Inflammatory gradients in the infrapatellar fat pad of knee osteoarthritis: implications for joint damage

Abstract

The infrapatellar fat pad (IFP) plays a pivotal role in the pathogenesis of knee osteoarthritis (KOA), exhibiting marked histological changes as the disease progresses. However, the intra-tissue variations within KOA-affected IFP remain poorly understood. In this study, we examined IFP tissues from KOA patients at different disease stages, assessing inflammatory damage through histological evaluation via H&E staining. Based on the extent of tissue damage, we classified IFP regions into inflammatory and non-inflammatory layers. Quantitative PCR (qPCR) and immunohistochemical analyses were then employed to compare the expression of joint damage-associated molecules and immune cell infiltration between these two regions. Our results reveal a pronounced inflammatory response in the IFP tissue adjacent to the synovium (inflammatory layer), while the deeper, non-synovial regions (non-inflammatory layer) showed relatively mild inflammation. Additionally, the inflammatory layer exhibited significantly higher secretion of pro-inflammatory cytokines (IL-1β, TNF-α, and IL-8) and adipokines (Leptin, Adiponectin, and FABP4) compared to the non-inflammatory layer. Notably, B cell infiltration was more prominent in the inflammatory layer than other immune cell types, highlighting its potential role in the progression of KOA. These findings underscore the heterogeneity within the IFP and suggest that localized inflammation, particularly B cell involvement, may contribute to the change of IFP and pathophysiology of KOA.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13075-025-03702-9.

Introduction

Knee osteoarthritis (KOA) is a chronic, low-grade inflammatory condition and represents the most commonly affected site in osteoarthritis [1, 2]. Clinically, KOA is characterized by pain, limited mobility, and joint deformity [1, 3]. With the rapid aging of the population and the extension of life expectancy worldwide, KOA prevalence is increasing year by year among middle-aged and elderly individuals [4]. In these populations, KOA not only serves as a major cause of disability but also has a profound impact on life expectancy [2, 5], while imposing significant economic burdens on families and healthcare systems [2, 4]. The incidence of knee osteoarthritis continues to rise annually, positioning it as one of the leading causes of chronic disability among the elderly. This condition has now emerged as a major global public health concern [1]. The pathophysiology of KOA encompasses a wide range of molecular and structural factors [6, 7], though the precise underlying mechanisms remain incompletely understood [8].

The infrapatellar fat pad (IFP), located within the anterior compartment of the knee joint and outside the synovium, is richly supplied with blood vessels and nerves [9]. In healthy individuals, the IFP serves as a cushion that reduces joint load, lubricates the knee, and supplies blood to the patella and patellar tendon. Consequently, the IFP plays a protective role in normal knee function by mitigating degeneration due to stress fluctuations during movement [10, 11]. However, there has been growing interest in the role of infrapatellar fat pad changes in KOA onset and progression, as imaging evidence of damage in this fat pad is correlated with knee structural damage and symptom exacerbation [12, 13]. Pathological studies have demonstrated marked inflammation, increased angiogenesis, and thickening of the interlobular septa in the IFP of KOA patients, along with elevated expression of cytokines such as VEGF, MCP-1, and IL-6 [14, 15]. Additionally, the IFP’s adipose tissue also secretes a variety of adipokines and cytokines that impact surrounding knee structures, including classic adipokines like leptin, adiponectin, and resistin [16, 17]. These large quantities of inflammatory mediators and adipokines synthesized and secreted from IFP disturbs the intra-articular immune microenvironment and interact with the cartilage, synovium, and subchondral bone, thereby accelerating KOA progression [16–18]. Partial resection of the IFP has been demonstrated to significantly alleviate pain and improve functional outcomes in patients with KOA [19, 20], while concurrently promoting cartilage health. However, in other surgical interventions involving IFP removal aimed at reducing the secretion of harmful cytokines, no substantial improvement in clinical prognosis was observed for KOA patients. This discrepancy may be attributed to the loss of the cushioning effects and nutritional support inherently provided by the IFP [21]. As a result, it is necessary to further define the extent of IFP pathology in KOA to better determine the appropriate scope of resection during surgery.

In this study, we have identified that the pathological alterations in the infrapatellar fat pad (IFP) of KOA patients are anatomically stratified, with the synovial-side IFP tissue showing higher levels of inflammatory and adipose-related factors as well as greater immune cell infiltration. These findings may offer a theoretical foundation for determining the appropriate extent of IFP resection during clinical surgery.

Method

Patient and tissue collection

Sixteen patients who met the American College of Rheumatology (ACR) classification criteria [22] for knee OA were recruited for the study. These patients underwent total knee arthroplasty (TKA) at the First People’s Hospital of Hefei. Written informed consent was obtained from all participants prior to inclusion in the study, which was conducted with approval from the Ethics Committee, following the principles of the Declaration of Helsinki. The following clinical data were collected: age, sex, and Body Mass Index (BMI). We primarily determine the pathological changes of the infrapatellar fat pad (IPFP) through intraoperative visual assessment combined with postoperative histopathological analysis. During the surgical procedure, we excised the IPFP that exhibited significant congestion and edema until normal adipose tissue characterized by a uniform granular yellow appearance was exposed. Thus, we define the inflammatory layer as the ‘visibly vascularized and congested region at the synovial-fat pad junction,’ while the non-inflammatory layer is taken from the central adipose tissue near the patellar tendon, away from the junction. Intraoperative infrapatellar fat pads (IFPs) and adjacent synovial tissues were harvested. Tissue were either fixed in 10% formalin and embedded in paraffin for histochemical analysis or snap-frozen in liquid nitrogen and stored at −80 °C for molecular assays.

Morphological studies

Intraoperative Magnetic Resonance Imaging of the knee joint in partial flexion (patella displaced inferolaterally) were captured to evaluate the pathological morphology of stratified IFP. Specimens from distinct IFP layers were fixed in formalin for 24 h and embedded in paraffin. Section (10 μm) were stained with hematoxylin and eosin (H&E) and analyzed microscopically to assess inflammatory cell infiltration, vascularization, and adipocyte morphology.

Immunohistochemistry (IHC)

Paraffin-embedded sections were deparaffinized in xylene and rehydrated through graded alcohols. Endogenous peroxidase was blocked using 3% hydrogen peroxide at 37 °C for 10 min, followed by three washes with phosphate-buffered saline (PBS). Antigen retrieval was performed by boiling sections in 0.01 M citrate buffer (pH 6.0) for 15–20 min, followed by natural cooling and three additional PBS washes. After blocking with normal sheep serum, primary antibodies were applied and incubated at 4 °C overnight. Negative controls were treated with PBS in place of the primary antibody. After washing, sections were incubated with biotinylated secondary antibodies at 37 °C for 30 min, followed by treatment with a DAB-H2O2 substrate for color development. Hematoxylin counterstaining was performed, and the slides were dehydrated, cleared, and mounted for microscopic observation (Table S2).

Real-time quantitative (qPCR)

Total RNA was extracted from tissues using TRIzol reagent (Invitrogen), and cDNA was synthesized using a high-capacity cDNA reverse transcription kit(ThermoFisher Scientific). qPCR was performed using SYBR Green Master Mix on a 7500-FAST PCR system (Applied Biosystems). The relative expression of the target genes (fold change) were determined by 2-ΔΔCT, where ΔCT = CT(target) – CT(18s), and ΔΔCT = ΔCT(experimental group) - ΔCT(control group). Primer information for the analyzed genes is provided in Table S1.

Sequencing of transcriptome

Total RNA was extracted from tissues using TRIzol (Invitrogen, Carlsbad, California, USA) according to manual instruction.

Appropriate amount of tissue was ground into powder under liquid nitrogen, then transferred into a 2 ml EP tube containing 1.5 ml of TRIzol, standing for 5 min. The mixture was centrifuged at 4 °C at 12,000xg for 5 min, and the supernatant was transferred to a new EP tube for extraction. 300 µl of chloroform/isoamyl alcohol (24:1) was added into the transferred supernatant and the mixture was vigorously vortexed to ensure thorough mixing. Then the mixture was centrifuged at 4 °C and 12,000xg for 8 min. The aqueous supernatant was carefully aspirated and subjected to an additional iteration of the aforementioned extraction protocol. After centrifugation, the final upper aqueous layer was carefully transferred to a new 1.5 ml EP tube with a 2/3 supernatant volume of isopropanol (for low amount of tissues/cell, 2µL 5 mg/mL glycogen can be added for precipitation), gently inverted to mix well, placed in a −20 °C refrigerator for 2 h.

Next, the precipitation mixture was centrifuged with 17500xg at 4 °C for 25 min. The supernatant was discarded, and the precipitation was washed with 0.9 ml of 75% ethanol. The precipitation was suspended by inverting the tube up and down several times. Then the precipitation was collected by centrifuging at 17,500 g for 3 min at 4 °C and discarding the supernatant. Repeat the centrifugation to remove supernatant completely and the precipitation was dried in the biosafety cabinet for 3–5 min. Finally, 20µL ~ 200µL of DEPC-treated or RNase-free water was added to dissolve the RNA. Subsequently, total RNA was qualified and quantified using Agilent 2100.

Library preparation is performed using Optimal Dual-mode mRNA Library Prep Kit (BGI-Shenzhen, China). A certain amount of RNA are denatured at 65–80℃ to open the secondary structure, and mRNA is enriched by oligo (dT) attached magnetic beads at 25℃. After reacting at 94℃/5–8 min for a fixed time period, RNAs are fragmented with fragmentation reagents.

Then First-strand cDNA is generated using random hexamer-primed reverse transcription, followed by a second-strand cDNA synthesis. The synthesized double strand cDNA is subject to end repairment reaction. After cDNA end repairment, a single ‘A’ nucleotide is added to the 3’ ends of the blunt fragments through A tailing reaction. Then the reaction system for adaptor ligation configured to ligate adaptors with the cDNAs, and finally, the library products are amplified through PCR reaction and subjected to quality control.

Next, the single-stranded library products are produced via denaturation. The reaction system for circularization is set up to get the single-stranded cyclized DNA products. Any uncyclized single stranded linear DNA molecules will be digested. The final single strand circularized library is amplified with phi29 and rolling circle amplification (RCA) to make DNA nanoball (DNB) which carries more than 300 copies of the initial single stranded circularized library molecule. The DNBs are loaded into the patterned nanoarray and PE 100/150 bases reads are generated on G400/T7/T10 platform (BGI-Shenzhen, China).

Data filtering

The sequencing data was filtered with SOAPnuke by (1) Removing reads containing sequencing adapter; (2) Removing reads whose low-quality base ratio (base quality less than or equal to 15) is more than 20%; (3) Removing reads whose unknown base (‘N’ base) ratio is more than 5%, afterwards clean reads were obtained and stored in FASTQ format. The subsequent analysis and data mining were performed on Dr. Tom Multi-omics Data mining system.

The clean reads were mapped to the reference genome using HISAT2. After that, Ericscript (v0.5.5) and rMATS (V4.1.2) were used to detect fusion genes and differential splicing genes (DSGs), respectively.

Bowtie2 was applied to align the clean reads to the gene set, in which known and novel, coding and noncoding transcripts were included.

Expression level of gene was calculated by RSEM (v1.3.1). The heatmap was drawn by pheatmap (v1.0.12) according to the gene expression difference in different samples. Essentially, differential expression analysis was performed using the DESeq2(v1.34.0) (or DEGseq(v1.48.0)or PoissonDis)with Q value ≤ 0.05 (or FDR ≤ 0.001).

To take insight to the change of phenotype, GO and KEGG enrichment analysis of annotated different expression gene was performed by Phyper based on Hypergeometric test. The significant levels of terms and pathways were corrected by Q value with a rigorous threshold (Q value ≤ 0.05).

Western blotting

Prepare the required number of centrifuge tubes, place grinding beads in each tube, and set them on an ice box for later use. Wash the extracted IFP tissue blocks of different layers 2–3 times with pre-cooled PBS to remove blood contamination. Then place them on filter paper to absorb excess PBS before transferring them into corresponding homogenization tubes. Add approximately 10 times the sample volume of RIPA lysis buffer. Place the tubes symmetrically in a tissue grinder, cover them, select the program, and start homogenization. After homogenization, place the homogenization tubes on ice and let stand for 30 min to fully lyse the tissue (or use ultrasonic treatment for 10 min). Centrifuge at 12,000 rpm for 10 min at 4 °C. Collect the supernatant, avoiding the upper oil layer and bottom impurities (immediately transfer the supernatant into a new centrifuge tube for storage), using infrared micro-quantitative analyser to determine the concentration of protein; add PBS and 5X sampling buffer diluted to the standard protein concentration After that, boil at 100℃ for 5 min, freeze at −20℃.

Electrophoresis and membrane transfer: according to the molecular weight of the target protein, 12% SDS-PAGE gel electrophoresis was configured. Configure electrophoresis buffer according to the requirements, mix the samples, and then take the samples, 20ul per well, electrophoresis at 100 V for 1h30min; after the electrophoresis process, take out the gel, and then stack the filter paper, PVDF membrane, electrophoresis gel and filter paper according to the direction of positive pole to negative pole of the electrode plate of the transfer membrane in order to remove the air bubbles, and then transfer the membrane at 220 A for 1h30min; after the electrophoresis is completed, the PVDF membrane will be placed in the sealing solution and closed at room temperature for 1 h. The PVDF membrane was put into the sealing solution after the transfer and closed at room temperature for 1 h.

Antibody incubation and detection: the closed PVDF membrane was incubated with the antibodies GAPDH, BAFF and CXCL13 (Table S2) diluted in antibody diluent, incubate at 4℃ overnight; the next day, wash the membrane with TBST for 3 times, transfer the PVDF membrane into 3% BSA diluted secondary antibody (dilution 1:5000) and incubate at room temperature for 1 h; then wash the membrane with TBST for 3 times, and then develop the target bands by chemiluminescence using ECL kit, and take pictures of the images with multifunctional imager. The images were analysed using ImageJ2X to compare the grey value of each protein band.

Cell extraction

Under strict aseptic surgical conditions, infrapatellar fat pad tissue samples were obtained. Immediately after collection, the tissues were transferred into sterile containers pre-filled with pre-cooled phosphate-buffered saline (PBS) supplemented with 1% penicillin-streptomycin (double-antibiotic solution). The samples were then transported to the laboratory as promptly as possible to minimize ex vivo time, thereby preserving cell viability.The collected infrapatellar fat pad tissues were transferred to a sterile petri dish within a laminar flow hood. The tissues were rinsed 3–5 times with pre-cooled PBS, gently agitated during each wash to effectively remove surface blood, debris, and residual tissue fluids. Subsequently, using sterile scissors, the tissues were minced into small fragments approximately 1–2 mm³ in size, ensuring uniformity to facilitate subsequent digestion.The minced tissue fragments were transferred into 50-mL centrifuge tubes, and an appropriate volume of 0.1% type I collagenase solution was added to completely submerge the tissues. The tubes were then placed in a 37 °C shaking incubator set at 120 revolutions per minute (rpm) for 60–90 min. During digestion, the tubes were inverted gently every 15–20 min to ensure thorough digestion. After completion, the digestion process was terminated by adding an equal volume of Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS), leveraging the protein components in FBS to neutralize the collagenase activity.The digested mixture was filtered through a 40-µm cell strainer into a new 50-mL centrifuge tube to remove undigested tissue pieces, yielding a single-cell suspension. The suspension was centrifuged at 1200 rpm for 10 min at 4 °C. The supernatant was discarded, leaving a cell pellet. The cell pellet was resuspended in PBS and centrifuged again under the same conditions. This washing procedure was repeated 2–3 times to eliminate residual collagenase and digestion by-products. Finally, the cells were resuspended in DMEM containing 10% FBS to prepare the cell suspension for subsequent use.

Statistical analysis

Statistical analyses were performed using GraphPad Prism 10 software. We employed paired analysis to leverage the correlation of homologous samples, thereby reducing errors and enhancing statistical power. Specific methods should be selected based on data types, including parametric tests (such as paired t-test) or nonparametric tests (such as Wilcoxon test), with strict adherence to statistical assumptions (such as normality and independence). For normally distributed data, Student’s t-test was used; for non-normally distributed data, non-parametric tests were applied. Categorical variables were analyzed using the chi-square test. Data are presented as mean ± standard deviation, and statistical significance was defined as P < 0.05.

Results

Clinical characteristics of included patients

This study recruited 16 patients with OA who underwent TKA treatment, including 4 males and 12 females (Table 1). The average age of the males was 66.50 years, with an average weight of 70.50 kg and an average BMI of 25.99; The average age of the females was 68.17years, with an average weight of 63.17 kg and an average BMI of 25.62. There were no statistically significant differences between the two groups in terms of weight, age and BMI (P > 0.05).

Table 1.

Clinical data analysis of weight, age, and BMI in OA patients

	Male	Female	P value
Number of patients	4	12
Average weight[kg]	70.50 ± 7.37	63.16 ± 10.33	0.163
Average age[years]	66.50 ± 7.85	68.16 ± 9.82	0.741
Average BMI[kg/m²]	25.99 ± 3.20	25.62 ± 3.00	0.848

Inflammatory gradients existed in the IFP of KOA

IFP tissues were collected from 15 patients(one sample was unavailable due to a technical issue) undergoing total knee arthroplasty (TKA) treatment. Intraoperative direct visualization showed that the inflammatory lesions in the IFP of patients with KOA had a distinct pattern: inflammation was more pronounced near the synovial side, while the area closer to the patellar tendon appeared almost normal (Fig. 1A). H&E staining further indicated that the extent of inflammation and vascularization progressively extended toward the patellar tendon side [23].A measure of joint damage severity, such as vascularization and fibrosis levels increased (Fig. 1B). Hematoxylin and eosin (H&E) staining also revealed more mononuclear cell infiltration on the synovial side. To further investigate morphological differences, we performed VEGF staining and Masson’s trichrome staining. VEGF staining showed a higher degree of VEGF staining on the synovial side, indicating a higher degree of vascularization. Masson’s trichrome staining found that the fibrosis level on the synovial side was also higher (Fig. 1C). Thus, we named IFP tissue adjacent to the synovium as inflammatory layer, while the deeper, non-synovial regions as non-inflammatory layer (Fig. 1D).

Fig. 1.

Morphological and histological features of the infrapatellar fat pad (IFP) in an osteoarthritis (OA) patient. The intraoperative photograph (A) demonstrates the distinct layers of the IFP. Hematoxylin and eosin (HE) staining (B) reveals that the inflammatory layer contains more adipocytes with smaller cell areas compared to the non-inflammatory layer, along with a clearly visible boundary between the two regions. Further analysis (C) shows that the inflammatory layer exhibits increased monocyte infiltration, higher vascular endothelial growth factor (VEGF) expression, and greater fibrosis relative to the non-inflammatory layer. A schematic diagram (D) illustrates the anatomical distribution of these layers within the IFP

Transcriptomic heterogeneity between inflammatory and non-inflammatory layers of IFP in osteoarthritis patients

We selected five KOA patients with normal BMI and performed transcriptome sequencing on the infrapatellar fat pads of OA patients, focusing on both inflammatory and non-inflammatory layers, and made more discoveries. The volcano plot shows that there are differentially expressed genes that are upregulated and downregulated, and most of them are related to immune and inflammatory responses. For this, we grouped the differentially expressed genes. A total of five major groups were divided: adipokines, inflammatory factors, fibrosis, angiogenesis, and pain. And we analyzed the differentially expressed genes in each group. In order to understand the magnitude of the impact of the differentially expressed genes in each group, we created a heatmap of the differentially expressed genes in each group for analysis (Fig. 2). In the KEGG pathway enrichment analysis, we identified several representative pathways of differentially expressed genes, such as IL-17 signaling pathway, PPAR signaling pathway, Rap1 signaling pathway, TNF signaling pathway, and cGMP-PKG signaling pathway. The signaling pathways of these differentially expressed genes will be further investigated in subsequent studies.

Fig. 2.

Transcriptomic profiling of differentially expressed genes (DEGs) between inflammatory and non-inflammatory layers of the infrapatellar fat pad in osteoarthritis (OA) patients. A Left panel: Volcano plot displaying DEGs (Treat vs. Control) with thresholds for statistical significance (|log < sub > 2</sub > FC| >1, adjusted *p*-value < 0.05). Red and blue dots represent significantly upregulated and downregulated genes, respectively. Right panel: Bubble plots of Gene Ontology (GO) enrichment analysis for DEGs, categorized by cellular component (top), biological process (middle), and molecular function (bottom). Key enriched terms include immune response pathways (e.g., MHC class II binding, inflammatory response) and membrane-related processes. B Heatmap of hierarchical clustering analysis for DEGs (Treat vs. Control), with rows representing genes and columns representing samples. Color intensity reflects normalized expression levels (log < sub > 2</sub > TPM). The heatmap highlights distinct transcriptional patterns between layers, with functional annotation for genes related to inflammation, adipokines, fibrosis, vascularization, and pain

qPCR validation of differential gene expression between IFP layers

Adipokines originating from the IFP are key mediators of joint damage and inflammation [16, 24]. To explore the differential expression of inflammatory and adipogenic factors between the inflammatory and non-inflammatory layers of IFP in patients with knee osteoarthritis (KOA), we measured the transcription levels of various inflammatory and adipogenic factors using RT-PCR. As illustrated in Fig. 3, both the transcriptional and protein expression levels of inflammatory and adipose-related factors were markedly elevated in the inflammatory layer of IFP from KOA patients. These results may suggest that the inflammatory layer of IFP in KOA is characterized by a heightened inflammatory state.

Fig. 3.

Quantitative PCR validation of adipokine-, inflammatory factor-, and extracellular matrix-related gene expression in inflammatory versus non-inflammatory layers of the infrapatellar fat pad from osteoarthritis patients. The qPCR analysis demonstrates significant molecular differences between tissue layers: (A) Inflammatory factors (IL-1β, TNF-α, and IL-8) show markedly higher expression in the inflammatory layer; (B) Adipokines (Adiponectin, FABP4, and Leptin) are similarly elevated in the inflammatory layer; while (C) Extracellular matrix components exhibit layer-specific regulation—MMP3 is upregulated but COL2A1 is downregulated in the inflammatory layer compared to the non-inflammatory layer. These results confirm the pro-inflammatory phenotype and metabolic dysregulation of the inflammatory layer

MMP3 and COL2A1 levels serve as indicators of the anabolic and catabolic activity in articular cartilage [24, 25]. To investigate whether the inflammatory layer of the IFP in KOA patients contains molecules linked to joint damage, we assessed the expression levels of MMP3 and COL2A1 in both the inflammatory and non-inflammatory layers of the IFP. The results revealed a statistically significant upregulation of MMP3 at both the transcriptional and protein levels, along with a notable downregulation of COL2A1 in the inflammatory layer of the IFP in KOA patients (Fig. 3). This suggests that the inflammatory layer exhibits higher levels of cartilage-degrading mediators.

Immunohistochemical characterization of distinct IFP stratifications

After conducting transcriptome sequencing, in order to further understand the differences between the inflammatory layer and the non-inflammatory layer, we performed Immunohistochemical staining. Immunohistochemical staining shows both the transcriptional and protein expression levels of inflammatory and adipose-related factors were markedly elevated in the inflammatory layer of IFP from KOA patients. Macrophages, T cells, and B cells are the primary immune cells involved in local infiltration during chronic inflammation [1, 26]. To investigate the immune cell infiltration patterns within the inflammatory layer of the IFP in patients with KOA, we conducted immunohistochemical staining targeting macrophages, T cells, and B cells. The findings revealed a significant enrichment of B cells in the IFP’s inflammatory layer, compared to T cells and macrophages, in KOA patients (Fig. 4).

Fig. 4.

Immunohistochemical characterization of inflammatory and structural markers in the infrapatellar fat pad layers from osteoarthritis patients. Immunohistochemical analysis reveals distinct molecular profiles between tissue layers: (1) Pro-inflammatory cytokines (IL-1β, TNF-α, and IL-8) show significantly stronger staining intensity in the inflammatory layer; (2) Extracellular matrix markers demonstrate layer-specific patterns with elevated MMP3 but reduced COL2A1 expression in the inflammatory layer; (3) Immune cell infiltration analysis shows higher CD19 + B cell presence in the inflammatory layer, while CD3 + T cells and CD68 + macrophages exhibit comparable levels between layers

Subsequently, we carried out immunofluorescence staining for inflammatory factors, adipokines, and related immune cells. The results indicated that the levels of inflammatory factors and adipokines in the inflammatory layer were higher than those in the non-inflammatory layer, and the level of CD19 was also higher than that in the non-inflammatory layer. However, there were no significant differences in the levels of CD3 and CD68 (Fig. 5).

Fig. 5.

Immunofluorescence analysis of adipokines, inflammatory mediators, and immune cell markers in the infrapatellar fat pad layers from osteoarthritis patients. A shows that the levels of the adipokine Leptin and the inflammatory cytokine IL-1β in the inflammatory layer are higher than those in the non-inflammatory layer. B shows that there is no significant difference in the CD68 levels between the inflammatory and non-inflammatory layers. C shows that the CD19 level in the inflammatory layer is higher than that in the non-inflammatory layer. D shows that there is no significant difference in the CD3 levels between the inflammatory and non-inflammatory layers. A-D, immunofluorescence staining. Scale: 20x

BAFF/CXCL13-Mediated B-Cell recruitment in the inflamed infrapatellar fat pad of knee osteoarthritis patients

To investigate the causes of B-cell aggregation and activation in the inflamed layer of infrapatellar fat pad (IFP) in KOA patients, we performed protein extraction on IFP tissues from different layers, followed by Western blot (WB) analysis using B-cell activating factor (BAFF) and B-cell chemokine (CXCL13). Results showed significantly increased expression of BAFF and CXCL13 in the inflamed layer (Fig. 6A). Additionally, we isolated adipose-derived precursor cells from IFP tissues and divided them into two groups: a normal control group and an IL-1β-stimulated group to mimic an inflammatory state. Notably, the expression of BAFF and CXCL13 was significantly upregulated in the inflammation-stimulated group (Fig. 6B). These findings suggest that increased expression of BAFF and chemokine CXCL13 in the inflamed layer of IFP in KOA patients may contribute to B-cell aggregation and activation in this region. Therefore, we hypothesize that the pro-inflammatory microenvironment at the synovial-adipose interface may guide B cells to accumulate on the synovial side through chemokines such as CXCL13.

Fig. 6.

Western blot analysis of B-cell-related factors in infrapatellar fat pad (IFP) tissues and cultured adipose-derived precursor cells. A Protein analysis of IFP tissue layers revealed significantly elevated expression of B-cell activating factor (BAFF) and the chemokine CXCL13 in the inflamed layer compared to the non-inflamed layer. B In vitro experiments using IFP-derived precursor cells demonstrated that IL-1β stimulation markedly increased BAFF and CXCL13 production compared to untreated controls

Discussion

The prevalence of KOA increased globally with expectancy life-time prolonged [1, 27], positioning KOA as a leading cause of significant socioeconomic burden [2]. Previous studies that have highlighted the heterogeneous nature of inflammation in joint tissues affected by KOA [7, 28], where the proximity to the synovium often correlates with heightened inflammatory activity. In this study, we explored the inflammatory microenvironment within the IFP of KOA and its potential contribution to joint damage. The findings demonstrate a distinct inflammatory gradient, with inflammation being more pronounced in the synovial side of the IFP, gradually decreasing towards the patellar tendon. Moreover, the disorder of immune cell infiltration was observed in synovial side in IFP of KOA patients.

The upregulation of inflammatory mediators and adipokines within the inflammatory layer of the IFP is particularly notable, suggesting that this region serves as an active site for the release of pro-inflammatory cytokines and adipose-related factors that may exacerbate joint damage [16, 29, 30]. Previous research has established adipokines as key contributors to cartilage degradation and synovial inflammation [14, 16], and our data further support this notion by demonstrating elevated levels of these mediators in the IFP’s inflammatory layer. This localized inflammatory state likely plays a critical role in the pathogenesis of KOA, reinforcing the hypothesis that the IFP is not merely a passive fat depot but an active participant in the disease process [31, 32]. Additionally, fibrosis is a defining pathological feature of the IFP in KOA, contributing to disease progression by modulating cytokine production and altering biomechanical properties, which disrupts local mechanical stability and fosters sustained joint damage [33]. This study also reveals that the IFP region adjacent to the synovial layer exhibits heightened vascularization and inflammatory cell infiltration compared to the joint capsule layer, along with denser fibrous tissue, underscoring the potential interplay between inflammation and fibrosis. The concurrent microstructural changes in the synovium and IFP suggest a shared pathogenic mechanism that may involve crosstalk between tissues or cells [34, 35]. This observation is in line with previous histological and imaging findings [7, 12, 28]. Additionally, the IFP exhibited more intricate and dense fibrous tissue, further substantiating the link between inflammatory processes and fibrotic responses [36].

The remodeling of the extracellular matrix (ECM) by chondrocytes is a critical event in the progression of KOA, characterized by enhanced catabolic activity and suppressed anabolic pathways [37, 38]. A key molecular signature of this process is the increased expression of MMP3 coupled with the reduced expression of COL2A1 [25]. In this study, we identified a significant upregulation of MMP3 and a downregulation of COL2A1 in the inflammatory layer of the IFP in KOA patients, compared to the non-inflammatory layer. Additionally, in vitro adipocyte culture experiments showed that adipocytes exposed to inflammatory cytokines exhibited similar expression patterns [39], implying that damaged IFP tissue or cells may drive changes in joint structure. Future research on osteoarthritis and potential therapies should account for the role of adipose-driven inflammation in disease progression.

Microtrauma to the joints, which exposes the extracellular matrix and activates the innate immune system, is considered a key factor in the progression of OA [40]. This initial injury sets off an inflammatory cascade, worsening synovitis and accelerating OA development [41]. Current understanding points to the chronic inflammation observed in OA as being driven by the overactivation of the innate immune system, particularly through the dysregulation of macrophage polarization [27], activation of T and B lymphocytes [42, 43]. Recently, research has concentrated on the proportion and subtype distribution of immune cells within the synovium of osteoarthritic joints or peripheral blood in patients with OA [43, 44]. In this study, immunohistochemical analysis demonstrated a markedly higher infiltration of B cells in the inflammatory layer of the IFP tissue near the synovium compared to the non-inflammatory layer closer to the patellar ligament. These results indicate a preferential localization of B cells within the inflammatory areas of the IFP and underscore the significant role B cells may play in orchestrating the inflammatory response in IFP tissue. We hypothesize that activated B cells may activate synovial fibroblasts by secreting pro-inflammatory factors such as TNFα and IL-1ß, leading to the progression of OA [45]. It is also possible that CD40-CD40L interactions may induce inflammation in adipocytes [46].

In contemporary surgical practice, the removal of the IFP during TKA is commonly performed to enhance surgical visibility and facilitate the procedure [47]. Recent animal studies suggest that early resection of the IFP can slow the progression of knee OA, while preserving the IFP may exacerbate OA development [48]. However, some reports indicate that IFP removal may increase the incidence of anterior knee pain [49], reduce patellar tendon blood supply, and heighten the risks of scar formation and subsequent tendon shortening [50, 51]. As such, the decision to remove or retain the IFP during TKA remains a topic of debate. Arthroscopic partial resection of the IFP has proven effective for Hoffa’s disease. Our study revealed significant damage to the IFP in KOA patients, with the damaged regions producing joint-damaging inflammatory and adipokine mediators. Thus, selectively removing only the damaged portions of the IFP during surgery is recommended for KOA patients, as it both alleviates the inflammatory load and preserves the cushioning and nutritional functions of the patellar tendon [19, 20].

Recent studies have emphasized the importance of the complement system in the inflammatory network of osteoarthritis (OA), particularly within the infrapatellar fat pad (IFP). In a recent mouse study [52], demonstrated that adipose-derived leptin and complement factor D (CFD, also known as adipsin) act as co-regulators of OA progression and pain, linking adipose tissue inflammation to joint pathology. Transcriptomic and proteomic profiling revealed elevated complement signaling—including C3, C5, and CFD—in adipose tissues associated with OA, suggesting that the alternative complement pathway contributes to the pro-inflammatory milieu of the IFP. In the current study, although complement factor expression was not directly evaluated at the mRNA or protein level, our findings of immune cell infiltration and cytokine upregulation are consistent with activation of innate immune mechanisms that could involve complement signaling. Complement activation products such as C3a and C5a are known to promote macrophage and fibroblast activation, thereby amplifying local inflammation and potentially contributing to the observed histological changes. Future studies should therefore explore the transcriptional and functional dynamics of complement components in the IFP, particularly how leptin–CFD cross-talk may shape the local immune microenvironment and pain sensitization in knee OA.

There are several limitations to this study that warrant consideration. First, the sample size was relatively small (n = 16), which limits the statistical power and generalizability of the findings. Future studies with larger, multi-center cohorts are needed to confirm the observed inflammatory gradients within the infrapatellar fat pad (IFP). Second, the study cohort predominantly consisted of female patients, reflecting the higher clinical incidence of KOA in women but also introducing potential sex-related bias in immune or adipose responses. Inclusion of a more balanced sex ratio in subsequent studies would improve representativeness. Third, the current analysis did not include obese or severely overweight BMI groups, despite obesity being one of the strongest risk factors for KOA and IFP inflammation. Given that adiposity and metabolic factors profoundly influence cytokine and adipokine signaling, future work should stratify patients according to BMI categories to explore obesity-specific inflammatory mechanisms within the IFP.

Moreover, the present study primarily focused on histological and molecular characterizations of IFP layers, without in-depth exploration of functional immune interactions. While B cell infiltration was observed and BAFF/CXCL13 signaling was identified as a potential recruitment axis, the mechanistic role of B cells in mediating local inflammation or cartilage damage remains insufficiently defined. Future investigations should incorporate flow cytometric phenotyping to determine B cell subsets (e.g., naïve, memory, or plasma cells), as well as co-culture assays with chondrocytes or synovial fibroblasts to elucidate their effector functions. Additionally, employing animal models such as B cell-deficient mice could further validate the causative role of B cells in IFP-driven joint pathology.

Lastly, the absence of healthy control samples limited our ability to compare IFP inflammation in KOA to physiological conditions. Future research should therefore include age- and sex-matched healthy controls to better contextualize the inflammatory gradient. Despite these limitations, our study provides a novel anatomical and molecular perspective on IFP heterogeneity and underscores the need for integrated, multi-factorial approaches that consider sex, obesity, and immune mechanisms in knee osteoarthritis.

Conclusion

In this study, we demonstrated the presence of distinct inflammatory gradients within the infrapatellar fat pad (IFP) of knee osteoarthritis (KOA) patients, with inflammation predominantly localized near the synovial side. Our findings revealed that the inflammatory layer of the IFP exhibited elevated expression of inflammatory mediators and adipokines, such as MMP3 and reduced COL2A1, suggesting an active role in promoting joint damage and cartilage degradation. Furthermore, the differential immune cell infiltration, particularly the enrichment of B cells in the inflammatory layer, highlights the involvement of adaptive immune responses in IFP-driven inflammation. These results underscore the critical role of the IFP in KOA pathology, particularly its inflammatory layer, which may contribute to both local joint inflammation and cartilage catabolism.

Authors’ contributions

Rende Ning and Bizhi Tu conceived the study idea, revised the manuscript, and provided financial support. Shuo Yang and Peizhi Lu collected the data and wrote the initial draft. Zheng Zhu, Ya Li and Shijie Wang contributed to the data collection and analysis. Bixuan Cao, Haoyu Yao and Miaoyang Liang took part in carrying out the cell and histological experiments. All authors approved the final draft of the manuscript. All authors are accountable for all aspects of the work in ensuring related questions’ accuracy or integrity. Any parts of the work are appropriately investigated and resolved. Rende Ning is the guarantor. All authors contribute equally. The corresponding author attests that all listed authors meet author-ship criteria and that no others meeting the criteria have been omitted.

Funding

This study was supported by Grants from Anhui Key Clinical Speciality Construction Project, The Basic and Clinical Cooperative research Promotion Program of the Third Affiliated Hospital of Anhui Medical University (2022sfy007), and Anhui Medical University Foundation (2023xkj109).

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding authors upon reasonable request.

Declarations

Ethics approval and consent to participate

for our study was granted by The Committee on Medical Ethics of The Third Affiliated Hospital of Anhui Medical University (Reference number 2024-292-01). All participants signed an informed consent form.

Competing interests

The authors declare no competing interests.

Conflict of interest

Declarations.