbed on the surface of hydroxyapatite because of its chemical compositions and physical structures, which affected the establishment of the cell microenvironment and further cell phenotype. By contrast, the hydrophobic surfaces of PMMA and PLLA could not respond to the biological clues and exhibit deficiency in constructing a niche for cells. Besides, the different degradation behaviors and products of the microspheres initialized various cell responses, among which calcium signaling and motor protein pathways activated by calcium ions released from HAps were mainly discussed in this study.

Calcium ions serve as one of the most significant intercellular and intracellular messengers. The calcium ion signaling pathway is an important cellular communication pathway found universally across all eukaryotic organisms [[60], [61], [62], [63], [64], [65], [66], [67], [68], [69]]. The gradually released calcium ions from HAps would upregulate the calcium ion signaling pathways by entering calcium ions into cells via complex cell signaling processes, including receptor activation and secondary signaling, voltage-gated calcium ion channels, ligand-gated calcium ion channels, intracellular calcium store release, and intracellular calcium-binding protein. The upregulations of partial signaling pathways were discovered as shown in Fig. 5. Thus, the released calcium ions from HAp could trigger the upregulation of calcium ion signaling pathways and influence many cellular functions, including cell proliferation, differentiation, migration, etc.

FGF2 was known as a multifunctional growth factor that played a significant role in fibroblast and endothelial cell proliferation and differentiation [[70], [71], [72], [73]]. Furthermore, it was especially important in the calcium ion signaling pathway. FGF2 could influence calcium channel activity to regulate the influx of calcium ions through the cell membrane. Also, it could impact the activity of intracellular calcium pumps and sodium/calcium exchangers to modulate the intracellular calcium ion concentrations [[74], [75], [76], [77]]. Under the appropriate stimulation of calcium ions released by HAps, the high expression of FGF2 was confirmed by immunofluorescence staining as shown in Fig. 6, and the FGF/FGFR signaling pathway was further activated, which targeted tyrosine kinase receptors to regulate immune processes and enhanced the expression of S100A4. As a calcium-binding protein could sense and transduce calcium signals, S100A4 played a central role in the calcium signaling pathway and was significantly more expressed in HAp groups than in other groups as shown in Fig. 6, Fig. 7. S100A4 also could interact with cytoskeletal proteins to influence cell movement and morphological changes, with the regulation of calcium ions. The highly expressed S100A4 and activated calcium ion signaling pathway could promote the expression of integrins and focal adhesion-related proteins such as Itg αV, Itg β1, and Vinculin, which could enhance cell adhesion, migration, and spreading [[78], [79], [80], [81], [82], [83], [84], [85], [86], [87], [88]]. Concurrently, the motor protein signaling pathway was activated and a representative protein MYH9 was upregulated. MYH9, also known as non-muscle myosin II A (NMII-A), was a subtype of non-muscle myosin II (NMII) which was an ATP-dependent motor protein and affected the processes requiring force generation and actin cytoskeleton translocation. The enhanced expression of MYH9 could strengthen the interaction with actin microfilaments, cellular microtubules, and calcium-binding proteins. Its participation in forming polar focal adhesions, stress fibers, and regulating cell contractility through phosphorylation of its heavy and light chains could finally impact cell adhesion and migration [[89], [90], [91], [92]].

It also noticed some abnormal high protein expressions in ELISA results (that is S100A4 and MYH9 in the HA/PLLA group), which might be caused by an unknown bypath or mechanism related to the degradation products of PLLA. Nevertheless, according to the evidence and analysis on signaling pathways and crucial signaling proteins, it could be preliminarily concluded the mechanism of dermal skin regeneration modulated by HAps as illustrated in Fig. 8. The implantation of HA/HAps accelerated the polarization of macrophages from M1 towards M2 via quick neutrophils apoptosis, which recruited and activated fibroblasts by enhancing the expression of FGF2. The FGF2/FGFR signaling pathway then induced the expression of S100A4 to modulate the calcium ion signaling pathway and following motor protein pathway. With the secretion of downstream proteins and molecules, the cell phenotype was regulated and finally achieved ECM deposition and remodeling.

Fig. 8.

Schematic of regenerative mechanisms induced by HAps encapsulated in hybrid hydrogels. Following the implantation of HA/HAps into the body, neutrophils underwent rapid and programmed apoptosis, which facilitated the polarization of M1 macrophages towards M2 macrophages. This polarization recruited and activated fibroblasts by enhancing the expression of FGF2. Subsequently, the activation of the FGF2/FGFR signaling pathway induced the expression of S100A4, thereby modulating the calcium ion signaling pathway. This process promoted the expression of integrins and focal adhesion-related proteins such as Itg αV, Itg β1, and Vinculin. Concurrently, it upregulated the motor protein signaling pathway, leading to increased expression of MYH9. Collectively, these mechanisms enhanced cell adhesion, migration, and spreading, ultimately facilitating skin regeneration and repair.

Finally, the mechanism framework of HAp on skin regeneration could be summarized. HA/HAp was first recognized by neutrophils and promoted plasma protein deposition post-implantation, which promoted neutrophil apoptosis in the short term and released signals to up-regulate macrophage phagocytosis-related gene expression. This process further promoted the conversion of M1 macrophages to anti-inflammatory M2 macrophages. M2-type macrophages enhanced FGF2 expression, activating the FGF2/FGFR signaling pathway to recruit fibroblasts. Subsequently, activation of the FGF2/FGFR signaling pathway induced S100A4 expression, which regulated the calcium signaling pathway, thus enhancing the expression of integrins and focal adhesion-associated proteins and consequently improving cell adhesion, proliferation, migration, differentiation, and angiogenesis. Meanwhile, the calcium signaling pathway activated the motor protein signaling pathway and up-regulated the expression of MYH9 protein, which enhanced the interaction with actin microfilaments, cellular microtubules, and calcium-binding proteins, participating in the formation of polar focal adhesion, stress fibers, and regulating cellular contractile force, ultimately affecting the cytoskeleton, adhesion, and migration. Furthermore, HAp, due to its unique physical structure properties, promoted cell adhesion and migration more effectively compared to PMMA and PLLA microspheres with smooth and hydrophobic surfaces. Collectively, these effects facilitated ECM assembly and remodeling.

5. Conclusions

Hydroxyapatite has a long and successful history of application in bone tissue regeneration; however, its safety and efficacy in soft tissue regeneration remain underexplored. Given the stereotypical application of HAp for bone regeneration, where it is characterized by difficulty or even non-degradability, concerns have been raised regarding its long-term safety and efficacy in soft tissue applications, particularly concerning the degradability of HAp microspheres as dermal fillers.

In this study, a long-term and in-depth investigation of HAp microspheres with tailored physicochemical properties was conducted to evaluate their performance in soft tissue regeneration. HAp microspheres were encapsulated within crosslinked hyaluronate hydrogels and subsequently injected subcutaneously to investigate the biocompatibility and skin regeneration effects, using PMMA and PLLA microspheres - commercially applied as dermal fillers in clinical practice - as controls. The results suggested that the HAps with controlled crystallinity could degraded in vivo at varying rates, exerting corresponding effects on collagen assembly and remodeling. Transcriptomics and molecular biological analysis preliminarily elucidated that calcium ions released from HAp activated and upregulated calcium signaling and motor protein pathways, thereby modulating cell adhesion, migration, proliferation, and gene and protein expression. The observed increased content of collagen with advanced structures suggested that skin regeneration was achieved to some extent. In contrast, tissue regeneration in the HA/PMMA and HA/PLLA groups was limited quantitatively and qualitatively. These findings revealed the potential of HAp microspheres in soft tissue regeneration applications.

In future studies, HAp microspheres-based subcutaneous implants could be further optimized for enhanced stability and ease of operation, supported by a deeper understanding of the mechanism underlying HAp's effects on soft tissue regeneration. Additionally, their safety and efficacy could be further verified through comparative analyses with commercially available dermal fillers.

CRediT authorship contribution statement

Shuo Liu: Writing – review & editing, Writing – original draft, Methodology, Investigation, Formal analysis, Data curation. Lu Song: Investigation. Shuwen Huang: Investigation. Zhanhong Liu: Investigation. Yang Xu: Investigation. Zhiyuan Wang: Investigation. He Qiu: Investigation. Jing Wang: Investigation. Zhiru Chen: Investigation. Yumei Xiao: Methodology. Hang Wang: Methodology, Conceptualization. Xiangdong Zhu: Writing – review & editing, Supervision, Project administration, Methodology, Data curation, Conceptualization. Kai Zhang: Resources. Xingdong Zhang: Supervision, Resources. Hai Lin: Writing – review & editing, Supervision, Project administration, Methodology, Data curation, Conceptualization.

Data and materials availability

All data are available in the main text or the supplementary materials.

Ethics approval and consent to participate

No human subjects and clinical samples were involved in the reported work. The use and care of experimental animals were approved by the Animal Care and Use Committee of Sichuan University (Approval No. WCHSIRB-D-2024-530) and adhered to the animal welfare guidelines established by the Chinese Society for Laboratory Animals.

Funding

This research was funded by the National Key Research and Development Program of China (Grant Nos. 2022YFC2409803 and 2022YFC2409804).

Declaration of competing interest

Kai Zhang is an editorial board member for Bioactive Materials and was not involved in the editorial review or the decision to publish this article. All authors declare that there are no competing interests.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Multimedia component 1

figs1.

figs2.

figs3.

figs4.

figs5.

figs6.

figs7.

figs8.

figs9.

figs10.