Abstract

   The effects of calcium phosphate (CaP) materials on macrophage
   polarization state vary with their physicochemical properties. The
   study aims to elucidate the impact of phosphate ion‐mediated energy
   metabolism on M2 macrophage polarization and the corresponding
   regulatory mechanism. The phosphate ions released from CaP ceramic as
   bioenergetic factor is identified; its concentration is closely
   associated with the polarized state. After being taken up by the
   sodium‐dependent phosphate transporter 1, extracellular phosphate ions
   produce energy via oxidative phosphorylation by facilitating
   tricarboxylic acid flux, thereby contributing to M2 macrophage
   polarization. Further mechanistic analysis reveals that the elevation
   of the bioenergetic basis can drive macrophage M2 polarization via the
   AMP‐activated protein kinase‐mammalian target of rapamycin (AMPK‐mTOR)
   axis. Another regulatory effect is that of the adenosine triphosphate
   (ATP), a signaling molecule. Intracellular ATP is released into the
   extracellular space and degraded to adenosine, which serves as a
   signaling molecule through the A2b adenosine receptor to activate the
   cyclic adenosine monophosphate (cAMP) pathway, thereby promoting M2
   macrophage polarization. Overall, these findings may transform the
   existing knowledge on cell metabolism and energy homeostasis from
   bystanders to pivotal factors guiding M2 macrophage polarization and
   have implications for the future design of biomimetic CaP scaffolds.

   Keywords: AMPK‐mTOR axis, CaP scaffold, energy metabolism, macrophage
   polarization, phosphorous
     __________________________________________________________________

   Phosphate ions released from CaP ceramics, after being taken up by
   SLC20a1, produce energy via OXPHOS by facilitating TCA flux, thereby
   promoting M2 macrophage polarization. This is likely due to increased
   energy availability, which activates mTOR signaling by decreasing AMPK
   activity. Moreover, ATP serves as an extracellular autocrine and
   paracrine signaling molecule that promotes M2 macrophage polarization
   via A2b adenosine receptor.

   graphic file with name ADVS-11-2306062-g001.jpg

1. Introduction

   The immune microenvironment of bone regeneration sites is crucial for
   reconstructing the tissue regeneration process. A substantial body of
   evidence supports the pivotal role of macrophages in various
   wound‐healing processes. Additionally, it has been shown that an
   advanced paradigm for regulating macrophages is essential for achieving
   successful bone regeneration.^[ [32]^1 , [33]^2 , [34]^3 ^]
   Biomaterials‐based bone substitutes are widely used for bone tissue
   regeneration in large bone defects. Our study, along with others, has
   shown that biomaterials incorporating calcium phosphate (CaP), a key
   inorganic component of natural bone tissue, can promote bone tissue
   formation in vivo.^[ [35]^4 , [36]^5 , [37]^6 ^] However, the
   effectiveness of these materials in repairing bone defects is closely
   linked to the response and polarization states of macrophages.
   Therefore, to effectively utilize CaP‐based materials for bone defect
   repair, it is crucial to understand both the osteoimmunological status
   and molecular mechanisms involved in macrophage polarization within a
   CaP mineral environment.

   CaP biomaterials usually tend to degrade calcium and phosphate ions
   (i.e., Ca^2+ and PO[4] ^3−), forming a microenvironment capable of
   regulating bone regeneration.^[ [38]^7 ^] Among the various
   biomaterials containing CaP, those that easily dissociate into Ca^2+
   and PO[4] ^3− ions have been found to enhance bone healing. This
   observation is further supported by research indicating that exposing
   osteoblasts and progenitor cells to a medium rich in either Ca^2+ or
   PO[4] ^3− promotes osteogenic differentiation.^[ [39]^8 , [40]^9 ,
   [41]^10 ^] Despite numerous studies highlighting the potential impact
   of CaP minerals, Ca^2+, and PO[4] ^3−, on osteogenic performance, a
   majority of molecular investigations have primarily focused on the
   regulatory role of Ca^2+. For example, Wen et al. have reported that an
   influx of extracellular Ca^2+ via L‐type calcium channels would
   accelerate osteogenic differentiation of osteoprogenitor cells and bone
   formation.^[ [42]^11 ^] Additionally, PO[4] ^3− has a critical, but
   often overlooked, role in supporting osteogenesis. In addition to
   apatite formation, studies have shown that PO[4] ^3− is an important
   driver of osteogenic processes caused by an increase in cellular energy
   metabolism.^[ [43]^9 , [44]^12 ^] Extracellular PO[4] ^3− are
   transported by sodium‐dependent phosphate transporter 1 (SLC20a1, or
   PiT‐1), a sodium‐phosphate symporter that transports PO[4] ^3− ions
   from the extracellular milieu into the cytoplasm and participates in
   the tricarboxylic acid (TCA) cycle and oxidative phosphorylation
   (OXPHOS), resulting in elevated intracellular adenosine triphosphate
   (ATP) levels.^[ [45]^13 , [46]^14 ^] The expansion of CaP matrices to
   investigate the impact of PO[4] ^3− on osteogenesis has revealed that
   an elevation in intracellular ATP levels is advantageous for initiating
   a signaling cascade that prompts MSC differentiation into the
   osteoblastic lineage in response to CaP.^[ [47]^15 , [48]^16 ^]
   Additionally, Liu et al. and Bossche et al., have substantiated that
   ATP release is an efficacious approach for regulating energy demand
   during the initial phase of tissue repair.^[ [49]^17 , [50]^18 ^]

   Recent advancements have underscored the pivotal significance of energy
   requirements in the activation state and related functions of
   macrophages.^[ [51]^19 ^] Cell transformation frequently involves
   changes in energy metabolism that facilitate adaptation to cellular
   functions and reprogramming interactions within the cellular
   environment.^[ [52]^20 , [53]^21 ^] M1 macrophages exhibit metabolic
   reprogramming toward glycolysis, whereas M2 macrophages rely on OXPHOS
   for energy production (i.e., ATP). Consequently, aerobic glycolysis and
   OXPHOS play distinct roles in regulating the macrophage phenotype and
   function.^[ [54]^22 ^] Despite PO[4] ^3− having shown positive
   enhancement in boosting osteogenic differentiation via modulating
   intracellular energy metabolism,^[ [55]^23 , [56]^24 ^] the role of
   PO[4] ^3− in modulating the metabolic programming of macrophages is
   unclear. Considering the positive role of PO[4] ^3‐ in OXPHOS, PO[4]
   ^3− ions from the extracellular milieu may provide biochemical and
   bioenergetic basis for M2 macrophage polarization. Therefore, it is
   imperative to investigate the impact of PO[4] ^3− on macrophage
   polarization, as well as the connection between the role of
   extracellular PO[4] ^3− in cell metabolism, final polarization fate
   selection, and the corresponding molecular mechanisms.

   Here, we identified this central link in which the CaP mediated‐rich
   PO[4] ^3− environment, taken up through SLC20a1, targeted the energy
   metabolism to shape the transition of macrophages into M2 subsets, an
   effect referred to as “metabolic regulation”. Our study showed that
   extracellular PO[4] ^3− was critical to promote polarization of
   macrophages toward regenerative M2 macrophages, by producing energy in
   the form of ATP through the TCA cycle and OXPHOS. Increased energy
   availability resulted in the activation of the mammalian target of
   rapamycin (mTOR) signaling by a decrease in AMP‐activated protein
   kinase (AMPK) activity, and subsequently switched on M2 macrophage
   polarization. Moreover, ATP, secreting into the extracellular
   environment and metabolizing into adenosine, served as extracellular
   autocrine and paracrine signaling molecules to promote the polarization
   of M2 macrophages via A2b adenosine receptor. Overall, this study
   presents a mechanism to demonstrate the beneficial role of phosphate
   ions in macrophage functional phenotypes and ultimately inspires the
   development of novel CaP biomaterial designs toward metabolic
   strategies.

2. Results and Discussion

2.1. Phosphate Degradability of CaP Ceramics

   Two CaP ceramics (hydroxyapatite (HA) and β‐tricalcium phosphate
   (β‐TCP)), with porosity of ≈70% with microporous structures are shown
   in Figure [57]1 . The energy dispersive spectroscopy (EDS) analysis
   confirmed the corresponding Ca/P ratio for HA and β‐TCP (Figure [58]S1,
   Supporting Information). X‐ray diffraction (XRD) analysis provided
   further confirmation that they were indeed HA and β‐TCP, as evidenced
   by the specific diffraction peaks observed at 2θ = 31.81° (HA) and 2θ =
   31.07° (β‐TCP) (Figure [59]1A). The surface morphologies of the CaP
   ceramics were examined before and after immersion. After 7 days of
   immersion in Tris‐HCl buffer solution (pH 7.4), the scaffold topography
   changed considerably to varying degrees (Figure [60]1B). Furthermore,
   the abilities of the HA and β‐TCP ceramics to release PO[4] ^3− in the
   Tris‐HCl buffer solution were evaluated (Figure [61]1C). The PO[4] ^3−
   released from β‐TCP ceramic was significantly higher than that from HA
   ceramic. The concentrations of PO[4] ^3− continued to increase with
   incubation time, reaching 15.1 ppm at day 7 in β‐TCP ceramics.
   Combining the results of surface change and PO[4] ^3− releasing
   abilities, it could be concluded that β‐TCP ceramic had stronger
   degradation abilities than HA ceramic, which was consistent with
   previous reports.^[ [62]^7 ^] Aside to the physical cues provided, it
   was important to consider the dissolution kinetics and the ability of
   CaP ceramics to modulate the extracellular mineral environment through
   dissolution/formation processes, as these factors significantly
   influenced the biological functions of CaP materials, which have often
   been overlooked. Our results indicated that the β‐TCP ceramic exhibited
   a stronger ability to create a CaP‐rich environment than HA ceramic,
   with more and faster PO[4] ^3− ions released.

Figure 1.

   Figure 1
   [63]Open in a new tab

   Characterization of CaP ceramics. A) XRD graph of HA and β‐TCP
   scaffolds. B) SEM morphologies of HA and β‐TCP scaffolds before and
   after 7‐day immersion in Tris‐HCl solution. C) Phosphate ionic
   concentrations released from HA and β‐TCP scaffolds (n = 3).

2.2. CaP Ceramics Enhance M2 Macrophage Polarization Through SLC20a1

   We then sought to uncover the role of phosphate ions released from the
   CaP ceramics in macrophage polarization toward the M2 subtype. The rate
   of M2 polarization was determined using flow cytometry (Figure
   [64]2A,B). The percentage of M2 polarization gradually increased with
   the increase of incubation time, indicating a shift in macrophage
   polarization toward the M2 subtype. Moreover, there was a significantly
   higher proportion of M2 polarization in the β‐TCP group, with high
   PO[4]^3−‐rich environment, compared with that in the HA group. At 5
   days, the percentage of CD206^+ cells in the β‐TCP group wase as high
   as 50%, with the ratio being ∼ 2‐fold greater for HA group, indicating
   that β‐TCP scaffolds outcompeted HA scaffolds for modulating the
   polarization of macrophages to M2 phenotype (Figure [65]2B). Meanwhile,
   the M2‐polarized‐related parameters, including IL‐10, IL‐1RA, and
   CD206, as detected by real‐time quantitative reverse transcription
   polymerase chain reaction (qRT‐PCR), were up‐regulated in the β‐TCP
   scaffolds, further ascertaining the effectiveness of β‐TCP scaffolds in
   inducing M2 macrophage polarization (Figure [66]2C). In addition to
   M2‐like markers, RAW264.7 cells cultured on β‐TCP scaffolds presented
   up‐regulation of sodium‐phosphate symporter SLC20a1 expression at 5
   days (Figure [67]2D). When incubation time was extended to 10 days, the
   expressions of SLC20a1 in RAW264.7 cells treated with β‐TCP scaffold
   were significantly higher than those in the HA scaffold (Figure [68]S2,
   Supporting Information). Notably, knockdown of SLC20a1 expression with
   siRNA (≈50% reduction) led to the decreased immunofluorescent staining
   for arginase‐1 (Arg‐1), suggesting that the increase in degraded PO[4]
   ^3− content from scaffolds may be related to the upregulation of M2
   polarization (Figure [69]2E; Figure [70]S3, Supporting Information).

Figure 2.

   Figure 2
   [71]Open in a new tab

   CaP ceramics enhance M2 macrophage polarization through SLC20a1 in
   vitro and in vivo. A) Flow cytometry analysis of CD206 (M2 marker) in
   macrophages treated with HA and β‐TCP scaffolds after 2 days and 5 days
   of culturing. B) Polarization trend of M2‐type macrophages on HA and
   β‐TCP scaffolds. C,D) Gene expressions of IL‐10, IL‐1RA, CD206 (C), and
   SLC20a1 (D) after 5 days of culture on HA and β‐TCP scaffolds. E)
   Immuno‐Fluorescence staining for M2 phenotype marker‐Arg‐1 with SLC20a1
   siRNA and corresponding scrambled siRNA after 5 days of culture. F)
   Immuno‐Fluorescence staining for M2 marker‐CD163 (red fluorescence), M1
   marker‐CD86 (green fluorescence) and HE staining at 7 and 14 days in
   vivo. G) Immuno‐Fluorescence staining for M2 marker‐CD163 (red
   fluorescence) and SLC20a1 (green fluorescence) at 7 and 14 days in
   vivo. (n = 5. The values presented are the mean ± SD. Statistical
   significance is assessed by the student's t‐test. ^* p < 0.05, ^**
   p < 0.01, and ^*** p < 0.001 vs HA).

   Our in vivo data also provided an affirmation of the in vitro results.
   M2 macrophage marker, CD163, was also significantly increased in the
   bone tissue of β‐TCP scaffold‐treated rats compared with those in
   HA‐treated rats, especially for 14 days. Many studies have shown that
   promoting macrophage polarization to the reparative M2 phenotype was
   important during the eradication of inflammation and tissue
   regeneration because it can tune an inflammatory microenvironment
   toward a pro‐regenerative niche.^[ [72]^25 , [73]^26 ^]
   Correspondingly, HE staining demonstrated that β‐TCP scaffolds with
   high expression of an M2 macrophage marker showed promising outcomes in
   fracture repair models (Figure [74]2F). More importantly, a noteworthy
   association was identified between the densities of SLC20a1‐positive
   and CD163‐positive cells. The SLC20a1‐positive cells were dispersed
   among the CD163‐positive cells. Additionally, a substantial correlation
   was observed between a higher proportion of SLC20a1‐positive cells and
   a higher proportion of CD163‐positive cells, thereby reinforcing the
   significance of the sodium‐phosphate symporter SLC20a1 expression in
   macrophage polarization (Figure [75]2G).

2.3. Inorganic Phosphate‐Regulated Macrophage Polarization uses SLC20a1

   As the CaP ceramics contributed to the extracellular PO[4] ^3− and
   SLC20a1 transported phosphate, further verification of the role of
   extracellular PO[4] ^3− on macrophage polarization of RAW264.7 was
   conducted by culturing the cells in different concentrations of
   phosphate‐containing medium. Similar to the above‐mentioned results,
   the RAW264.7 cells cultured in high phosphate (5 mm) medium displayed
   significantly higher expressions of various M2 markers, including IL‐10
   and IL‐1RA, relative to those cultured in low phosphate (1 mm) medium
   (Figure [76]3A). In parallel, RAW264.7 cells cultured in high phosphate
   (5 mm) medium showed higher gene expressions of the sodium‐phosphate
   symporter SLC20a1 (Figure [77]3B). This finding was further confirmed
   by immunofluorescent staining, which demonstrated clear accumulation of
   SLC20a1 fluorescent signals within the cells (Figure [78]3C,D). Akin to
   CaP ceramics, the knockdown of SLC20a1 with siRNA annulled the
   phosphate‐mediated macrophage polarization of RAW264.7, as evidenced by
   the downregulation of IL‐10 and IL‐1RA expressions quantified by
   qRT‐PCR and a greater decrease in fluorescent intensity
   (Figure [79]3E–G).

Figure 3.

   Figure 3
   [80]Open in a new tab

   Inorganic phosphate‐regulated macrophage polarization uses SLC20a1.
   A,B) Gene expressions of IL‐10, IL‐1RA (A), and SLC20a1 (B) after 5
   days of culture in growth medium with low (1 mm) and high
   (5 mm)‐phosphate concentrations. C) Immuno‐Fluorescence staining for
   SLC20a1. D) Semi‐quantitative analysis of immunofluorescent signals of
   SLC20a1. E,F) Gene expressions of IL‐10 and IL‐1RA after 5 days of
   culture in low‐ and high‐phosphate medium with SLC20a1 siRNA and
   corresponding scrambled siRNA. G) Immuno‐Fluorescence staining for
   Arg‐1 with SLC20a1 siRNA and corresponding scrambled siRNA after 5 days
   of culture in low‐ and high‐phosphate medium. (n = 5. The values
   presented are the mean ± SD. Statistical significance is assessed by
   student's t‐test or by one‐way ANOVA with Tukey's post hoc test. ^*
   p < 0.05, ^** p < 0.01, and ^*** p < 0.001).

   Overall, the finding that M2 macrophage polarization via CaP ceramics
   and high PO[4] ^3− medium can be reversed by knocking down the SLC20a1
   gene demonstrated that PO[4] ^3− content in the extracellular
   environment and its transportation through SLC20a1 were critical
   mediators of CaP ceramic‐induced M2 polarization in macrophages.

2.4. Increase of Intracellular ATP Through the Actions of the TCA and
Oxidative Phosphorylation, Which Is Dependent on SLC20a1

   One fundamental question remains, however, as to how extracellular
   PO[4]^3−influences the transformation of macrophages into M2‐type
   macrophages. We observed a significant increase in intracellular PO[4]
   ^3− of macrophages cultured on β‐TCP scaffolds with high PO[4] ^3− ion
   release compared with those on HA scaffolds as measured by the
   phosphate assay (Figure [81]4A). Considering that intracellular
   inorganic phosphate was an essential constituent of ATP, ATP production
   was examined. As evidenced by a luminescent assay, there was a notable
   increase in intracellular ATP levels when RAW264.7 cells were cultured
   on β‐TCP scaffolds for 5 days (Figure [82]4B). To further confirm
   whether PO[4] ^3− was directly involved in ATP synthesis, we cultured
   macrophages in medium containing 5 mm PO[4] ^3−. Similar to cells grown
   on CaP ceramics, a significant increase was noted in intracellular ATP
   in the cells cultured in high PO[4] ^3− (5 mm) medium compared with
   that in the cells cultured in low PO[4] ^3− (1 mm) medium; however,
   this increase was attenuated upon knockdown of SLC20a1 with siRNA
   (Figure [83]4C). This finding was corroborated by the quinacrine
   staining for intracellular ATP. The signal intensity was weak in low
   PO[4] ^3− (1 mm) group, indicating low ATP levels. Conversely, the
   signal intensity was greatly enhanced in high PO[4] ^3− (5 mm) group,
   and this enhancement was reduced upon SLC20a1 knockdown
   (Figure [84]4D).

Figure 4.

   Figure 4
   [85]Open in a new tab

   Increase of intracellular ATP through the actions of the TCA and
   oxidative phosphorylation which is dependent upon SLC20a1. A) Content
   of intracellular PO[4] ^3− ion after 5 days of culture on HA and β‐TCP
   scaffolds. B) Concentration of intracellular ATP after 5 days of
   culture on HA and β‐TCP scaffolds. C) Concentration of intracellular
   ATP after 5 days of culture in low‐ and high‐phosphate medium with
   SLC20a1 siRNA and corresponding scrambled siRNA. D) intracellular ATP
   staining with quinacrine after 5 days of culture in low‐ and
   high‐phosphate medium with or without SLC20a1 knockdown. E) A radar
   chart of metabolite abundance via targeted energy metabolite analysis.
   F–K) Content of 3‐phenyllactic‐acid (F), lactate (G),
   alpha‐ketoglutaric‐acid (H), succinic acid (I), fumaric‐acid (J), and
   citric‐acid (K). L) The radar plot to display the effect of β‐TCP
   scaffolds on the metabolite abundance. M) Schematic showing the changes
   in targeted key metabolites involved in TCA cycle. The red and blue
   represent the HA and β‐TCP scaffolds, respectively. Blue from light to
   deep indicates the more obvious difference in metabolite for β‐TCP
   scaffold compared to HA scaffold. (n = 3. The values presented are the
   mean ± SD. Statistical significance is assessed by student's t‐test or
   by one‐way ANOVA with Tukey's post hoc test. ^* p < 0.05 and ^**
   p < 0.01).

   Cellular uptake of phosphate ions represented a key prerequisite as a
   response to the degradation of CaP ceramics, resulting in changes in
   the metabolic state. Next, to monitor the impact of the increased
   intracellular inorganic phosphate on cellular energy metabolism, we
   tested the abundances of the key metabolites in the macrophages with
   the stimulation of β‐TCP scaffolds, and HA scaffolds were selected as
   control. Figure [86]4E shows a radar chart highlighting the outstanding
   effects of pyruvic acid on energy metabolism. Pyruvic acid plays a
   central role in energy metabolism and could be regarded as an important
   node for the switch between aerobic and anaerobic metabolism.^[ [87]^27
   ^] Pyruvic acid would be reduced to lactate in the cytoplasm or be
   oxidized as acetyl coenzyme A in the mitochondria, entering the TCA
   cycle.^[ [88]^28 ^] To identify the pathway by which pyruvic acid was
   used to produce energy, we first measured the levels of lactate and
   3‐phenyllactic‐acid (Figure [89]4F,G; Table [90]S1, Supporting
   Information). No noticeable difference in the abundance of lactic acid
   and 3‐phenyllactic‐acid was found between HA samples and β‐TCP samples.
   Therefore, we hypothesized that pyruvic acid may be implicated in the
   mitochondrial TCA cycle to regulate energy metabolism. Of note, key
   intermediate products in the TCA cycle, including
   alpha‐ketoglutaric‐acid, succinic acid, fumaric acid, and citric acid
   were significantly elevated when the cells were cultured on β‐TCP
   ceramic (Figure [91]4H–K; Table [92]S1, Supporting Information). This
   suggested an enhancement of TCA cycle by β‐TCP scaffolds while
   decreasing the flux toward lactate. It was perhaps relevant that
   intramitochondrial phosphate increased. Varghese et al. have been
   confirmed that cells may use extracellular phosphate to increase
   intramitochondrial phosphate, which could serve as the primary
   substrate for ATP synthesis in the electron transport chain of
   mitochondria.^[ [93]^9 ^] Increasing mitochondrial membrane potential
   can boost TCA cycles. Notably, the TCA cycle is a critical event in
   OXPHOS, and is a more efficient method for ATP generation.^[ [94]^29 ^]
   Metabolic adaptation that occurs during macrophage activation is an
   essential aspect of macrophage polarization and is instrumental in
   their function in inflammation.^[ [95]^30 , [96]^31 ^] Several lines of
   evidence have suggested that M2 macrophages underwent a metabolic shift
   from the less energy‐efficient glycolytic pathway to the TCA cycle to
   meet the increasing demand for attenuating inflammation and promoting
   tissue regeneration.^[ [97]^32 , [98]^33 ^] Here, highly activated
   energy production, via TCA, enabled β‐TCP scaffold to induce macrophage
   polarization into M2 phenotypes. We also found that the increase in
   pyruvic acid was accompanied by the upregulation of asparagine and
   glutamine abundances (Figure [99]4L). M2‐type macrophages are more
   dependent on glutamine influx into the TCA cycle than M1 macrophages.^[
   [100]^34 , [101]^35 ^] Asparagine and glutamine can be used to
   replenish the intermediates in the TCA cycle after deamination.^[
   [102]^36 ^]

   Overall, inorganic phosphate exhibited a dose‐dependent effect on ATP
   accumulation and TCA flux, potentially serving as the underlying
   mechanism through which it influenced macrophage polarization. β‐TCP
   ceramics created a CaP‐rich environment that was conducive to the
   energy requirements necessary for M2 macrophage polarization, achieved
   through a metabolic shift toward the TCA cycle for ATP production. This
   stimulation of the TCA cycle not only facilitated the generation of
   bioenergy essential for macrophage polarization but also yielded
   metabolites that influenced the phenotype and functionality of
   macrophages (Figure [103]4M).

2.5. Intracellular ATP Directly Modulates Macrophage Polarization via
AMPK‐mTOR Signaling Pathway

   Next, the dual‐omics associative analysis, including genomics and
   non‐targeted metabolomics analysis was employed to elucidate the
   molecular mechanism by which increased metabolic levels promoted
   macrophage polarization. A total of 11 268 transcripts and 673
   metabolites were quantified in the transcriptome and metabolome,
   respectively. In comparison to the HA group, the transcriptome of the
   β‐TCP group revealed the presence of 3055 differentially expressed
   genes, with 1914 being upregulated and 1141 being downregulated.
   Additionally, the metabolome analysis identified a total of 310
   differentially expressed metabolites, with 206 being upregulated and
   104 being downregulated (Figure [104]5A). Out of these 116 were
   significantly changed at both the transcriptome and metabolome levels,
   indicating the coordination of transcriptome and metabolome remodeling,
   as shown in the heatmap (Figure [105]5B; Table [106]S2, Supporting
   Information).

Figure 5.

   Figure 5
   [107]Open in a new tab

   Bioinformatic analysis of overall gene and metabolite expression
   changes on HA and β‐TCP scaffolds. A) The up‐and‐down‐regulation of
   gene and metabolite. B) Heat map illustrating the common changed
   substances in genes and metabolites from HA and β‐TCP scaffolds. C) The
   Sankey diagram links metabolic pathways and transcriptional pathways
   through metabolic substances. D) Pyruvic acid is associated with a
   variety of energy metabolites. E) Subnetworks of the notable enrichment
   maps created from the selected pathways. F) The Pearson's correlation
   analysis. The correlation coefficient ranged from −1 to 1 (blue to red)
   depending on the color intensity. G) Immuno‐Fluorescence staining for
   mTOR and AMPK after 5 days of culture on HA and β‐TCP scaffolds. H)
   Gene expressions of mTOR after 5 days of culture on HA and β‐TCP
   scaffolds. I) Gene expressions of mTOR after 5 days of culture on β‐TCP
   scaffolds with or without SLC20a1 knockdown. J) Gene expressions of
   IL‐10 after 5 days of culture on β‐TCP scaffolds with or without
   Torin1. (n = 5. The values presented are the mean ± SD. Statistical
   significance is assessed by student's t‐test. ^* p < 0.05, ^**
   p < 0.01, and ^*** p < 0.001).

   In addition, the conjoint Kyoto Encyclopedia of Gene and Genomes (KEGG)
   pathway enrichment analysis was conducted to examine the transcriptome
   and metabolome data. This analysis revealed the emergence of common
   genes and corresponding metabolites in both genomic and metabolic
   pathways (Figure [108]5C). Notably, macrophage activation necessitated
   metabolic support, thereby requiring polarizing signals to energize and
   regulate the metabolic pathways. As expected, arginine biosynthesis was
   also observed. M2 macrophage polarization was characterized, in part,
   by the ability to metabolize L‐arginine.^[ [109]^37 ^] Arginine
   biosynthesis was associated with PI3K‐Akt signaling pathway, AMPK
   signaling pathway, cyclic adenosine monophosphate (cAMP) signaling
   pathway, and mTOR signaling pathway through alpha‐ketoglutaric‐acid,
   indicating that macrophage polarization was related to the above
   signaling pathways in β‐TCP‐treated cells. Moreover, TCA cycle and
   oxidative phosphorylation that were associated with M2‐polarized
   metabolism state were linked with AMPK and mTOR signaling pathways,
   indicating that AMPK‐mTOR axis may be involved in β‐TCP‐mediated
   bioenergy to promote macrophage polarization. In addition, we found
   indications of the existence of additional regulatory mechanisms.
   Pyruvate, an important precursor metabolite involved in energy
   production, was implicated in cAMP signaling pathway, in addition to
   AMPK‐mTOR axis (Figure [110]5D).

   Furthermore, we sought to establish the interaction networks involved
   in predicting alterations in cellular signaling cascades using the
   GeneMANIA. Notably, an interactive network was established between
   arginine and proline metabolism and oxidative phosphorylation. Numerous
   genes were associated with oxidative phosphorylation through relatively
   few downstream pathways, often converging on the AMPK‐mTOR axis and
   cAMP signaling (Figure [111]5E).

   AMPK is an important energy receptor in cells and decreased energy
   availability results in AMPK activation and restoration of energy
   homeostasis. When energy is sufficient, mTOR signaling is activated by
   a decrease in AMPK activity.^[ [112]^38 , [113]^39 ^] Further PPI
   analysis using GeneMANIA analysis showed the category of
   “energy‐dependent regulation mTOR by LKB1‐AMPK”. Specially, the β‐TCP
   scaffolds increased the synthesis of PIPs at the plasma membrane and
   PLD activity, leading to greater AKT activation. Activated AKT then
   phosphorylated mTOR and activated its downstream signaling pathway to
   promote the polarization of macrophages toward the M2 phenotypes. This
   finding aligned with previous reports highlighting the importance of
   AMPK‐mTOR pathway in modulating macrophage polarization toward the
   M2‐phenotype.^[ [114]^40 , [115]^41 ^]

   Subsequently, the Pearson's correlation analysis provided additional
   evidence supporting the significance of the AMPK‐mTOR signaling pathway
   in β‐TCP‐mediated macrophage polarization. Of particular interest were
   three regions in the confusion matrix (heatmap) that displayed strong
   correlation coefficients (red or blue). 1) The energy metabolism was
   significantly correlated with the AMPK signaling pathway. 2) mTOR was
   modulated by both AKT and AMPK, which were to regulate mTOR positively
   and negatively, respectively (Figure [116]5F; Table [117]S3–[118]S8,
   Supporting Information).

   Finally, the role of AMPK‐mTOR pathway in β‐TCP‐mediated‐energetic
   level promoting M2 polarization was further confirmed.
   Immunofluorescence staining and qRT‐PCR results demonstrated that cells
   cultured on β‐TCP scaffolds exhibited a significant increase in mTOR
   expressions (Figure [119]5G,H) and a simultaneous decrease in AMPK
   intensity, compared with HA scaffolds (Figure [120]5G). Conversely,
   mTOR intensity was reduced, following blocking with siSLC20a1
   (Figure [121]5I; Figure [122]S4, Supporting Information). Furthermore,
   the pharmaceutical inhibition of mTOR signaling by Torin1 attenuated
   the polarization of macrophages into the M2 phenotype, as evidenced by
   the downregulated gene expressions of IL‐10; this confirmed that
   inorganic phosphate increased energy metabolism in macrophages via
   AMPK‐mTOR signaling pathway, promoting M2 macrophage polarization
   (Figure [123]5J).

   Overall, elevations in the intracellular PO[4] ^3− concentration in the
   β‐TCP scaffolds induced the increased bioenergy level, resulting in
   AMPK turning off. This was accompanied by phosphorylation modification
   via AKT, resulting in the release of its inhibitory effect on AMPK and
   the upregulation of mTOR activity.^[ [124]^42 ^] The influence of
   AMPK‐mTOR‐mediated sensing of environmental and metabolic cues on
   macrophage polarization was expected, given that macrophage
   polarization was regulated not only by cytokines and growth factors but
   also by environmental cues, including nutrients, fatty acids, and other
   metabolites.^[ [125]^43 ^] The AMPK lies at the crossroads of
   metabolically driven macrophage inflammation. The mTOR activation is
   maintained, within a range, by AMPK, which is known to activate mTOR
   during conditions of increased energy supply.^[ [126]^44 ^] Most
   studies have reported a strong association between mTOC1 and the
   regulation of fatty acid oxidation, further supporting the notion that
   the cellular metabolism of macrophages was intricately connected to
   their activation status.^[ [127]^45 , [128]^46 , [129]^47 ^]

2.6. Extracellular ATP Modulates Macrophage Polarization via A2b Adenosine
Receptor

   An association study between metabolomics and genomics suggested that
   cAMP signaling pathway was also enriched. Pearson's correlation
   analysis revealed a positive association between M2 macrophages, cAMP
   levels, and energy metabolism (Figure [130]5F). This was not surprising
   since ATP acted as a signaling molecule in addition to being an energy
   source.^[ [131]^9 , [132]^24 ^] Excessive ATP was degraded into AMP and
   subsequently into adenosine, which may be released from the cells and
   appear in the extracellular space. Macrophage activation has been
   reported to be controlled by extracellular ATP and associated adenosine
   signaling.^[ [133]^9 ^]

   To investigate the impact of extracellular ATP on macrophage
   polarization, we used the vesicular transport inhibitor N‐ethyl
   maleimide (NEM) to inhibit ATP transport to the extracellular
   environment. The results showed that the use of NEM led to the
   down‐regulation of IL‐10 expressions in β‐TCP scaffolds, suggesting
   that inhibition of ATP transport negatively influenced β‐TCP
   scaffold‐assisted M2 macrophage polarization (Figure [134]6A). However,
   no significant inhibitory effect was observed on HA scaffolds. This
   might be because the inhibition of ATP transport may compensate for the
   intracellular energy deficiency, resulting in an increase in IL‐10
   expressions. HPLC measurements revealed a substantial quantity of
   adenosine in cell culture involving β‐TCP scaffolds. The presence of
   adenosine, a metabolite of ATP, in the extracellular milieu indicated
   that ATP was released from cells and rapidly metabolized to adenosine
   through the action of membrane‐bound ectonucleotidases (CD39),
   including ectonucleoside triphosphate diphosphohydrolase,
   ectonucleotide pyrophosphatase/phosphodiesterase, and
   ecto‐5′nucleotidases (CD73).^[ [135]^48 ^] Moreover, the presence of
   extracellular adenosine was abolished through the SLC20a1 knockdown and
   the use of NEM, respectively (Figure [136]6B), demonstrating the
   correlation between PO[4] ^3−‐modulated ATP and adenosine.

Figure 6.

   Figure 6
   [137]Open in a new tab

   Extracellular ATP modulates macrophage polarization via A2b adenosine
   receptor. A) Gene expressions of IL‐10 after 5 days of culture on HA
   and β‐TCP scaffolds with or without vesicle transport inhibitor NEM. B)
   HPLC measurement of adenosine in culture medium after 5 days with or
   without SLC20a1 and NEM. C,D) Gene expressions of IL‐10 (C) and IL‐1RA
   (D) after 5 days of culture on HA and β‐TCP scaffolds with or without
   alloxazine. E) Immuno‐Fluorescence staining for Arg‐1. (n = 5. The
   values presented are the mean ± SD. Statistical significance is
   assessed by one‐way ANOVA with Tukey's post hoc test. ^* p < 0.05, ^**
   p < 0.01, and ^*** p < 0.001).

   To further confirm the involvement of adenosine in ATP‐mediated
   macrophage polarization, we investigated the effect of adenosine
   receptors on the modulation of macrophage polarization toward the M2
   subtype. Specific pharmacological inhibition of adenosine receptors by
   alloxazine inhibitors demonstrated that the alloxazine down‐regulated
   the expressions of IL‐10 and IL‐1RA genes in the cells treated with
   β‐TCP scaffolds (Figure [138]6C,D). Immunofluorescent staining for
   Arg‐1 also revealed a decrease in Arg‐1 expression in the presence of
   alloxazine (Figure [139]6E). These findings aligned with the previous
   reports indicating the beneficial effects of adenosine signaling via
   the A2b adenosine receptor on macrophage polarization.^[ [140]^49 ,
   [141]^50 ^] Studies have reported that activation of the A2b adenosine
   receptor generated cAMP, which can stimulate protein kinase A (PKA).^[
   [142]^51 ^] This would be in line with our results from GeneMANIA
   analysis that β‐TCP scaffold increased intracellular cAMP, resulting in
   the polarization of macrophages toward M2, probably via PKA‐cAMP
   responsive element‐binding protein (CREB) signaling (Figure [143]5E).

   The findings collectively suggested a molecular mechanism, as
   illustrated in Figure [144]7 , wherein the fluctuating dissolution and
   precipitation of CaP minerals from HA and β‐TCP ceramics regulated the
   levels of calcium ions (Ca^2+) and phosphate ions (PO[4] ^3−) in the
   extracellular environment. The extracellular PO[4] ^3− was transported
   into macrophages via the SLC20a1 transporter and subsequently entered
   the intracellular compartment. High levels of intracellular PO[4] ^3−
   drove the TCA cycle, followed by oxidative phosphorylation, resulting
   in rapid and efficient energy production. Importantly, our study
   indicated that the promotion of oxidative phosphorylation was a
   promising strategy for modulating macrophage polarization. In contrast
   to M1 macrophages, M2 macrophages preferred oxidative phosphorylation
   as a means of storing energy and supporting metabolism. Alterations in
   metabolic pathways played a crucial role in determining the distinct
   macrophage polarization phenotypes. Additionally, macrophage
   polarization led to metabolic reprogramming, with macrophages
   selectively utilizing metabolic pathways to enhance the activation of
   phenotypes and effector functions.

Figure 7.

   Figure 7
   [145]Open in a new tab

   The potential molecular mechanism behind the enhanced M2 macrophage
   polarization of phosphate ions released from CaP ceramics.

   The positive effects of TCA cycle and OXPHOS on macrophage polarization
   were not only limited to supporting bioenergetics and biosynthesis but
   also altered transcription by modulation of epigenetic modifications
   leading to changes in transcriptional programs. The potential molecular
   mechanism behind the enhanced M2 macrophage polarization of phosphate
   ions released from CaP ceramic was attributed to the following two
   points. On the one hand, the active function of PO[4] ^3− in macrophage
   polarization involved AMPK‐mTOR axis, likely through the increased
   energy metabolism. AMPK‐mTOR axis acted as central metabolic switch in
   cell metabolism and served as a signaling molecule in macrophages. When
   cell energy was increased, there was a decrease in the activation of
   AMPK, resulting in turning on the mTOR signaling cascade. On the other
   hand, intracellular ATP could be secreted into the exocellular space
   and metabolized into adenosine, which subsequently promoted the
   polarization of macrophages toward an M2 phenotype through the A2b
   adenosine receptor via autocrine and/or paracrine signaling.

   Notably, although this study primarily focused on phosphate metabolism
   in a mineralized environment, it did not discount the positive effects
   of Ca moieties found in CaP ceramics. Previous research has already
   elucidated the molecular mechanism through which calcium influenced M2
   macrophage polarization via PI3K‐Akt and Wnt/β‐catenin.^[ [146]^52 ^]
   Moreover, the CaP moieties of the CaP ceramics usually underwent
   dissolution/precipitation responding to the concentration of Ca^2+ or
   PO[4] ^3‐ ions in the surrounding environment, resulting in the
   creation of a dynamic environment. This dynamic mineral environment
   held significant importance, as it demonstrated the interdependency
   between extracellular Ca^2+, PO[4] ^3−, and Ca.

   In summary, we have gained insights into the impact of phosphate ions
   on the progression of polarization, specifically highlighting the
   crucial expression pattern of the SLC20a1 phosphate transporter during
   macrophage polarization and elucidating a mechanism that emphasized the
   metabolic regulation of phosphate ions to enhance cellular energy
   status for M2 phenotype polarization. These findings not only
   established phosphate ion as a favorable metabolic factor for
   macrophage polarization toward the M2 phenotype but also underscored
   the importance of considering the regulation of PO[4] ^3− content in
   the design of CaP‐biomaterials.

3. Conclusion

   Here, we report the molecular mechanism by which phosphate ions promote
   macrophage polarization toward the M2 phenotype via metabolic
   regulation. In vitro and in vivo studies demonstrate that extracellular
   phosphate ions enter the cells through SLC20a1 and generate large
   amounts of ATP through the TCA cycle and OXPHOS. Furthermore, the
   integration of metabolome and transcriptome analysis has established a
   link between phosphate‐mediated macrophage polarization, metabolism,
   and intracellular signal transduction. Activation of the intracellular
   AMPK‐mTOR axis has been identified as a contributing factor to
   M2‐macrophage polarization, potentially due to the enhanced energy
   supply from increased phosphate ion levels. In addition, extracellular
   phosphate regulates the synthesis of ATP synthesis, which is
   subsequently secreted and metabolized to adenosine, thereby promoting
   M2 macrophage polarization via A2b adenosine receptors. This research
   significantly contributes to our understanding of how phosphate governs
   macrophage phenotype through metabolic processes and provides ideas for
   the design of CaP ceramic materials in subsequent bone repair.

4. Experimental Section

Characterization of HA and β‐TCP Scaffolds

   Engineering Research Center in Biomaterials, Sichuan University
   provided porous HA and β‐TCP scaffolds in two different sizes
   (Ф14 × 2 mm for cell culture and Ф2 × 6 mm for animal experiment). The
   microtopography and elemental composition of the CaP ceramics were
   observed using field emission scanning electron microscopy (FE‐SEM,
   S‐4800, Hitachi, Japan) and EDS (Phenom, Philips, Netherlands). XRD
   (X'Pert 1, Philips, Netherlands) was performed to analyze the phase
   composition of CaP ceramics. To evaluate the degradation of HA and
   β‐TCP ceramics, samples were immersed in Tris‐HCl (0.1 mol L^−1, PH
   7.4) at 37 °C. After 1, 3, 5, and 7 days of incubation, the immersion
   solution (0.5 mL) was collected and added to a new 0.5 mL Tris‐HCl
   solution. The concentrations of PO[4] ^3− ions were then measured using
   an inductively coupled plasma mass optical emission spectrometer
   (ICP‐OES, SPECTRO ARCOS, Spectro Analytical, Germany). To evaluate the
   microtopography after the degradation of the CaP ceramics, they were
   rinsed with deionized water, dried, and examined by SEM after 7 days of
   incubation.

Cell Culture and Media

   The RAW 264.7 macrophages from mice were cultured in Dulbecco's
   modified Eagle's medium (DMEM, Gibco, USA) supplemented with 10% fetal
   bovine serum (FBS, BI, ISR) and 1% penicillin/streptomycin (HyClone,
   USA) and cultured with a fully humidified atmosphere of 5% CO[2] at
   37 °C. The phosphorus‐rich medium was prepared as follows. To obtain
   final concentrations of 1 and 5 mm PO[4] ^3− ions, various
   concentrations of PO[4] ^3−, NaH[2]PO[4] and Na[2]HPO[4] were mixed at
   a ratio of 1:4 to supplement the growth medium.

Flow Cytometry

   RAW264.7 (1 × 10^4) cells cultured with HA and β‐TCP samples for 2 and
   5 days were collected and then respectively incubated with peridinin
   chlorophyll protein (Percp, BioLegend, USA)‐conjugated anti‐mouse CD11c
   and activated protein C (APC, eBioscience, USA)‐conjugated anti‐mouse
   CD206 at room temperature for 30 min without light after fixation,
   membrane breaking and blocked with CD16/32 (BD pharmingen) steps.
   Finally, the cells were resuspended in 200 µL PBS and subjected to
   FACSCalibur flow cytometry (Beckman, CytoFLEXS, USA).

Quantitative Real‐Time PCR Analysis

   Total RNA was extracted and cDNA was generated using Triozl reagent and
   the Evo MMLVRT Mix kit (Accurate Biotechnology (Hunan) Co., Ltd),
   respectively, following the manufacturer's protocol. The qRT‐PCR was
   conducted at 95 °C (30 s), followed by 40 cycles at 95 °C (5 s)/60 °C
   (30 s). Signals were detected using SYBR green intercalating dye
   (Accurate Biotechnology (Hunan) Co., Ltd). Sequences of the primers
   used in this study were provided in (Table [147]S9, Supporting
   Information).

Immunofluorescence Staining

   For immunofluorescence staining, RAW264.7 macrophages plated on the
   materials for 5 days were washed, fixed, and permeabilized with 4%
   paraformaldehyde (0.1% Triton X‐100). After washing with PBS, blocking
   was performed by 30‐min incubation with 1% bovine serum albumin (BSA).
   Next, the cells were incubated overnight at 4 °C with primary antibody
   against Arg‐1 (1:200; Abclonal, China). On day 2, the cells were
   incubated at room temperature with a secondary antibody conjugated to
   Alexa Fluor 488 for 1 h. Finally, cell nuclei were stained with DAPI
   for 5 min.

Measurement of SLC20a1 Expression

   For the examination of SLC20a1 expression, total RNAs from the cells
   treated with HA, β‐TCP, phosphorus‐rich medium (1 mm), and
   phosphorus‐rich medium (5 mm) were extracted with Trizol reagent and
   the relative expressions of SLC20a1 gene were determined using qRT‐PCR
   following the method mentioned above. Moreover, immunofluorescence
   staining was used to determine the expression levels of SLC20a1
   protein.

siRNA Knockdown

   As directed by the manufacturer, RAW264.7 cells were transfected with
   siRNA oligonucleotides to knockdown SLC20a1. Briefly, RAW264.7 cells
   were transfected under serum‐free conditions with 30 nm siRNA targeting
   SLC20a1 and scrambled control siRNA (medium GC content) using the
   RNAimax transfection reagent (Invitrogen). The siRNA sequences of siRNA
   were given in (Table [148]S10, Supporting Information).

Effect of Knockdown of SLC20a1 on Macrophage Polarization

   After knocking down SLC20a1, the expressions of IL‐10 and IL‐1RA were
   assessed by quantitative qRT‐PCR. Besides, Arg‐1 was stained by
   incubation with an Arg‐1 antibody (1:200; Abclonal, China) to observe
   its expression of Arg‐1 in vitro.

Inorganic Phosphate Assay

   The effects of HA and β‐TCP ceramics on intracellular phosphate
   contents were measured by malachite green phosphate detection kit. In
   brief, RAW264.7 was seeded on HA and β‐TCP ceramics in 24‐well culture
   plates. On day 5, cells were rinsed three times with physiological
   saline solution before adding deionized water (200 µL per well). The
   cells were frozen and thawed three times to lyse them. Subsequently,
   the assay was performed in accordance with the manufacturer's
   instructions (Beyotime).

Luminescent ATP Analysis and Fluorescent ATP Staining

   The intracellular ATP concentration was quantified using an ATP assay
   kit (Beyotime). In brief, after counting the cells, they were
   completely lysed by adding lysate. Then, the supernatants were
   harvested by centrifugation at 4 °C, 12 000 g for 5 min. Afterward, the
   ATP content of individual cell was determined according to the
   manufacturer's instructions.

   For subsequent fluorescent staining of intracellular ATP, the cells
   were gently rinsed with PBS three times and then were incubated in
   serum‐free DMEM dissolved in 30 µm quinacrine dihydrochloride
   (Sigma–Aldrich) for 30 min before imaging with an inverted fluorescence
   microscope (Leica, DMi8, Germany).

Detection of Targeted Energy Metabolites

   After 5 days of HA and β‐TCP treatment respectively, RAW264.7 cells
   were collected and centrifuged at 1200 r min^−1 for 5 min at 4 °C,
   before washing them in cold PBS. Further centrifugation at
   1200 r min^−1 for 5 min at 4 °C was followed by vigorous vortexing with
   1 mL of 80% cold methanol. The supernatant was collected and normalized
   to the protein concentration. Subsequently, the sample extracts were
   analyzed using an LC‐ESI‐MS/MS system (Waters ACQUITY H‐Class; MS,
   QTRAP 6500+ System). The AB 6500+ QTRAP LC‐MS/MS system was equipped
   with an ESI Turbo Ion‐Spray interface and can operate in both positive
   and negative ion modes according to Analyst 1.6 software (A B Sciex).
   Detection and determination of the relative abundances of the targeted
   energy metabolites were performed using MetWare based on the AB Sciex
   QTRAP 655 LC‐MS/MS platform.

Observed Metabolite Alterations Using Untargeted Metabolomics

   The preparation and collection of samples, and untargeted metabolomics
   analysis were performed as described previously. In short,
   intracellular metabolites were analyzed using an LC‐ESI‐MS/MS system
   (UPLC, ExionLC AD; MS, QTRAP System). LIT and triple quadrupole (QQQ)
   scans were acquired using a triple quadrupole‐linear ion trap mass
   spectrometer (QTRAP), QTRAP LC‐MS/MS System which was equipped with an
   ESI Turbo Ion‐Spary interface and could operate in positive and
   negative ion mode by Analyst 1.6.3 software (Sciex).

RNA‐Sequencing and Data Analysis

   RNA‐sequence (RNA‐seq) analysis was performed as described previously.
   Total RNA isolated from macrophages was selected for RNA sequencing.
   The KEGG pathway and GO analysis were used for the enrichment analysis
   of differentially expressed genes. Genes with differential expression
   were identified by calculating the fold‐change over two and an adjusted
   P ‐value of 0.05.

Verification of the AMPK‐mTOR Axis

   After 5 days of culture on HA and β‐TCP scaffolds, macrophages were
   stained to visualize AMPK (1:200; Abclonal) and mTOR (1:200; Abclonal)
   using immunofluorescence staining as previously described.
   Additionally, qRT‐PCR was used to detect the mRNA expression levels of
   mTOR as previously described. To further confirm that phosphate ions
   were involved in AMPK‐mTOR axis, SLC20a1 expression was knocked down.
   After 5 days, immunofluorescence staining was used to explore the
   expressions of AMPK and mTOR. In addition, mTOR gene expression levels
   were determined using qRT‐PCR. Finally, Torin 1 (a specific mTOR
   inhibitor, sigma) was used at a concentration of 250 nm to block the
   mTOR activity. Afterward, IL‐10 expressions were measured via qRT‐PCR.

Adenosine Content

   Prior to the experiments, the cell culture media (1 mL) was lyophilized
   for 24 h and frozen at −80 °C. Using HPLC‐UV, the concentration of
   adenosine was quantified. Specifically, chromatic separations were
   conducted on an Aeris Peptide 3.6u Xb‐C18 phenomenex column
   (150 mm × 2.10 mm) with the assistance of a Hitachi Elite LaChrom
   L‐2130 pump equipped with a UV–vis detector (Hitachi Elite
   LaChromL‐2420). Adenosine levels were detected at 254 nm under
   isocratic conditions, with a flow rate of 0.23 mL·min^−1 and an
   injection volume of 20 µL. The mobile phase for adenosine consisted of
   0.5% acetonitrile, 5% (vol/vol) methanol, and 94.5% (vol/vol) sodium
   acetate buffer (0.25 m, pH 6.3).

Blockade of ATP Transport

   NEM was added to the culture medium (final concentration, 0.5 µm) to
   inhibit ATP transport, and the fresh medium was changed after 2 days.
   Subsequently, the expression of IL‐10 was determined using qRT‐PCR as
   described above.

Alloxazine Inhibit the A2b Adenosine Receptor Expression

   The A2b receptor inhibitor alloxazine (Sigma–Aldrich) was added at a
   concentration of 100 nm to the medium for pharmacological inhibition of
   adenosine receptors. After 2 days, culture medium was replaced. Next,
   the expressions of IL‐10 and IL‐1RA genes were determined using qRT‐PCR
   following the method mentioned above. Moreover, immunofluorescence
   staining was used to determine the expression level of Arg‐1 protein.

In Vivo Evaluation

   The animal in vivo experiments were performed in accordance with the
   Institutional Animal Care and Use Committee of Sichuan University
   (Chengdu, China) guidelines (approval No. K2023011). Male
   Sprague‐Dawley rats (weighing 350–400 g), purchased from DaShuo
   Experimental Animal Co. Ltd. (Chengdu, China), were used in this study.
   The rats were anesthetized by intraperitoneal injection of chloral
   hydrate at a dosage of 0.4 mL/100 g, and dental trephine bars were used
   to create 2 mm in diameter and 4.5 mm in‐depth critical‐sized femur
   defects. After washing the holes with sterile saline, the implants were
   inserted into the defects. After all the surgical procedures, the rats
   were housed in standard cages and fed a regular laboratory diet.

   The rats were sacrificed after 7 and 14 days of implantation, and the
   tibias with implants were harvested for subsequent analysis. The
   hematoxylin & eosin (H&E) staining and immunofluorescence staining for
   CD163 (NOVUS, 1:100), CD 86 (HUAAN, 1:200), and SLC20a1 (BIOSS, 1:100)
   were used to detect macrophage polarization and new bone formation.

Statistical Analysis

   All data were present as mean ± standard deviation (SD) from at least
   three repeated experiments. Student's t‐test was used to determine the
   significance between the two groups. One‐way analysis of variance
   (ANOVA) followed by Tukey's post hoc test was utilized to perform
   statistical comparisons between multiple groups. The detailed sample
   sizes were labeled in figure caption, and the sample size for each
   statistical analysis (n) was at least three. GraphPad Prism 8.00
   (GraphPad Software Inc., USA) was utilized for data analysis. The
   significance levels were denoted by asterisks: ^* p < 0.05, ^**
   p < 0.01, and ^*** p < 0.001, respectively. A value of p < 0.05 was
   considered indicative of a statistically significant difference.

Conflict of Interest

   The authors declare no conflict of interest.

Supporting information

   Supporting Information
   [149]ADVS-11-2306062-s001.pdf^ (320.8KB, pdf)

Acknowledgements