Pstpip2-deficient mouse models in SAPHO syndrome research: Utility, mechanistic insights, and translational limitations Authors: Yu Wang, Bowen Feng, Yunuo Wang, Yi Wang, Yingai Xu, Mengjiao Gu, Yuanhao Wu Published: 2026 ABSTRACT Synovitis, acne, pustulosis, hyperostosis, and osteitis (SAPHO) syndrome is a rare autoinflammatory disorder characterized by combined cutaneous and osteoarticular involvement. Its etiology and pathogenesis remain incompletely understood, and an ideal animal model capable of fully recapitulating its phenotype is still lacking. Pstpip2-deficient mice exhibit sterile osteitis, inflammatory bone destruction, abnormal bone remodeling, and varying degrees of extraosseous inflammation, accompanied by myeloid immune abnormalities and IL-1β-associated inflammatory dysregulation. These mice have therefore become an important experimental model for investigating SAPHO-related myeloid inflammation and inflammatory bone remodeling. This review focuses on the biological functions of Pstpip2 and the different models of Pstpip2 deficiency, systematically summarizing the major model types and disease-course characteristics, osteoarticular and cutaneous phenotypes, systemic myeloid inflammation, and mechanisms involving aberrant neutrophil activation, IL-1β production and processing, osteoclast differentiation, abnormal bone formation, and the functional division between hematopoietic and non-hematopoietic compartments. We also discuss the effects of breeding efficiency, genetic background, diet, and gut microbiota on phenotypic stability, as well as the translational value and limitations of current therapeutic targets and pharmacological intervention evidence. However, Pstpip2 has not been established as a definitive causative gene for human SAPHO syndrome, and these models remain limited in terms of lesion distribution, the extent to which hyperostosis is reproduced, and phenotypic stability. Therefore, Pstpip2-deficient mice are better regarded as mechanistic models of SAPHO-related myeloid inflammation and inflammatory bone remodeling rather than as fully homologous surrogates for the human disease. Introduction Synovitis, acne, pustulosis, hyperostosis, and osteitis (SAPHO) syndrome was first described by the rheumatologist Chamot in 1987 and is a rare autoinflammatory disorder characterized by cutaneous manifestations and osteoarticular involvement [1,2]. SAPHO syndrome represents an important clinical entity within the spectrum of autoinflammatory bone diseases. Its skeletal lesions are fundamentally characterized by sterile inflammation and may be accompanied by extraosseous manifestations such as synovitis, acne, and palmoplantar pustulosis [3,4]. The precise etiology and pathogenesis of SAPHO syndrome remain unclear, although current evidence suggests a multifactorial origin closely associated with dysregulation of innate immunity [5]. Despite substantial advances in understanding its clinical phenotypes, inflammatory pathways, and therapeutic strategies in recent years, several key issues—including the lack of a clearly defined causative gene, marked clinical heterogeneity, and the absence of stable animal models—continue to hinder mechanistic investigations, therapeutic target validation, and clinical translation. With increasing research interest, Pstpip2-deficient mice have gained attention as experimental tools for studying SAPHO syndrome. These models include the Pstpip2cmo, Pstpip2Lupo, and Pstpip2−/− models, which reproduce different aspects of SAPHO-related pathology, including sterile osteitis, inflammatory bone destruction, abnormal bone remodeling, and systemic myeloid inflammation. In this narrative review, we summarize, compare, and critically integrate the available evidence regarding the utility, mechanistic basis, and translational limitations of Pstpip2-deficient mice in SAPHO syndrome research [6]. Particular emphasis is placed on comparing the phenotypic and research characteristics of the Pstpip2cmo, Pstpip2Lupo, and Pstpip2−/− models and on evaluating their research value, scope of applicability, and translational limitations in terms of inflammatory responses, bone remodeling, therapeutic interventions, breeding performance, and phenotypic stability. Methods PubMed was used as the primary source for literature retrieval. Relevant English-language publications were identified using combinations of the following keywords: “SAPHO syndrome,” “PSTPIP2/Pstpip2,” “Pstpip2 cmo,” “Pstpip2 Lupo,” “autoinflammatory osteomyelitis,” and “bone remodeling.” Additional studies were identified by screening the reference lists of included articles and relevant reviews. Priority was given to original studies of Pstpip2-deficient mouse models, while clinical studies of SAPHO syndrome and literature on related autoinflammatory bone diseases were also included. The literature covered model phenotypes, pathogenic mechanisms, therapeutic interventions, breeding performance, and phenotypic stability. During evidence synthesis, direct experimental evidence from Pstpip2-deficient models was distinguished from indirect evidence derived from other diseases or experimental systems. The consistency and limitations of the available findings were evaluated in the context of model type, genetic background, and experimental conditions. Biological functions of Pstpip2 Molecular structure and myeloid-specific expression of Pstpip2 Pstpip2 encodes proline-serine-threonine phosphatase-interacting protein 2 (PSTPIP2) [7]. It is predominantly expressed in myeloid lineages of the hematopoietic system, including monocytes/macrophages, neutrophils, and osteoclast precursors [8,9]. As an important negative regulatory molecule, PSTPIP2 interacts with inhibitory molecules to restrain aberrant macrophage activation as well as neutrophil migration and hyperresponsiveness, thereby maintaining innate immune homeostasis [[10], [11], [12], [13]]. Loss of PSTPIP2 function can disrupt myeloid cell responses and is closely associated with the development of autoinflammatory diseases. Establishment, major types, and disease-course characteristics of Pstpip2-deficient models The major Pstpip2-deficient models include the spontaneously arising Pstpip2cmo model, the chemically mutagenized Pstpip2Lupo model, and the Pstpip2−/− model. All three models share impaired PSTPIP2 expression or function but differ in the nature of the genetic alteration, residual protein status, and principal research applications. The Pstpip2cmo model results from a spontaneous missense mutation in Pstpip2 [14]. Subsequent studies showed that PSTPIP2 protein is nearly undetectable in macrophages, mast cells, and osteoclasts in this model, indicating a state of near-complete functional deficiency [15]. Among the available models, the Pstpip2cmo model has been the most extensively studied and has been widely used to investigate myeloid immune abnormalities, IL-1β-associated inflammation, and inflammatory bone remodeling caused by PSTPIP2 dysfunction. The Pstpip2Lupo model was generated by N-ethyl-N-nitrosourea (ENU) mutagenesis. Its Pstpip2 missense mutation markedly reduces PSTPIP2 protein expression and impairs the function of the residual protein [8]. Disease penetrance in this model is strongly dependent on genetic background; accordingly, it has primarily been used to investigate macrophage-intrinsic abnormalities and the effects of host genetic background on PSTPIP2-associated autoinflammatory phenotypes. The Pstpip2−/− model was generated by targeted gene disruption, resulting in complete loss of PSTPIP2 protein. This model concurrently exhibits multiple SAPHO-related inflammatory phenotypes involving bone, joints, and skin, thereby providing an important experimental tool for evaluating the relationship between complete PSTPIP2 deficiency and SAPHO-like multisystem manifestations [9]. All three models exhibit spontaneous and progressively worsening disease courses, although the timing of disease onset and the initial manifestations are not identical. The Pstpip2cmo model typically develops early disease manifestations at 4–6 weeks of age [14]. Disease onset in the Pstpip2Lupo model is strongly influenced by genetic background, with a median onset of approximately 42 days on the C3H background and a delayed onset of approximately 124 days on a mixed C3H/C57BL/6 background [8]. The Pstpip2−/− model develops clinical manifestations at approximately 4 weeks of age, which become progressively more severe with increasing age [9]. Thus, although different Pstpip2-deficient models share similar features of progressive autoinflammation, they do not follow an entirely uniform disease-course timeline. Overall, although all three models arise from loss or impairment of PSTPIP2 expression or function, they differ in their principal research applications and can therefore serve as complementary experimental tools for investigating PSTPIP2-associated myeloid inflammation and bone damage from different perspectives. Given their shared pathological basis of impaired PSTPIP2 function, these models are collectively referred to hereafter as “Pstpip2-deficient mice (models).” Model-specific phenotypes or experimental findings are identified by the corresponding model where appropriate. Major phenotypic characteristics and model positioning of Pstpip2-deficient mice Bone, joint, soft-tissue, and cutaneous phenotypes of Pstpip2-deficient models Sterile osteitis and abnormal bone remodeling are among the core phenotypes consistently observed in Pstpip2-deficient models, although the specific features of bone lesions vary among models. In the Pstpip2cmo model, bone lesions predominantly involve the caudal vertebrae and paws and are characterized by osteitis, bone erosion, and local infiltration of mixed inflammatory cells. As the lesions progress, the inflamed areas may gradually be replaced by newly formed bone and fibrous tissue [14,15]. In the Pstpip2Lupo model, inflammation of the paws can progress to local osteolysis and tissue necrosis [8]. Further studies have shown that both Pstpip2cmo and Pstpip2Lupo mice develop systemic trabecular bone loss, accompanied by increased numbers of tartrate-resistant acid phosphatase (TRAP)-positive osteoclasts and enhanced differentiation capacity of osteoclast precursors [16]. These findings indicate that their skeletal phenotypes are not limited to local inflammatory bone destruction but also involve more widespread abnormalities in bone remodeling. In addition to osteitis and osteolysis, the Pstpip2−/− model can also develop synovitis and local hyperostosis [9]. Collectively, these findings indicate that Pstpip2 deficiency can lead to inflammatory bone destruction and abnormal bone remodeling, although the specific manifestations and supporting evidence differ among models. Notably, clear evidence of local hyperostosis has thus far been reported primarily in the Pstpip2−/− model [9]. Although late-stage lesions in the Pstpip2cmo model may show new bone formation and replacement by fibrous tissue, these histological changes cannot be directly equated with the local hyperostosis described in the targeted whole-body knockout model. Therefore, current evidence more strongly supports bone erosion, reduced bone mass, and osteoclast abnormalities as well-established bone-remodeling phenotypes of the Pstpip2cmo and Pstpip2Lupo models, whereas hyperostosis cannot currently be regarded as a shared phenotype across different Pstpip2-deficient models. In addition to skeletal involvement, Pstpip2-deficient models can exhibit varying degrees of periarticular soft-tissue and cutaneous lesions. Some Pstpip2cmo mice develop erythema and thickening of the ears, accompanied by auricular cartilage destruction [14]. The Pstpip2Lupo model can develop inflammation of the paws and ears, which may subsequently progress to ulceration, exudation, crusting, and tissue necrosis [8]. In the Pstpip2−/− model, infiltration of neutrophils and macrophages into the skin has also been observed [9]. Systemic myeloid inflammatory phenotypes of Pstpip2-deficient models In addition to local bone, joint, and soft-tissue lesions, Pstpip2-deficient models can exhibit varying degrees of systemic myeloid immune abnormalities, among which the Pstpip2cmo model has been the most extensively characterized. Pstpip2cmo mice develop splenomegaly, extramedullary hematopoiesis, and expansion of splenic myeloid progenitor cells, while mature myeloid cells also display enhanced responsiveness. Serum macrophage inflammatory protein-1α (MIP-1α/CCL3) and IL-6 levels are elevated even during the asymptomatic stage, and additional inflammatory mediators, including TNF-α, may increase after clinical manifestations emerge [8]. These findings support the presence of systemic myeloid immune priming in the Pstpip2cmo model and indicate that local sterile osteitis is accompanied by broader systemic immune abnormalities. The Pstpip2Lupo model also exhibits splenomegaly and abnormalities in certain circulating inflammatory mediators, although its inflammatory cytokine profile does not completely overlap with that of the Pstpip2cmo model. For example, although MIP-1α levels are elevated in Pstpip2Lupo mice, no significant increase in serum IL-6 has been observed at either 3 or 8 months of age [8,16]. The Pstpip2−/− model can likewise develop splenomegaly and lymphadenopathy, accompanied by increased neutrophil numbers in the bone marrow, spleen, and lymph nodes [9]. Collectively, these findings indicate that systemic myeloid immune abnormalities occur across different Pstpip2-deficient models, although the profiles of inflammatory mediators and the magnitude of these abnormalities vary among models. It should be noted that existing studies of these models differ in genetic background, age at assessment, and phenotypic evaluation methods, and some parameters lack consistently reported quantitative data. Therefore, the differences in local tissue and systemic inflammatory phenotypes described here are based primarily on an integrated comparison of independent studies rather than on direct head-to-head comparisons performed under identical experimental conditions. Phenotypic similarities and differences between Pstpip2-deficient mice and human SAPHO syndrome and model positioning Although Pstpip2-deficient mice can recapitulate SAPHO-related skeletal phenotypes, including sterile osteitis, bone erosion, and abnormal bone remodeling (Table 1), the anatomical distribution of lesions differs from that observed in human disease. In patients with SAPHO syndrome, osteoarticular lesions frequently involve the anterior chest wall, particularly the sternoclavicular joint and sternocostoclavicular region [17]. In contrast, skeletal lesions in Pstpip2-deficient mice predominantly affect the paws, distal limbs, and caudal vertebrae.Table 1Major phenotypic similarities and differences between human SAPHO syndrome and Pstpip2-deficient mice.Table 1Comparison dimensionPatients with SAPHO syndromePstpip2-deficient modelsModel significanceDisease courseChronic, relapsing, or progressive courseProgressive inflammation and bone damage over the disease courseRecapitulates some features of chronic bone remodelingSkeletal involvementSterile osteitis, bone destruction, and hyperostosisPstpip2cmo: osteitis, bone erosion, and new bone formation; Pstpip2Lupo: local osteolysis; both models show systemic trabecular bone loss; whole-body knockout: hyperostosisRecapitulates sterile osteitis and abnormal bone remodelingSoft-tissue/cutaneous manifestationsMay be accompanied by cutaneous involvementPstpip2cmo: ear inflammation and auricular cartilage destruction; Pstpip2Lupo: ear and paw inflammation; whole-body knockout: inflammatory-cell infiltration in the skinPartially recapitulates extraosseous inflammatory manifestationsSystemic inflammatory backgroundNeutrophil activation and IL-1-related inflammatory abnormalitiesPstpip2cmo: splenomegaly, extramedullary hematopoiesis, and myeloid progenitor expansion; Pstpip2Lupo: splenomegaly and abnormalities in circulating inflammatory mediators; whole-body knockout: splenomegaly, lymphadenopathy, and neutrophiliaExhibits systemic myeloid inflammatory featuresDistribution of involved sitesThe anterior chest wall is commonly involved, particularly the sternocostoclavicular regionPredominantly involves the paws, distal limbs, and caudal vertebraeAnatomical distribution differs, but the model remains useful for studying related inflammatory and bone-remodeling mechanismsInflammatory-cell characteristicsEnhanced activation of peripheral blood neutrophilsNeutrophil-dominant myeloid immune abnormalitiesShows similarities in myeloid inflammatory featuresInflammatory mediator profileElevated IL-1 family cytokinesPstpip2cmo: increased MIP-1α and IL-6; Pstpip2Lupo: increased MIP-1α without a significant increase in IL-6Shows mechanistic overlap in IL-1-related inflammation Table 1Major phenotypic similarities and differences between human SAPHO syndrome and Pstpip2-deficient mice.Table 1Comparison dimensionPatients with SAPHO syndromePstpip2-deficient modelsModel significanceDisease courseChronic, relapsing, or progressive courseProgressive inflammation and bone damage over the disease courseRecapitulates some features of chronic bone remodelingSkeletal involvementSterile osteitis, bone destruction, and hyperostosisPstpip2cmo: osteitis, bone erosion, and new bone formation; Pstpip2Lupo: local osteolysis; both models show systemic trabecular bone loss; whole-body knockout: hyperostosisRecapitulates sterile osteitis and abnormal bone remodelingSoft-tissue/cutaneous manifestationsMay be accompanied by cutaneous involvementPstpip2cmo: ear inflammation and auricular cartilage destruction; Pstpip2Lupo: ear and paw inflammation; whole-body knockout: inflammatory-cell infiltration in the skinPartially recapitulates extraosseous inflammatory manifestationsSystemic inflammatory backgroundNeutrophil activation and IL-1-related inflammatory abnormalitiesPstpip2cmo: splenomegaly, extramedullary hematopoiesis, and myeloid progenitor expansion; Pstpip2Lupo: splenomegaly and abnormalities in circulating inflammatory mediators; whole-body knockout: splenomegaly, lymphadenopathy, and neutrophiliaExhibits systemic myeloid inflammatory featuresDistribution of involved sitesThe anterior chest wall is commonly involved, particularly the sternocostoclavicular regionPredominantly involves the paws, distal limbs, and caudal vertebraeAnatomical distribution differs, but the model remains useful for studying related inflammatory and bone-remodeling mechanismsInflammatory-cell characteristicsEnhanced activation of peripheral blood neutrophilsNeutrophil-dominant myeloid immune abnormalitiesShows similarities in myeloid inflammatory featuresInflammatory mediator profileElevated IL-1 family cytokinesPstpip2cmo: increased MIP-1α and IL-6; Pstpip2Lupo: increased MIP-1α without a significant increase in IL-6Shows mechanistic overlap in IL-1-related inflammation Major phenotypic similarities and differences between human SAPHO syndrome and Pstpip2-deficient mice. This difference in lesion distribution may be related to distinct biomechanical environments in humans and mice. In autoinflammatory bone diseases, repetitive biomechanical stress has been proposed to contribute to inflammation at specific anatomical sites [18]. The sternoclavicular joint is the only bony articulation connecting the upper limb to the axial skeleton in humans and serves as an important site for mechanical force transmission between the upper limb and trunk [19]. Anterior chest wall lesions in SAPHO syndrome show a marked regional predilection, primarily involving the sternocostoclavicular region and the first rib [20]. In contrast, mice exhibit quadrupedal locomotion, during which body weight and ground reaction forces are transmitted predominantly through the forelimbs and hindlimbs [21]. Accordingly, differences in locomotion and local skeletal loading patterns between humans and mice may contribute to their distinct lesion distributions, although this explanation has not yet been directly validated by mechanical loading or unloading experiments. Moreover, whether PSTPIP2 expression, bone development, and bone-remodeling characteristics differ across skeletal sites between humans and mice, and whether host microbiota contribute to anatomical lesion selection, remain unclear. Therefore, the specific contributions of these factors to the differences in lesion distribution between Pstpip2-deficient mice and human SAPHO syndrome require further investigation. Beyond differences in lesion distribution, the extent to which Pstpip2-deficient mice recapitulate the hyperostotic phenotype of SAPHO syndrome also requires further evaluation. Long-term follow-up studies have shown that entheseal involvement may occur as an early manifestation of human SAPHO syndrome and may gradually progress to hyperostosis and bony fusion over the disease course [22]. A histomorphometric study of a hyperostotic lesion from a patient with SAPHO syndrome revealed high bone turnover, accompanied by abundant osteoid and woven bone formation and increased numbers of alkaline phosphatase (ALP)-positive osteoblasts along the bone surface. These findings suggest that, at least in some cases, SAPHO-associated hyperostosis is related to active new bone formation [23]. In contrast, studies of Pstpip2-deficient models have focused predominantly on myeloid inflammation, enhanced osteoclast differentiation, and inflammatory bone resorption. Although Pstpip2−/− mice can develop local hyperostosis and late-stage lesions in the Pstpip2cmo model may exhibit new bone formation, osteoblast activity and bone turnover status in these models have not been adequately characterized. Therefore, it remains uncertain whether local hyperostosis in the systemic Pstpip2−/− model and late-stage new bone formation in the Pstpip2cmo model faithfully recapitulate the chronic hyperostotic process observed in human SAPHO syndrome. From a genetic perspective, a direct association between Pstpip2 and human SAPHO syndrome has not been established. Candidate gene analysis of PSTPIP2, NOD2, and LPIN2 in 38 patients with SAPHO syndrome identified no clear association between variants in these genes and the disease [24]. Thus, Pstpip2 cannot currently be regarded as a definitive causative gene for human SAPHO syndrome. Nevertheless, human SAPHO syndrome and Pstpip2-deficient models share certain features of myeloid inflammation. Patients with SAPHO syndrome may exhibit enhanced activation of peripheral blood neutrophils and elevated levels of IL-1 family cytokines, while IL-1 antagonists have shown therapeutic effects on osteoarticular inflammation in some patients [3]. These findings support mechanistic overlap between human SAPHO syndrome and Pstpip2-deficient models in terms of myeloid innate immune dysregulation and IL-1-related inflammatory pathways. Against this background, comparison of Pstpip2-deficient mice with other models of related autoinflammatory diseases can further clarify their value as experimental models. Pyogenic arthritis, pyoderma gangrenosum, and acne (PAPA) syndrome is caused by pathogenic variants in PSTPIP1, among which A230T and E250Q are well-established disease-associated variants [25]. However, neither mice ectopically expressing PAPA-associated mutant PSTPIP1 proteins nor subsequently developed PSTPIP1 mutant knock-in models have fully reproduced the characteristic purulent arthritis and cutaneous inflammatory phenotypes of human PAPA syndrome [26,27]. Similarly, although LPIN2 mutations have a well-established genetic association with human Majeed syndrome, Lpin2−/− mice do not spontaneously develop osteomyelitis [28]. These findings indicate that genetic correspondence alone is insufficient to ensure complete recapitulation of the human clinical phenotype in mice. On the other hand, spontaneous sterile osteitis is not unique to Pstpip2-deficient models. Chronic recurrent multifocal osteomyelitis (CRMO) and SAPHO syndrome both belong to the spectrum of autoinflammatory bone diseases and share skeletal manifestations such as sterile osteitis, together with articular and extraosseous inflammatory features [29]. Ali18 mice harboring a gain-of-function mutation in Fgr spontaneously develop paw inflammation, sterile osteomyelitis, and systemic reductions in bone mineral density, and variants within the coding region of FGR have also been identified in patients with CRMO [30]. These findings indicate a certain degree of genetic and phenotypic relevance of this model to human CRMO. However, the established phenotypes of the Ali18 model remain focused mainly on CRMO-like sterile osteitis and abnormalities in bone mass, with relatively limited coverage of the multidimensional bone, joint, skin, and abnormal bone-remodeling manifestations characteristic of SAPHO syndrome. In comparison, across different Pstpip2-deficient models, phenotypes including osteitis, bone erosion, hyperostosis, and synovitis have been reported, together with inflammatory-cell infiltration in bone, joint, and skin tissues. In addition, these models share certain mechanistic features with human SAPHO syndrome, particularly abnormalities in myeloid innate immunity and IL-1-related inflammatory pathways. The principal value of Pstpip2-deficient models therefore lies not in reproducing the genetic basis of SAPHO syndrome, but in their ability to recapitulate multiple disease-relevant features involving bone, joints, skin, and abnormal bone remodeling. Accordingly, they are particularly suitable as mechanistic models for investigating SAPHO-related myeloid inflammation, sterile osteitis, and inflammatory bone remodeling. Therapeutic targets, pharmacological intervention evidence, and translational value of Pstpip2-deficient models A growing body of evidence regarding therapeutic targets and pharmacological interventions has accumulated in Pstpip2-deficient models, although current studies remain dominated by genetic target-validation approaches, with relatively few direct pharmacological intervention studies. Among these targets, the IL-1β/IL-1R1 axis is supported by strong genetic evidence for a causal role, whereas direct in vivo pharmacological studies have mainly involved the CSF-1R/c-Kit inhibitor PLX3397 and Wilforine [16,31,32]. These findings indicate that Pstpip2-deficient models have potential utility for therapeutic target validation and pharmacological evaluation (Table 2).Table 2Comparison of therapeutic targets in Pstpip2-deficient models with clinical treatment evidence in SAPHO syndrome.Table 2Intervention target/drugEvidence in Pstpip2-deficient modelsClinical treatment evidence in SAPHO syndromeModel-clinical correspondenceCSF-1R/c-Kit inhibitor: PLX3397Ameliorates extramedullary hematopoiesis, inflammation, and bone loss [16]Corresponding targeted clinical evidence is currently lackingEffective in the animal model; clinical translation has not yet been establishedWilforineReduces bone erosion, bone marrow edema, and inflammatory infiltration and inhibits osteoclast fusion [32]Tripterygium wilfordii preparations may improve symptoms and imaging findings in some patients [33,34]An indirect pharmacological association exists, but the two cannot be considered directly equivalentIL-1β/IL-1R1Genetic deletion delays disease onset and markedly reduces bone damage [31]IL-1 blockade is effective for osteoarticular manifestations in some patients [35]Shows a degree of mechanistic-clinical correspondenceJAK inhibitorsWilforine involves inhibition of JAK-STAT signaling, but direct validation with a specific JAK inhibitor is lackingJAK inhibitors have shown some clinical efficacy [36,37]Mechanistic association exists, but direct model validation is lackingTNF inhibitorsCorresponding pharmacological validation is lackingShow relatively high response rates for osteoarticular manifestations and may improve some cutaneous manifestations [35,37]Clinical efficacy is relatively well established, but model validation remains insufficientBisphosphonatesCorresponding pharmacological validation is lackingCan improve osteoarticular manifestations [38,39]Clinical efficacy is relatively well established, but model validation remains insufficient Table 2Comparison of therapeutic targets in Pstpip2-deficient models with clinical treatment evidence in SAPHO syndrome.Table 2Intervention target/drugEvidence in Pstpip2-deficient modelsClinical treatment evidence in SAPHO syndromeModel-clinical correspondenceCSF-1R/c-Kit inhibitor: PLX3397Ameliorates extramedullary hematopoiesis, inflammation, and bone loss [16]Corresponding targeted clinical evidence is currently lackingEffective in the animal model; clinical translation has not yet been establishedWilforineReduces bone erosion, bone marrow edema, and inflammatory infiltration and inhibits osteoclast fusion [32]Tripterygium wilfordii preparations may improve symptoms and imaging findings in some patients [33,34]An indirect pharmacological association exists, but the two cannot be considered directly equivalentIL-1β/IL-1R1Genetic deletion delays disease onset and markedly reduces bone damage [31]IL-1 blockade is effective for osteoarticular manifestations in some patients [35]Shows a degree of mechanistic-clinical correspondenceJAK inhibitorsWilforine involves inhibition of JAK-STAT signaling, but direct validation with a specific JAK inhibitor is lackingJAK inhibitors have shown some clinical efficacy [36,37]Mechanistic association exists, but direct model validation is lackingTNF inhibitorsCorresponding pharmacological validation is lackingShow relatively high response rates for osteoarticular manifestations and may improve some cutaneous manifestations [35,37]Clinical efficacy is relatively well established, but model validation remains insufficientBisphosphonatesCorresponding pharmacological validation is lackingCan improve osteoarticular manifestations [38,39]Clinical efficacy is relatively well established, but model validation remains insufficient Comparison of therapeutic targets in Pstpip2-deficient models with clinical treatment evidence in SAPHO syndrome. However, intervention evidence from Pstpip2-deficient models does not fully correspond to the clinical efficacy observed in SAPHO syndrome. Although IL-1-related findings show broadly consistent effects between experimental models and clinical observations, the experimental efficacy of Wilforine has only an indirect relationship with previous clinical use of Tripterygium wilfordii preparations in SAPHO syndrome. Moreover, clinically important therapies such as TNF inhibitors and bisphosphonates have not yet undergone corresponding pharmacological validation in Pstpip2-deficient models [[33], [34], [35], [36], [37], [38], [39]]. Overall, current evidence provides a preliminary basis for using Pstpip2-deficient models to investigate therapeutic targets and pharmacological interventions, but the correspondence between preclinical evidence and clinically used therapies for SAPHO syndrome remains incomplete. Future studies could prioritize pharmacological validation of treatments with stronger clinical evidence, such as TNF inhibitors and bisphosphonates, to further evaluate and extend the translational value of these models for therapeutic research in SAPHO syndrome. Mechanistic basis of inflammation and bone remodeling in Pstpip2-deficient models Aberrant neutrophil activation and initiation of inflammation Aberrant neutrophil activation is an important early pathological feature of Pstpip2-deficient models. Under physiological conditions, PSTPIP2 participates in the negative regulation of NADPH oxidase 2 (NOX2)-mediated reactive oxygen species (ROS) production in neutrophils. Loss of PSTPIP2 function results in excessive ROS generation and a sustained state of elevated ROS production (Fig. 1A), which precedes the onset of clinical manifestations [40]. To further define the role of NOX2-derived ROS in disease initiation and bone damage, Pstpip2cmo mice were genetically crossed with gp91phox-deficient mice. Despite the combined deficiency, these mice still exhibited high levels of IL-1β release and soft-tissue swelling, whereas osteolysis was suppressed [40]. These findings indicate that NOX2-derived ROS are not required for IL-1β release or soft-tissue inflammation in the context of Pstpip2 deficiency, but play an important role in the development and progression of osteolysis.Fig. 1Mechanistic basis of inflammation and bone remodeling in Pstpip2-deficient models.(A) Pstpip2 deficiency leads to aberrant neutrophil activation, involving NOX2/ROS signaling, SYK/NF-κB-mediated pro-IL-1β production, and IL-1β maturation, and may promote neutrophil recruitment to local lesions through the CXCL2–CXCR2 axis. (B) Pstpip2 deficiency promotes myeloid progenitor expansion, recruitment of osteoclast precursors, and osteoclast differentiation, thereby enhancing bone resorption and contributing to osteolysis. (C) Evidence from other inflammatory bone diseases suggests that osteoclast–osteoblast coupling and inflammation-associated osteogenic signaling may contribute to pathological new bone formation in Pstpip2-deficient models. (D) Pathogenic IL-1β is derived predominantly from the hematopoietic compartment and acts on the IL-1R1-expressing non-hematopoietic compartment. Local stromal cells, synovial fibroblasts, and endothelial cells may participate in translating inflammatory signals into bone-remodeling responses. Solid arrows indicate relationships directly supported by evidence from Pstpip2-deficient models, whereas dashed arrows indicate potential mechanisms supported indirectly or inferred from related experimental systems. Dashed boxes indicate candidate mechanisms or cell types that have not yet been directly established in Pstpip2-deficient models. Mechanistic basis of inflammation and bone remodeling in Pstpip2-deficient models. (A) Pstpip2 deficiency leads to aberrant neutrophil activation, involving NOX2/ROS signaling, SYK/NF-κB-mediated pro-IL-1β production, and IL-1β maturation, and may promote neutrophil recruitment to local lesions through the CXCL2–CXCR2 axis. (B) Pstpip2 deficiency promotes myeloid progenitor expansion, recruitment of osteoclast precursors, and osteoclast differentiation, thereby enhancing bone resorption and contributing to osteolysis. (C) Evidence from other inflammatory bone diseases suggests that osteoclast–osteoblast coupling and inflammation-associated osteogenic signaling may contribute to pathological new bone formation in Pstpip2-deficient models. (D) Pathogenic IL-1β is derived predominantly from the hematopoietic compartment and acts on the IL-1R1-expressing non-hematopoietic compartment. Local stromal cells, synovial fibroblasts, and endothelial cells may participate in translating inflammatory signals into bone-remodeling responses. Solid arrows indicate relationships directly supported by evidence from Pstpip2-deficient models, whereas dashed arrows indicate potential mechanisms supported indirectly or inferred from related experimental systems. Dashed boxes indicate candidate mechanisms or cell types that have not yet been directly established in Pstpip2-deficient models. In addition to dysregulated ROS production, Pstpip2 deficiency may influence neutrophil recruitment to local lesions by altering chemokine expression. Previous studies have shown that PSTPIP2 regulates chemokine production through interactions with PEST-PTPs (PEST-family protein tyrosine phosphatases), and that loss of PSTPIP2 markedly increases CXCL2 expression in neutrophils [10]. CXCL2 specifically binds to CXCR2 on the neutrophil surface and promotes neutrophil transendothelial migration [41]. These observations suggest that the CXCL2–CXCR2 signaling axis may contribute to the formation of local inflammatory lesions in the setting of Pstpip2 deficiency, although a causal role has not yet been directly demonstrated. IL-1β production and inflammasome-independent maturation Within the local inflammatory microenvironment of Pstpip2-deficient models, IL-1β is considered a key pathogenic mediator driving bone destruction. Genetic evidence has shown that crossing Pstpip2cmo mice with IL-1β- or IL-1R1-deficient mice delays disease onset and almost completely reverses skeletal damage [42], whereas crossing with IL-1α-deficient mice does not confer a similar protective effect [31]. These findings indicate that IL-1β is a central effector molecule driving the initiation of bone inflammation and the progression of bone destruction in this model. With respect to cellular sources, Pstpip2cmo neutrophils exhibit markedly increased IL-1β release in response to sterile stimulation, whereas bone marrow-derived macrophages do not show the same exaggerated response [42], further supporting an important role for neutrophils in IL-1β-driven inflammation in this model. At the level of IL-1β production, aberrant signaling in myeloid cells may promote the accumulation of pro-IL-1β. In the Pstpip2cmo background, LysM-Cre-mediated conditional deletion of Syk prevents disease development and reduces NF-κB activation and pro-IL-1β expression in paw tissues [43], indicating that myeloid SYK/NF-κB signaling contributes to the abnormal production of pro-IL-1β associated with Pstpip2 deficiency. Further studies have shown that Pstpip2cmo mice still develop disease with either Nlrp3 or Casp1 deficiency. Casp8 deficiency alone likewise fails to prevent disease; however, bone disease is markedly suppressed when Casp8 deficiency is combined with Casp1 or Nlrp3 deficiency [42,44,45]. These findings suggest functional redundancy between the canonical NLRP3–caspase-1 inflammasome pathway and caspase-8-related pathways in IL-1β processing. In addition, neutrophil-derived serine proteases (NSPs) may contribute to noncanonical IL-1β processing. The abnormally increased release of IL-1β from Pstpip2cmo neutrophils can be inhibited by serine protease inhibitors; together with previous evidence that NSPs such as proteinase 3 (PR3) and neutrophil elastase (NE) can cleave pro-IL-1β [42,46,47], these findings suggest that NSPs may participate in the generation of mature IL-1β in this model (Fig. 1A). Myeloid progenitor expansion, osteoclast differentiation, and osteolysis Bone lesions resulting from Pstpip2 deficiency reflect disruption of osteoimmune homeostasis. PSTPIP2 functions as a negative-feedback regulator of macrophage colony-stimulating factor 1 receptor (CSF-1R) and c-Kit signaling. Under physiological conditions, phosphorylated PSTPIP2 recruits PEST-family phosphatases to restrain proliferative and proinflammatory signaling [16]. Pstpip2 deficiency leads to aberrant expansion of myeloid progenitor cells and enhances the responsiveness of osteoclast precursors to RANKL and CSF-1. Even at physiological concentrations of RANKL and CSF-1, these precursors can differentiate into multinucleated osteoclasts with high bone-resorptive capacity (Fig. 1B), suggesting that Pstpip2 deficiency provides a cellular basis for enhanced bone resorption [15,16]. The dual CSF-1R/c-Kit inhibitor PLX3397 suppresses the expansion of these myeloid progenitors and their differentiation into osteoclasts while alleviating bone damage, providing pharmacological support for this mechanism [16]. In addition, Pstpip2 deficiency is associated with increased secretion of MIP-1α/CCL3, which further promotes the recruitment of osteoclast precursors to affected sites and exacerbates osteolysis [15]. Taken together, aberrant expansion of myeloid progenitors, recruitment of osteoclast precursors, and enhanced osteoclast differentiation constitute an important cellular basis for osteolysis in Pstpip2-deficient models. Potential mechanisms of abnormal bone formation and hyperostosis As described above, abnormal bone-forming phenotypes, including new bone formation, osteosclerosis, and hyperostosis in some models, have been reported across Pstpip2-deficient models. [9,14,48], together with markedly enhanced osteoclast differentiation and bone-resorptive activity [16]. Studies of other inflammatory bone diseases provide two potential mechanistic explanations for this phenotype. First, enhanced osteoclastic activity may promote subsequent bone formation through osteoclast–osteoblast coupling. A study integrating facet joint tissues from patients with ankylosing spondylitis (AS), human osteoprogenitor cells, and curdlan-induced SKG mice showed that osteoclast-derived DEL1 promotes osteoblast differentiation and pathological new bone formation through αVβ3–RUNX2 signaling [49]. Studies of spinal ligament tissues from patients with AS have also shown that TGF-β activation following osteoclastic bone resorption can recruit osteoprogenitor cells and promote ectopic bone formation [50], suggesting that abnormal osteoclast activity may facilitate subsequent pathological osteogenesis through coupling mechanisms. Second, pathological bone formation may also be directly driven by inflammation-associated osteogenic signaling. In the DBA/1 spontaneous ankylosing enthesitis model, BMP signaling promotes the participation of entheseal progenitor cells in endochondral ossification, whereas BMP blockade markedly suppresses joint ankylosis [51]. Studies of bone tissues and primary bone-derived cells from patients with AS further demonstrated that IL-17A can directly promote osteoblast differentiation and mineralization through JAK2/STAT3 signaling [52]. Collectively, these findings indicate that abnormal new bone formation in inflammatory bone diseases may arise either secondary to osteoclast–osteoblast coupling or through direct stimulation of osteogenesis by local inflammatory signals, providing two potential explanations for the coexistence of bone destruction and bone proliferation in Pstpip2-deficient models. However, because current studies remain largely based on endpoint pathological observations, it is still unclear whether hyperostosis in these models primarily reflects coupled repair following bone damage or a relatively independent pathological osteogenic process distinct from bone erosion (Fig. 1C). Functional division between hematopoietic and non-hematopoietic compartments Bone marrow chimera experiments have further revealed distinct roles for hematopoietic and non-hematopoietic compartments in Pstpip2-deficient models. These studies showed that pathogenic IL-1β is derived predominantly from radiosensitive hematopoietic cells, whereas IL-1R1 expression in the radioresistant non-hematopoietic compartment is essential for disease development [45]. These findings indicate that, in the context of Pstpip2 deficiency, pathogenic IL-1β is generated mainly by hematopoietic cells and acts on the non-hematopoietic compartment. However, bone marrow chimera experiments do not further identify the specific non-hematopoietic cell populations responsible for IL-1R1-mediated responses. Studies of other inflammatory bone diseases provide some clues regarding the cell types that may mediate these responses. Bone marrow stromal cells can respond to IL-1 through IL-1R1 and promote osteoclast formation by upregulating RANKL; in IL-1R1-deficient stromal cells, TNF-induced RANKL expression and osteoclastogenesis are markedly attenuated [53]. Synovial fibroblasts can likewise regulate inflammatory bone erosion. In collagen antibody-induced arthritis and collagen-induced arthritis models, fibroblast-specific deletion of RANKL significantly reduces osteoclast formation and bone erosion [54]. In addition, in vitro studies have shown that inflammatory mediators such as IL-1 can modulate RANKL/OPG expression in human microvascular endothelial cells and promote osteoclast formation, while also enhancing the adhesion and transendothelial migration of CD14+ osteoclast precursors [55,56]. These findings suggest that bone marrow stromal cells, synovial fibroblasts, and vascular endothelial cells may all contribute to the translation of inflammatory signals into local bone-remodeling responses. Overall, current evidence provides an initial framework in which pathogenic IL-1β is produced by hematopoietic cells and sensed through IL-1R1 in the non-hematopoietic compartment, while suggesting that the specific responding cells may reside in local tissues. However, the precise cell types involved remain to be defined (Fig. 1D). Challenges in model establishment: factors affecting breeding efficiency and phenotypic stability Breeding impairment under chronic inflammatory conditions Although Pstpip2-deficient mice can reproduce certain pathological features of SAPHO syndrome, their practical application is limited by factors such as a low yield of homozygous offspring and insufficient phenotypic stability. This low yield is unlikely to reflect breeding technique alone and may also be related to multisystem pathological changes that develop during disease progression (Fig. 2).Fig. 2Potential mechanisms underlying breeding impairment caused by chronic systemic inflammation in Pstpip2-deficient mice.Neuroendocrine regulation: Chronic inflammation and elevated IL-1β may suppress the hypothalamic–pituitary–gonadal (HPG) axis, thereby affecting GnRH secretion and reproductive endocrine function. Reproductive function: Abnormal Pstpip2 expression in females may impair endometrial receptivity and embryo implantation, whereas chronic inflammation in males may suppress Leydig cell function, leading to reduced testosterone levels and impaired sperm motility. Immunometabolism: Persistent inflammation may induce energy reallocation toward immune responses, thereby reducing energy investment in reproductive processes. Together, these neuroendocrine, reproductive, and immunometabolic abnormalities may contribute to impaired reproductive function and reduced breeding efficiency. Potential mechanisms underlying breeding impairment caused by chronic systemic inflammation in Pstpip2-deficient mice. Neuroendocrine regulation: Chronic inflammation and elevated IL-1β may suppress the hypothalamic–pituitary–gonadal (HPG) axis, thereby affecting GnRH secretion and reproductive endocrine function. Reproductive function: Abnormal Pstpip2 expression in females may impair endometrial receptivity and embryo implantation, whereas chronic inflammation in males may suppress Leydig cell function, leading to reduced testosterone levels and impaired sperm motility. Immunometabolism: Persistent inflammation may induce energy reallocation toward immune responses, thereby reducing energy investment in reproductive processes. Together, these neuroendocrine, reproductive, and immunometabolic abnormalities may contribute to impaired reproductive function and reduced breeding efficiency. Skeletal lesions themselves may directly impair breeding efficiency. These mice typically develop severe bone destruction at 6–8 weeks of age, a period that overlaps with sexual maturation, and limb swelling and restricted mobility markedly reduce mating success [42]. Even after successful conception and delivery, persistent inflammation, limb lesions, and disease-associated wasting may impair maternal nesting and lactation behaviors [57], further reducing overall breeding efficiency. At the neuroendocrine level, persistently elevated IL-1β associated with Pstpip2 deficiency may suppress the hypothalamic–pituitary–gonadal (HPG) axis [31]. As a potent inhibitor of the HPG axis, IL-1β can cross the blood–brain barrier and act on hypothalamic circumventricular regions, altering the distribution and activity of neurotransmitters such as norepinephrine and thereby suppressing pulsatile secretion of gonadotropin-releasing hormone (GnRH) [58]. In addition, Pstpip2 has been identified as a gene associated with endometrial receptivity, and abnormal expression may directly interfere with embryo implantation [59]. In males, chronic inflammation has been shown to suppress steroidogenesis in testicular Leydig cells in a dose-dependent manner, leading to reduced testosterone levels and impaired sperm motility [60]. Persistent inflammatory exposure may also drive systemic energy reallocation, further suppressing nonessential processes such as reproduction [61]. In addition to disease-related reproductive impairment, the recessive inheritance pattern itself limits the efficiency of generating homozygous animals. Heterozygous mice (wt/cmo) can generally be used to maintain breeding colonies, but intercrossing heterozygotes yields a theoretical homozygous proportion of only 25%. When the adverse effects of progressive disease on breeding performance and survival of homozygous animals are taken into account, the practical use of this model may involve substantial breeding costs and a very low effective model yield. Modulation of disease phenotypes by gut microbiota, diet, and genetic background Even in the presence of the same Pstpip2 functional defect, disease phenotypes remain jointly modulated by the gut microbiota and host genetic background (Fig. 3). These effects are reflected in multiple aspects, including disease onset and the severity of bone destruction, highlighting the strong gene–environment dependence of phenotypes associated with Pstpip2 deficiency.Fig. 3Mechanisms by which gut microbiota, diet, and host genetic background jointly modulate disease phenotypes in Pstpip2-deficient mice.Diet–microbiota axis: A low-fat diet (LFD) is associated with Prevotella expansion and more severe bone lesions, whereas a high-fat diet (HFD) suppresses Prevotella expansion and alleviates inflammation and bone erosion. Antibiotic treatment and fecal microbiota transplantation studies suggest that the gut microbiota primarily acts as a phenotypic modifier in the context of genetic susceptibility. Genetic background: The same Pstpip2cmo mutation can result in different times of disease onset, inflammatory intensity, and severity of bone damage across distinct genetic backgrounds. Together, the diet–microbiota axis and host genetic background influence the threshold for disease manifestation and thereby modulate disease onset and severity. Mechanisms by which gut microbiota, diet, and host genetic background jointly modulate disease phenotypes in Pstpip2-deficient mice. Diet–microbiota axis: A low-fat diet (LFD) is associated with Prevotella expansion and more severe bone lesions, whereas a high-fat diet (HFD) suppresses Prevotella expansion and alleviates inflammation and bone erosion. Antibiotic treatment and fecal microbiota transplantation studies suggest that the gut microbiota primarily acts as a phenotypic modifier in the context of genetic susceptibility. Genetic background: The same Pstpip2cmo mutation can result in different times of disease onset, inflammatory intensity, and severity of bone damage across distinct genetic backgrounds. Together, the diet–microbiota axis and host genetic background influence the threshold for disease manifestation and thereby modulate disease onset and severity. Diet and the gut microbiota can first influence disease incidence and the severity of bone damage. Under a low-fat diet (LFD), Pstpip2-deficient mice exhibit progressive expansion of Prevotella and reduction of Lactobacillus over the disease course, accompanied by a higher incidence of inflammatory bone disease and marked bone erosion. In contrast, a high-fat diet (HFD) suppresses Prevotella expansion and alleviates paw inflammation, bone erosion, and inflammatory-cell infiltration. Antibiotic treatment and fecal microbiota transplantation further demonstrate that different microbial states are associated not only with disease severity but also with the course of osteitis development. However, microbiota derived from diseased mice are insufficient to independently induce osteitis in wild-type hosts, suggesting that the gut microbiota primarily acts as a phenotypic modifier in the context of Pstpip2-associated genetic susceptibility [62]. Thus, diet–microbiota interactions are more likely to modulate the threshold for disease development and the intensity of inflammation rather than independently determine the disease phenotype. From an immunological perspective, modulation of the Pstpip2-deficient phenotype by diet and the gut microbiota may also involve intestinal barrier function and immunometabolic pathways. In general, chronic dysbiosis can be accompanied by alterations in intestinal mucosal barrier function, allowing microbial pathogen-associated molecular patterns (PAMPs) and metabolites to exert greater effects on peripheral innate immune responses [63]. In the absence of the negative regulatory effects of PSTPIP2, such microbial signals may further lower the activation threshold of neutrophils and macrophages and amplify IL-1β-related inflammation. In addition to reshaping the gut microbiota, an HFD may also participate in inflammatory regulation by altering the host metabolic environment. β-Hydroxybutyrate is an important metabolite generated during fatty acid oxidation and ketogenesis and has been shown in other inflammatory models to promote STAT6-dependent M2 macrophage polarization and facilitate inflammation resolution [64]. These findings provide a potential mechanistic explanation for the combined modulation of Pstpip2-deficient phenotypes by diet–microbiota interactions and host immunometabolism, although this mechanism has not yet been directly demonstrated in this model. In addition to environmental factors, the genetic background of mice is an important intrinsic determinant of disease severity in this model. When the same Pstpip2cmo mutation is introduced into BALB/c, C57BL/6J, and C57BL/6NCrl backgrounds, all three strains eventually develop osteomyelitis; however, BALB/c and C57BL/6J mice show later disease onset and milder bone destruction and soft-tissue swelling, whereas C57BL/6NCrl mice develop earlier onset and more severe bone damage. Correspondingly, lesions in C57BL/6NCrl mice exhibit greater neutrophil and monocyte infiltration, higher levels of CCL3 and CXCL2, and more pronounced neutrophil activation. In contrast, BALB/c mice show higher expression of inhibitory receptors such as PIR-B and SIRP-α, together with weaker neutrophil activation and degranulation [65]. These findings indicate that phenotypes resulting from Pstpip2 deficiency are not determined solely by a single major-effect mutation; additional genetic variants in the host background can further modulate disease onset, inflammatory intensity, and the severity of bone damage. Summary and future perspectives Overall, Pstpip2-deficient mice exhibit myeloid immune abnormalities, IL-1β-driven bone destruction, and abnormal bone remodeling. Current evidence supports aberrant neutrophil activation, dysregulated IL-1β production and processing, and enhanced osteoclast differentiation as major pathogenic mechanisms in these models. The functional division between hematopoietic cell-derived IL-1β and IL-1R1-mediated responses in non-hematopoietic compartments further reveals how inflammatory signals are transmitted from hematopoietic cells to local tissues to drive bone damage. Although these models share certain features of osteitis, bone remodeling, and myeloid inflammation with SAPHO syndrome, Pstpip2 has not been established as a definitive causative gene for human SAPHO syndrome, and differences remain between mice and patients in lesion distribution and several phenotypic characteristics. Moreover, disease manifestations in different models are modulated by factors such as genetic background, diet, and gut microbiota. Therefore, Pstpip2-deficient models are more appropriately used to investigate mechanisms of SAPHO-related myeloid inflammation and inflammatory bone damage rather than being regarded as fully homologous surrogates for the human disease. Future studies should further enhance the research value of Pstpip2-deficient models at three levels: mechanistic investigation, model optimization, and clinical translation. At the mechanistic level, priority should be given to identifying the specific cellular basis through which IL-1β signals are transmitted from hematopoietic cells to local non-hematopoietic cells and to clarifying the relationship between enhanced bone resorption and abnormal new bone formation. At the model level, key variables such as genetic background, diet, and gut microbiota should be more clearly defined and standardized to improve phenotypic stability and interstudy reproducibility. In addition, direct comparisons among the Pstpip2cmo, Pstpip2Lupo, and Pstpip2−/− models should be conducted under experimental conditions that are as closely matched as possible to further define their shared phenotypes, model-specific features, and appropriate research applications. Breeding and colony-maintenance strategies should also be optimized to address the low efficiency of obtaining homozygous animals and the breeding impairment associated with disease progression, thereby improving the reliable availability of experimental animals. At the translational level, bidirectional validation between intervention studies in these models and clinical treatment evidence from SAPHO syndrome should be strengthened to further define the applicability and translational value of these models for investigating disease mechanisms, validating therapeutic targets, and evaluating candidate therapies. CRediT authorship contribution statement Yu Wang: Writing – review & editing, Writing – original draft. Bowen Feng: Writing – review & editing, Writing – original draft. Yunuo Wang: Writing – original draft. Yi Wang: Investigation. Yingai Xu: Investigation. Mengjiao Gu: Writing – original draft, Supervision. Yuanhao Wu: Supervision, Funding acquisition. Funding This work was supported by the 10.13039/501100012166National Key R&D Program of China (Grant No. 2024YFC3506200, 2024YFC3506201), and the Innovative Team Cultivation Program of 10.13039/100022809First Teaching Hospital of Tianjin University of Traditional Chinese Medicine (Grant No. 2025060301). Declaration of competing interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Yuanhao Wu reports was provided by National Key R&D Program of China. Yuanhao Wu reports was provided by First Teaching Hospital of Tianjin University of Traditional Chinese Medicine. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.