Crosstalk between lymphatic system and synovial inflammatory cells in osteoarthritis: molecular mechanisms and potential cell-based therapies

Abstract

The lymphatic system plays a significant part in interstitial fluid balance and immune surveillance throughout the whole body. In addition, recent studies have elucidated the role of the lymphatic system in maintaining the normal function of joints, and its dysfunction is related to the pathogenesis of osteoarthritis (OA). In this review, we described the composition and function of the lymphatic system in the knee joints and summarized the research progress of lymphatic networks in the development of OA. The synovial lymphatic system (SLS) is the most well-characterized lymphatic network, and we elaborated the pathological changes of SLS which resulted in the disturbance of knee joint homeostasis and the progression of OA. We summarized the underlying molecular mechanisms of lymphatic disorders and focused on the crosstalk between the SLS and synovial inflammatory cells, especially the abnormal macrophage polarization and regulatory T cells dysfunction. Cell-based therapies have been proven to restore lymphatic function. This review proposed potential cell-based therapies to treat OA by targeting the SLS, and mesenchymal stem cells and Tregs based therapies are currently the most promising approaches to treat OA by restoring lymphatic flow.

Introduction

Osteoarthritis (OA) is the most common degenerative joint disease characterized by cartilage destruction, subchondral bone remodeling, osteophyte formation, and synovitis. An estimated 240 million people are affected by symptomatic OA worldwide [1, 2]. There still lacks effective treatments because of the poor understanding of the pathophysiology of OA [3]. Researchers have identified several risk factors of OA, including age, obesity, inflammation, trauma, and genetic factors [4].

The traditional view considers OA to be a wear-and-tear disease. However, current researches suggested that immune abnormalities were the key factors in the pathogenesis of OA, and the main pathological changes were caused by chronic low-grade inflammation [5]. To ameliorate the inflammatory level, the extravasated fluid, inflammatory cytokines, and immune cells would be removed by the lymphatic vessels in human body [6]. And in the synovial fluid (SF) of arthritic joint, the kinetics of inflammatory molecules clearance is also controlled by the lymphatic flow [7]. Lymphatic vessels are widely distributed in the components of knee joints, and the synovial lymphatic system (SLS) plays a major role in the clearance of inflammatory mediators and maintenance of intra-articular homeostasis [8]. Abnormalities of SLS and its draining function have been observed in OA joints, and restoring lymphatic function exhibited therapeutic effects on post-traumatic osteoarthritis (PTOA) and age-related OA in animal models [9–12].

In this review, we introduced the basic knowledge of the lymphatic system and described the distribution and function of the lymphatic vessels in various components of the joints. We concentrated on the SLS because of its significant role in maintaining the homeostasis of the joint. Then we summarized the researches unveiling the molecular mechanisms behind the lymphatic dysfunction in OA. Recent published reviews about lymphatic system and OA focused on lymphatic dysfunction itself and the interactions between lymphatic vessels and blood vessels [8, 13, 14]. Compared with these reviews, this review focused on the interactions between the SLS with synovial inflammatory cells, which may account for the lymphatic dysfunction and aggravated joint inflammation. Cell-based therapies targeting the SLS to restore lymphatic function may be promising strategies for the treatment of OA. All literatures included in this review were searched and acquired from the PubMed Database. We searched the database with the keywords “osteoarthritis” and “lymphatic system”, combined with the components of the joint, such as “cartilage”, “ligament”, or combined with related cell types, such as “macrophages”, “T cells”. Original articles and reviews from 2010 to 2025 were included in this review.

Basic structure and function of the lymphatic system

The lymphatic system consists of lymphatic vessels, organs, and tissues that are widely distributed in the human body. It functions as the executor for immune surveillance, interstitial fluid balance, lipid transportation, and interorgan communication [15]. Lymphatic capillaries and collecting vessels are the constituents of lymphatic vessels. Lymphatic capillaries consist of a thin layer of lymphatic endothelial cells (LECs) whose transcriptional or surface markers include lymphatic vessel endothelial receptor-1 (LYVE-1), podoplanin (PDPN), prospero homobox 1 (PROX1), and vascular endothelial growth factor receptor 3 (VEGFR3). Primary valves prevent the backflow of lymphatic fluid, which are formed by the overlapping of LECs. The initial lymphatic vessels with blind ends are highly permeable due to their button-like cell junctions and discontinuous basement membrane (BM). If the interstitial pressure surpasses the intraluminal pressure, tensioned anchoring filaments open the button-like structures between the LECs to facilitate interstitial fluid into the initial lymphatic vessels to form the lymph [16]. Lymphatic collecting vessels receive lymph from capillaries. Compared with capillaries, the permeability is lower because of the tight zipper-like junctions between LECs [17]. Mature lymphatic vessels are composed of LECs, continuous BM, secondary valves for one-way lymph flow, and 1–2 layers of lymphatic muscle cells (LMCs) providing the dynamical source of lymphatic drainage [18]. LMCs are positively stained with α-smooth muscle actin (α-SMA) for their structural similarities with smooth or straited muscle [19]. Phasic and tonic contractions of LMCs with extrinsic tissue force propel the lymph through afferent lymphatics to the lymph node (LN), where the lymph is further conveyed to the venous circulation [20].

Under certain circumstances, there are precollectors between the collecting vessels and capillaries. Precollectors possess secondary valves to prevent the backflow of lymph. Owing to the deficiency of LMCs in precollectors, the intrinsic pressure in individual segments is the determinant of lymph drainage [21] (Fig. 1).

Fig. 1.

Fundamental structure and function of the lymphatic system. A Lymphatic capillaries are made up of LECs with button-like intercellular junctions and discontinuous BM, which are anchored to the extracellular matrix by anchoring filaments. Increased interstitial fluid pressure can open the junctions between LECs and promote lymph generation. The lymph is transported back to the venous circulation eventually. Sometimes the intermediate component precollector can be detected. B LECs in lymphatic capillaries positively express LYVE-1, VEGFR-3, PDPN, and PROX1 without α-SMA. Lymphatic collecting vessels covered by LMCs are additionally positive for α-SMA, and the expression of LYVE-1 in LECs is relatively lower compared with that in capillaries

Lymphatic system in the knee joint

In the lower limbs, the distal lymphatic vessels drain the ankle joint to the popliteal lymph node (PLN), and the proximal lymphatic vessels drain lymph from the PLN and knee joint to the iliac lymph node [22, 23]. Lymphatic vessels were traditionally thought to be absent in bone, cartilage, cornea, and the central nervous system. However, recent studies have identified their presence in these tissues and organs, which reshapes our understanding of lymphatic networks. The distribution and function of the lymphatic system in joints have not been fully elucidated yet. In this review, we provided an overview of the peri-articular and intra-articular lymphatic system and described their roles in the onset and progression of OA.

Synovium

The synovium contains two layers: the lining layer (intima) and the sublining layer (subintima). The lining layer is mainly comprised of fibroblast-like synoviocytes and macrophage-like synoviocytes, and the relatively acellular sublining layer is composed of connective tissue, including blood vessels, fibroblasts, lymphatic vessels, and few infiltrating cells [24]. The SLS can be visualized with a whole-slide imaging system in arthritic mice, which separates the blood vessels and lymphatic vessels and reveals the difference between the initial and collecting vessels [21]. In recent studies, dysfunction of the SLS has been confirmed to accelerate the progression of OA. In mouse OA models, several studies have implicated the morphological and quantitative changes in synovial lymphatic vessels and corresponding undermined lymphatic function [9–12, 25]. Blocking lymphangiogenesis with VEGFR-3 neutralizing antibodies impaired lymphatic clearance and aggravated OA progression in mice. On the contrary, the progression of mouse OA was alleviated when synovial lymphatic drainage was restored using anti-inflammatory reagents or vascular endothelial growth factor (VEGF)-C [11, 12]. In addition to animal research, clinical research concerning the alternations of the SLS exhibited controversial outcomes. An early study reported that the number of lymphatic vessels increased in the inflammatory OA synovium [26]. However, Shi et al. reported decreased numbers of capillaries and mature lymphatic vessels in OA patients, suggesting impaired function of the SLS [9]. Walsh et al. described that the synovium in OA patients had lower lymphatic vessel density (LVD) and lower LEC fractional areas, and these parameters were independent of the synovial inflammation level [10]. Most of the published studies focused on the quantity and morphology of the SLS in patients with OA instead of investigating the molecular mechanisms underlying the observed changes, which should be the key research contents in the further investigation.

Bone

In the progression of OA, the subchondral bone undergoes microstructural remodeling, including sclerotic changes, cystic lesions, and osteophyte formation. The presence or absence of lymphatic vessels in bone is controversial. Previous investigations only found lymphatic vessels in the periosteum surrounding bone, indicating that the existence of lymphatics in bone was considered as a pathological phenomenon in diseases such as Complex Lymphatic Anomalies [27, 28]. With the development of imaging techniques, initial and collecting lymphatic vessels can be distinguished in the periosteal region, and lymphatic vessels can sometimes be discovered in the cortical bone of mice [25]. Recently, lymphatic vessels were detected both in cortical bone and bone marrow cavity using light-sheet imaging, proving the existence of lymphatics in the bone tissue of human and mouse [29].

Bone and lymphatic vessels are both modulated by homologous cytokines and signaling pathways. The interaction between receptor activator of NF-κB ligand (RANKL) and RANK is essential for both osteoclast differentiation and lymphatic development [30, 31]. Bone morphogenetic protein (BMP) can also mediate the signaling cascades involved in lymphatic development and lymphatic valve homeostasis [32]. Similarly, the lymphangiogenic factor VEGF-C can also function as a paracrine or autocrine cytokine to increase osteoclast activity, while its role in osteoblast reproduction and differentiation is comparatively negligible [33, 34]. Lymphatic vessels in bone can do harm to the bone homeostasis. Lymphatic invasion into the bone caused by the overexpression of VEGF-C can lead to abnormal bone resorption in mice [35]. Secretory factors produced by LECs restrict osteoblast reproduction and osteoblastic differentiation, ultimately resulting in reduced osteogenesis in vitro [36]. On the contrary, VEGF-C can also alleviate inflammatory bone loss through enhancing the lymph flow accompanied by decreased osteoclasts and increased osteoblasts in mice [37]. A reasonable explanation is that promoted lymphatic drainage helps reduce inflammatory cytokines and regulate bone metabolism, contributing to bone repair in complex inflammatory environment. A recent study reported that LECs were capable of releasing CXCL12 to promote the differentiation of myosin heavy chain 11-positive cells into osteoblast lineages and induce bone regeneration in mice [29]. The role of lymphatic vessels in bone is still intriguing until now. Though some studies have suggested the presence of lymphatic vessels in bone, the exact molecular mechanisms behind the pathological bone remodeling require further exploration.

Tendon

Tendon attaches muscle to bone and is made up of collagen fiber fascicles and interfascicular connective tissue. It is involved in the transmission and storage of force, and helps sustain joint flexion and extension, maintaining joint stability [38]. Instability caused by tendon injury is considered as the initiating and progressive factor of OA. The lymphatic flow of tendon has been identified in the last century [39]. With the identification of various lymphatic markers, lymphatic vessels were observed in the injured tendon lesions, and in the peritendineum or musculo-tendineal transition zone of normal tendon in rats [40]. Immune cells participating innate and adaptive immunity were engaged in inflammatory response in Achilles tendon injury or healing process in mice, and may potentially be modulated by the lymphatic system as well [41]. Boosting the lymphatic drainage was proved to mitigate the inflammatory levels of the impaired Achilles tendon, eventually alleviating the generation of heterotopic ossification in mice [42]. The lymphatic intrusion in the injured tendon might be ascribed to VEGF-C released by the mesenchymal progenitor cells and tenocytes, the former of which predominantly facilitated the occurrence of lymphangiogenesis and heterotopic ossification in mice model [43].

Infrapatellar fat pad

The infrapatellar fat pad (IFP) is the intra-articular adipose tissue located underneath the patella, which is extensively innervated by blood vessels and nerve fibers. The IFP is involved in the pathogenesis of OA, while its function is still controversial [44]. It could function as an endocrine organ to secrete various kinds of cytokines and adipokines to accelerate the progression of OA. But the mesenchymal stem cells (MSCs) originated from the IFP exhibited anti-inflammatory and regenerative capabilities to protect the joint from degeneration [45]. The IFP of knee OA patients was featured with increased inflammatory cell infiltration, augmented vascular invasion, and incrassated interlobular septa [46]. The lymphatic system is significant for preserving homeostasis in adipose tissue because of its participation in lipid metabolism and inflammation modulation [47]. Expanded lymphangiogenesis through the VEGF-D signaling pathway was beneficial to the transmigration of immune cells and the metabolism of glucose and lipid in mice [48]. Although previous studies have identified the existence of lymphatic vessels in the IFP and other adipose tissues of the joint, the morphological and functional alternations of lymphatic system in the IFP of patients with OA have not been investigated until now [9, 49].

Ligament

Ligament serves as the stabilizer of the joint and the bridge between two bones. Ligament is made up of collagen structures and non-collagenous components. The abnormal load bearing and pro-inflammatory reactions caused by ligament injury are closely associated with the progression of OA [50]. Approximately 50% of the patients with anterior cruciate ligament injury will progress to OA in the next decade [51]. Lymphatic vessels have been found in the intra-articular ligaments of several animal models [9, 25]. A cadaver study revealed that in the posterior cruciate ligament, the lymphatics were discernible in loose connective tissue intervals of longitudinal collagen fibers, and the distribution of the lymphatics was heterogeneous [52].

Muscle

Muscle weakness and inflammation caused by adipose deposition facilitate the structural and functional alterations of OA. Skeletal muscle can also interact with peri-articular tissues in the progression of OA through the myokines and cytokines secreted by muscle and other inflammatory cells [53]. Lymphatic networks and lymphangiogenic factors are found in the skeletal muscles, and lymphatic vessels have been identified in the perimysial, epimysium, and capillary beds between muscle fibers in mice [54, 55]. In muscular disorders, interstitial fluid and immune cells enter initial lymphatics and transmigrate to LNs. Intramuscular lymphatic vessels lack LMCs, therefore, the lymph drainage is maintained by the extrinsic muscle contractions [52]. An early study reported that LYVE-1 was an effective biomarker for the inflammatory response in the initiation of muscle pathogenesis [56]. Promoting lymphangiogenesis with VEGFR-3 or enhancing lymph flow through exercise-induced muscular contraction exhibited the potential effects to alleviate lymphatic-associated muscular disorders and OA in mice [57–59].

Cartilage

Cartilage is made up of chondrocytes and surrounding extracellular matrix, which reduces the friction during joint movement, diminishes the contact stress, and dissipates the loading energy. Under normal conditions, articular cartilage is absence of blood vessels, nerves, and lymphatic vessels [60], and in pathological circumstances, there remains no evidence of lymphatic vessels invasion into articular cartilage. In the articular cartilage of OA patients, blood vessels and nerves infiltration into the articular cartilage have been demonstrated [61]. Lymphangiogenesis and angiogenesis may be accompanied with each other because these two processes share common growth factors. Lymphatic sprouts were found in murine tracheal cartilage with airway inflammation [62]. Therefore, we speculate that lymphatic vessels may also exist in articular cartilage under certain conditions (Fig. 2).

Fig. 2.

The distribution of lymphatic system in articular tissue and its relationship with OA. A Under physiological circumstances, lymphatic vessels are distributed in nearly all kinds of tissues in the knee joint, excluding articular cartilage. B Tendon injury, ligament injury, pathological changes of infrapatellar fat pad, and muscle inflammation are involved in the initiation and progression of OA. Lymphatic vessels participate in these pathological changes. C Alterations in the subchondral bone of OA joints include sclerosis and the formation of cysts and osteophytes, which are caused by the interactions among osteocytes, osteoblasts, and osteoclasts

Pathological changes of the SLS in OA

The SLS is the most studied lymphatic system in knee joint. The predominant functions of lymphatic vessels in knee joint homeostasis are interstitial fluid balance and immune cell surveillance [63]. Under normal circumstances, intact cartilage is impermeable for metabolic molecules and acts as a physiological barrier. Molecules and cells in SF can penetrate the synovial lining and diffuse into the synovium, extravasating from the joint via blood and lymphatic vessels in the synovium. The trans-synovial clearance of cells and molecules is determined by their own physical properties including molecular weight, steric hindrance, charge, and carrier molecules. In general, molecules of low molecular weight are usually eliminated by blood capillaries in the manner of passive diffusion. On the contrary, high molecular weight molecules and cells leave the joint through synovial initial lymphatics [64]. Immune cells and antigens transmigrate via the lymphatic capillaries towards draining LNs, where the immune response occurs [65]. With the onset and progression of OA, the clearance of solutes from the joint will be affected. Confined solutes clearance may result from increased intra-articular pressure, thicker synovium, and impaired lymphatic function in rat OA joint [66]. Meanwhile, the hydraulic conductance of osteochondral tissue is enhanced with the erosion of cartilage in ex vivo experiments, indicating that subchondral bone may serve as another channel for intra-articular substance clearance [67].

The pathological changes of the SLS in OA consist of alternations in the number of lymphatic vessels, impairment of drainage function, and the increase of lymphatic permeability. As aforementioned, alterations in the number of lymphatic networks in OA synovium remain controversial [10, 26]. This discrepancy may be attributed to the difference in sampling location or patient population. With the development of detecting technology, capillary lymphatic vessels and collecting lymphatic vessels can be detected separately. At the early stage of OA, there are more lymphatic vessels which is predominantly composed of lymphatic capillaries. While at the end stage, the SLS is mainly characterized by the reduction of mature lymphatic vessels. Defective drainage function of the SLS was found at both stages in OA joints [9]. SF isolated from OA patients can alter lymphatic contractile activity both in vitro and ex vivo [68]. The possible mechanism may be that inflammatory cells and catabolic cytokines traveling within the lymphatic vessels can dramatically hinder the function of LECs and LMCs, and nitric oxide (NO) released by immune cells, chondrocytes, and inflamed endothelium may play a significant role [69].

In addition to the lymphatic pumping activity, permeability is another decisive factor for the drainage function of the SLS [6]. Changes in lymphatic permeability may explain the reason why increased lymphangiogenesis of lymphatic capillaries does not promote the drainage function. Primary and collecting lymphatics exhibit inherent permeability matching their physiological functions [70]. Under inflammatory conditions, the barrier integrity of lymphatic networks can be disrupted, and lymphatic vessels express a leaky phenotype [22]. Multiple pro-inflammatory and anti-inflammatory molecules are involved in the progression of OA [71]. Pro-inflammatory cytokines including interleukin-1β (IL-1β), IL-6, tumor necrosis factor-α (TNF-α), and interferon-γ (IFN-γ), were proven to increase lymphatic endothelial permeability through modulating the junctional components and cytoskeletal structures in vitro [72]. Though these changes were not verified in OA patients, we speculated that similar phenomena can be observed in the SLS. Other lymphatic-associated pathologies such as lymphatic valves dysfunction, lymphatic zippering, congestion within the lymphatic system have been identified in rheumatoid arthritis (RA) and other diseases [73]. Whether they could recapitulate in OA is a mystery.

Although researches on the SLS have made some achievements, there is still a long way to go. On the one hand, quantitative biomarkers for lymphatic function are needed in OA patients. Indocyanine green-near infrared (ICG-NIR) lymphatic imaging is a technique to examine the lymphatic function, which has been used to evaluate the function in mouse models of arthritis and joints of large animals [23, 74]. ICG-NIR was also used to assess the lymphatic flow in hands of RA patients [75]. Therefore, it is likely that this imaging technique can also be used to evaluate the lymphatic function of OA patients. On the other hand, the mechanisms underlying lymphatic dysfunction within joint inflammation remain largely unknown. The inhibitory effects of peri-lymphatic inflammatory cells on lymphatic function have attracted increasing attentions [76]. Synovitis is a characteristic of OA, which is accompanied by the infiltration of various kinds of inflammatory cells. Macrophages and T cells are abundant in the infiltrated synovium, while the percentages of other types of inflammatory cells are comparatively low [77]. As a result, we summarized the confirmed or possible interactions between the SLS and synovial inflammatory cells and their roles in the progression of OA (Fig. 3).

Fig. 3.

The structure and function of the SLS in homeostatic and OA joint. A The synovium has two layers. The lining layer is composed of fibroblast-like synoviocytes and macrophage-like synoviocytes, and the sublining layer is mainly made up of lymphatic vessels, blood vessels, and connective tissues. Molecules and cells in the synovial fluid are drained from the joint through blood and lymphatic vessels after penetrating through the synovial lining layer. B Pathological changes of the synovium in OA patients include synovial lining hyperplasia, stromal vascularization, and inflammatory cells infiltration. C The morphology and function of the SLS are impaired in OA. Reduced lymphangiogenesis leads to reduced density of lymphatic capillaries. Various inflammatory cytokines do harm to LMCs and reduce lymph flow. Lymphatic permeability is enhanced because of the loss of cellular adhesions. Extravasate lymph can promote adipose deposition and ultimately contribute to fibrosis

The interactions between the SLS and synovial inflammatory cells

Macrophage

Macrophages can be categorized into three distinct phenotypes, namely unstimulated macrophages (M0), classically activated macrophages (M1), and alternatively activated macrophages (M2) [78]. Synovial macrophages are typically dichotomized into two sources: tissue resident macrophages and blood circulating monocytes [79]. Macrophages promote lymphangiogenesis in two different ways. They can transdifferentiate into LECs and integrate with the lymphatic vessels, and they can also secrete lymphangiogenic cytokines to stimulate the proliferation and sprouting of LECs in a paracrine manner [80].

A subset of M2 macrophages co-expressing lymphatic markers and markers of the myeloid lineage exhibited lymphvasculogenic and lymphangiogenic capabilities in inflammatory lymphatic neovascularization. They are derived from bone marrow monocytes and function as lymphatic endothelial progenitor cells (LEPCs) in vitro [81, 82]. Tissue resident M2 macrophages could also express lymphatic markers. For example, cardiac LYVE-1 + macrophages can promote lymphangiogenesis through the paracrine effects of pro-lymphangiogenic factors [83]. Increased M1/M2 macrophage ratio has been found in human OA synovium, which is associated with OA progression [84]. LYVE-1 + M2 macrophages with anti-inflammatory capabilities have been identified in human OA synovium [85]. In inflammatory microenvironment of the dental pulp, LYVE-1 + macrophages quickly disappeared, while anti-inflammatory signals promoted M2 polarization of macrophages and upregulated the expression of LYVE-1 in vitro [86]. Therefore, we speculate that inflammatory cytokines in the synovium could damage the pro-lymphangiogenic capacity of M2 macrophages, accounting for the diminished lymphangiogenesis and impaired SLS functions.

Aggregating M1 macrophages can impair lymphatic function through secreting inflammatory cytokines, which enhance the expression of inflammatory genes in LECs and induce the apoptosis of LMCs [11]. The interaction between M1 macrophages and the SLS involves exosomes. Exosomes released by M1 macrophages impaired the function of SLS through restraining the proliferation and promoting the apoptosis of LECs. Exosomes generated by M2 macrophages exhibited reparative effects on synovial lymphatic function in rats [87]. To conclude, reprogramming of M1 to M2 macrophages may help restore the function of SLS and alleviate the progression of OA, and LYVE-1 + macrophages may participate in this process.

T cell

T cells are another type of cells that infiltrate into the synovium of OA patients. T cells can be classified into cytotoxic T cells, helper T cells (Th), and regulatory T cells (Tregs) [88]. CD4+, CD8 + T cells, Th1, Th17, and Tregs are the predominant subsets in the synovium involved in the progression of OA [89]. Except for Tregs, these subsets of T cells can secrete various kinds of catabolic cytokines to facilitate the progression of OA, including IL-17, IFN-γ, TNF-α, and tissue inhibitor of metalloproteinase-1 (TIMP-1) [90–92]. Decreased anti-inflammatory response mediated by Tregs is another non-negligible cause for OA [89]. The interaction between T cells and lymphatic vessels has been discovered in recent years. Different T cell subsets regulate lymphatic system function differently. CD4 + and CD8 + T cells suppress lymphatic function via IFN-γ, which inhibits LEC-specific gene expression and lympangiogenesis, impairs lymphatic drainage, and enhances lymphatic permeability by mislocalizing VE-cadherin in mice [93, 94]. Likewise, in vitro experiments found that IL-17 A released by Th17 cells can decrease LEC surface or transcriptional markers through a Traf6-dependent manner and suppress lymangiogenesis [95]. Conversely, LECs in different microenvironment exert different impacts on T cells. LECs support the existence and bioactivation of T cells in homeostatic environment. Under pathological conditions, cytokine-stimulated LECs can result in dysfunctional T cell immune response [96]. In a mouse model of RA, defective autophagy in LECs inhibited the quantities and functions of Tregs and facilitated the transmigration of Th17 cells from LNs to inflamed joints [97]. These phenomena may also be observed in OA because of the similarity between these two types of arthritis.

Improving lymphatic function can alleviate the progression of OA. Many types of pharmacotherapies can restore lymphatic function through modulating the T cells in cellular or animal experiments, such as Fingolimod, CD4 + T cell suppressants, and IL-17 A-neutralizing antibody [95, 98, 99]. The effectiveness of these pharmacotherapies on OA needs further investigation. Adoptively transferring Tregs is a promising immunotherapeutic strategy to refrain T cell response, alleviate inflammation and fibrosis, palliate lymphatic vessel enlargement, and facilitate lymphatic drainage function in mice [100]. Inducing Tregs with specially designed nanoparticles exhibited therapeutic effects in mouse model of OA [101].

Mast cell

Mast cells (MCs) are normal components in the sublining layer of synovium. MCs are classified into mast cell tryptase and chymase phenotype (MCTC) and mast cell tryptase phenotype (MCT). In the synovium of OA patients, MCs exhibit a striking shift in the relative proportion from MCTC dominant to MCT dominant phenotype, accompanied by increased number and degranulation ratio [102, 103]. Once activated, MCs can release pro-inflammatory mediators including histamine, tryptases, cytokines, and chemokines, which are closely and positively related to the pathogenesis of OA [104]. Due to their anatomical proximity with lymphatic vessels, these mediators can influence the lymphatic endothelium and regulate lymphatic function. Histamine can alter lymphatic contractile activity through activating H1 receptor, and induce VE-cadherin disruption and lymphatic vessel hyperpermeability through Ca2 + release-activated Ca2 + channel [105, 106]. Normal functioning MC/histamine/NF-κB axis is essential for the maintenance of lymphatic transportation and barrier function, and can be impaired by ageing-associated fundamental activation of MCs [107].

Genetic elimination or pharmacologic inhibition of MCs are promising strategies to alleviate OA in mice, but the therapeutic effects may not achieve through restoring the lymphatic functions [108]. In RA joint, two subpopulations of MCs with distinct functions were identified in modulating the lymphatic function. The activated peri-lymphatic MCT+/MCPT1+/MCPT4 + MCs exhibited pro-inflammatory effects. Intra-lymphatic MCT+/MCPT1+/MCPT4- MCs played a homeostatic role in lymphatic function, and depletion of MCs resulted in confined lymphatic flow and exacerbated joint destruction in mice [109]. Both the number and degranulation degree of MCs are higher in human OA synovium compared to RA [110, 111]. Therefore, we speculate that the interaction between MCs and the SLS may be more complex in OA.

B cell

B cells only make up a small portion of total inflammatory cells in OA synovium. In earlier studies, clonally expanded B cells and antigen triggered immune response might be involved in the development of OA [112, 113]. Recently, the phenotype and function of B cells in OA synovium were identified. B cells exhibited gene expression patterns similar to those of mature plasma cells, and their activation and maturation occurred in the synovium, which may be the explanation for the relative retention of auto-antibodies in OA joints [114, 115]. The crosstalk between B cells and lymphatic networks is concentrated in LNs. Activated B cells are the source of VEGF-A, which is responsible for the induction of lymphangiogenesis [116]. B cell-specific overexpression of VEGF-A can promote LN lymphangiogenesis and enhance LPS tolerance [117]. However, CD23+/CD21hi B cells can translocate from peripheral follicles to the lymphatic sinus, leading to the obstruction within the lymphatic system and hampering passive lymphatic flow from mouse model of RA joint [118]. Accordingly, we suppose that B cells exhibit pathogenic or protective roles depending on their phenotype. Reciprocally, LECs can modulate the homing of B cells to LNs through the secretion of CXCL13 in mouse [119].

B cell activating factor can enhance the production of VEGF-A and VEGF-C of B cells in murine LNs [120]. It can also be secreted by synovial stromal cells, indicating that this phenomenon can also take place in the synovium [121]. Whether B cells can be the target for the intervention of OA and the detailed mechanisms underlying the interaction between B cells and the SLS remain to be determined in the future [122].

Dendritic cell

Only a small number of dendritic cells (DCs) are found in OA synovium [123]. DCs are categorized into myeloid DCs and plasmacytoid DCs, or immature DCs and mature DCs, depending on their phenotype and function [124]. In OA joint, DCs were mainly originated from circulating inflammatory monocytes and exhibited a pro-inflammatory phenotype [125]. In a rabbit OA model, mature DCs were involved in the pathological process of synovium in early stage of OA [126]. The pro-inflammatory effects of DCs may be mediated by Toll-like receptor [127]. Diverse subtypes of DCs exhibit either stimulating or inhibiting effects on the immune response. Transferring the suppressive oligodeoxynucleotides-induced autologous or tolerogenic DCs can be promising strategies to eliminate OA-related inflammation in mouse models [128, 129].

After maturation, peripheral DCs will mobilize to draining LNs, where they contribute to the activation of T cells and the subsequent immune response [130]. The CCR7-CCL21 interaction between LECs and DCs affected the entry of DCs into lymphatic capillaries and their subsequent transmigration to LNs [131, 132]. DCs also contribute to lymphangiogenesis and could interact with collecting vessels [133]. Depending on the physiological permeability of collecting vessels, surrounding DCs can recognize antigens and facilitate their CCR7-dependent recruitment to the draining LNs [134]. Increased lymphatic leakage from collecting vessels hampered the migration of DCs to LNs and impaired the immune function [135]. A subset of CD11b + DCs co-expressing CCR7 and IFN regulatory factor 4 (IRF4) can modulate lymphatic permeability and restrain the fibrosis of collecting vessels [136].

Neutrophil

Neutrophils (NEUs) are the least abundant inflammatory cells in the synovium of OA patients. NEUs are more abundant in SF compared to the synovium [137]. Various types of proteases, pro-inflammatory mediators, and cytokines produced by NEUs may play a prerequisite role in the deterioration of OA [138, 139]. Ex vivo and in vitro experiments found that NEU elastase could induce the apoptosis of chondrocytes by activating the caspase signaling pathway and facilitate the degradation of collagen in a matrix metalloproteinase dependent manner [140, 141]. The activated immunophenotypic and functional characteristics of SF-derived NEUs from patients with OA were identified in recent study [142].

Activated NEUs may enter lymphatic vessels and subsequently migrate to draining LNs during the metabolic dynamics of the SF. The transmigration of NEUs within lymphatic vessels is modulated by a series of adhesion molecules and chemokines [143]. Emerging evidence indicated that NEUs had the capability to regulate lymphangiogenesis in inflamed tissue and LNs. In inflamed airways, the pro-lymphangiogenic factor VEGF-D was highly expressed in NEUs in accordance with the increased lymphatic vessels [144]. When B cells were depleted, NEUs could compensate for promoting LN lymphangiogenesis by modulating the bioavailability and bioactivity of VEGF-A and the secretion of VEGF-D [145].

Fibroblast

In addition to these immune cells mentioned above, fibroblasts are the major components of synovium and play a significant role in modulating lymphatic function in OA. Single-cell sequencing revealed that synovial fibroblasts are comprised of heterogenous subsets with distinct phenotypes exhibiting beneficial or detrimental effects on the progression of human OA [146, 147]. Fibroblasts in inflammatory status can secrete lymphangiogenic factors including VEGF-A, VEGF-C, and VEGF-D [148, 149]. Their lymphangiogenic effects have been verified in other tissues and organs [150, 151]. Diminished expression of VEGF-C and VEGF-B within synovial fibroblasts of OA patients were discovered, and the synovial COL6A1 + fibroblasts were the predominant cellular source of VEGF-C [12]. The role of fibroblasts in modulating the SLS needs to be investigated and verified in future research (Fig. 4).

Fig. 4.

The crosstalk between the SLS and synovial inflammatory cells. A Macrophage: Increased M1 macrophages can produce pro-inflammatory molecules such as TNF-α and NO, which cause the inflammation of LECs and induce LMCs apoptosis. Reduced and defective M2 macrophages may account for the dimished lymphangiogenesis and impaired lymphatic function. Exosomes may participate in the process of modulating lymphatic function by macrophages. B T cell: The predominant pro-inflammatory T cells in OA synovium include CD4 + T cells, CD8 + T cells, Th1 cells, and Th17 cells. They suppress lymphangiogenesis and enhance lymphatic permeability through secreting IFN-γ and IL-17 A. On the contrary, Tregs can reduce the release of inflammatory factors, alleviate fibrosis, and maintain lymphatic homeostasis. Reciprocally, cytokine-stimulated LECs can inhibit the immune response of T cells, whereas homeostatic LECs can promote their activation and transmigration. C MC: Activated peri-lymphatic MCs can release pro-inflammatory cytokines including histamine and tryptases, which can affect lymphatic contraction and increase lymphatic permeability. Intra-lymphatic MCT+/MCPT1/MCPT4- MCs may help maintain the homeostasis of lymphatic system in synovium. D DC: DCs can promote lymphangiogensis and interact with collecting vessels. The physiological permeability of collecting vessels allows the migration of CCR7 + DCs to LNs. CCR7 and IRF4 co-expressing DCs can regulate the permeability of collecting vessels and inhibit fibrosis. E B cell, NEU, fibroblast: These types of cells can promote lymphangiogensis through releasing VEGFs. The mobilization of B cells and NEUs to LNs is mediated by various kinds of adhesion molecules and chemokines. COL6A1 + fibroblasts may be the major source of VEGF-C in OA synovium

Potential cell-based therapies targeting lymphatic function to treat OA

Promoting lymphangiogenesis and improving lymphatic draining function are feasible treatments for OA. In the last decade, various types of lymphatic-modulating treatments aiming to alleviate the inflammation and restore lymphatic function in OA have been discovered [14]. Based on the intercellular crosstalk discussed above, we summarized cell-based therapies to treat OA by improving lymphatic vascular function. The cell types used for restoring lymphatic function include stem or progenitor cells (LEPCs, embryonic stem cells, induced pluripotent stem cells, and MSCs) and differentiated cells (LECs and Tregs) [152]. MSCs and LECs have attracted most attention in recent years because of their efficiency in lymphatic regeneration and less ethical and immunological controversies.

MSCs have been considered as the hotspot for OA treatment because of their potential to facilitate chondrocyte regeneration and suppress inflammatory response [153, 154]. A phase IIb clinical trial reported that intra-articular injection of MSCs can significantly improve the joint function and relieve pain in patients with OA without adverse effects [155]. Human MSCs can differentiate into chondrocytes or produce matrix components to promote the repair of cartilage after stimulated with chondrogenic induction in vitro [156]. Similarly, MSCs in lymph-inductive environment can obtain lymphatic characteristics and help restore the lymphatic networks [157]. These lymph-inductive MSCs can secrete lymphangiogenic factors to increase LVD and restore lymphatic drainage function by promoting the regeneration of collecting vessels in vitro and in mouse model [158, 159]. Compared with other components of joints, there have been few research on the effects of MSCs on SLS currently. Further studies are needed to explore whether the therapeutic effect of MSCs on OA is achieved through acting on SLS, and to design corresponding therapeutic tools based on these discoveries.

LEPCs are another major cell source for lymphangiogenesis. The heterogenetic origins of LEPCs include veins, hemogenic endothelium, myeloid lineage cells, endothelial progenitor cells, and MSCs [160–164]. LEPCs participate in the formation of lymphatic vessels through differentiating into LECs or secreting growth factors to promote lymphangiogenesis [152]. Recent studies reported that LEPCs isolated from human umbilical cord blood were capable of promoting lymphangiogenesis both in vitro and in vivo [165]. The application of LEPCs is relatively limited because of its poor availability. Improved acquisition, isolation, and purification methods are needed to broaden the application of LEPCs in future.

Tregs can rescue lymphatic dysfunction through immune modulation. Treg-inducing nanoparticles have been proven to alleviate the progression of murine OA, while the researchers did not elucidate their direct effect on SLS [101]. The autologous transplantation of LECs can restore lymph flow and augment the density of lymphatic vessels in rats, while their lymphangiogenic capabilities only existed in a short period of time [166]. Although cell-based therapies have achieved great success in preclinical research, there are still several problems to be solved, including dosage, administration method, and validation of the therapeutic effect, before covert these modalities into clinical practice. The potential cell-based therapies targeting lymphatic function for the treatment of OA are summarized in Table 1.

Table 1.

Potential cell-based therapies targeting lymphatic function to treat OA

Cell type	Source	Mechanism	References
MSC	MSCs isolated from bone marrow and peripheral blood	Attain lymphatic phenotypes and promote revival of lymphatic vessels	[157]
	Adipose tissue	Release lymphangiogenic factors or promote regeneration of lymphatic vessels	[158, 159]
LEPC	Human fetal liver	Differentiate into LECs and participate in lymphatic networks; Produce growth factors to boost lymphangiogenesis	[162]
	Human cord blood		[163, 165]
	Myeloid cells		[164]
	Endothelial progenitor cells		[161]
Treg	Mouse Tregs (Amplified by IL-2/anti-IL-2 antibody complex)	Suppress T cell responses, reduce inflammation and fibrosis, alleviate lymphatic vessel enlargement and promote lymphatic drainage function	[100]
LEC	Human LEC	Enhance lymphatic vessels density and improve lymph flow	[166]

MSC mesenchymal stem cell, LEPC lymphatic endothelial progenitor cell, Treg regulatory T cell, LEC lymphatic endothelial cell

Discussion

In the current exploration of OA pathogenesis and treatment, studies concerning the lymphatic system accounted for only a small portion of the total. For example, the review summarizing the progressions on the mechanisms of OA in 2024 published on Osteoarthritis and Cartilage journal did not mention the lymphatic function, underscoring that this filed remains underexplored and in its infancy [167]. At the in vitro level, well-established protocols exist for isolating and culturing chondrocytes, fibroblast/ macrophage-like synoviocytes, osteoblasts, and osteoclasts. LECs and LMCs, by contrast, are difficult to extract from joint and culture in vitro. In animal models, cartilage, synovium, and subchondral bone can be readily evaluated with histological or immunohistochemical staining, whereas joint-associated lymphatic structures are challenging to visualize, quantify, or evaluate the function. In humans, mature imaging and functional assessments are available for cartilage, synovium, and subchondral bone, yet no comparable systematic tools exist for evaluating joint lymphatic function. Present findings are confined largely to cellular and animal experiments, with only a handful of clinical studies. Translating these observations into mature therapeutic strategies remains a distant goal.

However, these challenges do not diminish the importance of this field. Recent animal studies and early-stage clinical trials have already yielded encouraging results. In this review, we provided a comprehensive summary of how the lymphatic system influences OA and detail the underlying mechanisms. Among these complicated mechanisms, we found that abnormal macrophage polarization and Tregs depletion were investigated in depth and may be the main reason. Therefore, future laboratory research should focus on further discovering the underlying signaling pathways and molecular mechanisms modulating these interactions. Advanced molecular biological techniques such as single-cell RNA sequencing and spatial single cell transcriptome may help achieve these aims. Designing pharmacological or cell-based therapies according to these therapeutic targets may become the promising treatment for OA. In addition, there is no established clinical methods for assessing joint lymphatic function. The imaging techniques used to evaluate lymphatic function in animal models should be adapted for clinical use in future, which could facilitate early detection of lymphatic dysfunction and early targeted treatment.

Conclusion

The lymphatic system participates in maintaining joint homeostasis through modulating interstitial fluid balance and immune response. Dysfunction of the SLS contributes to the progression of OA through impairing the drainage of SF and aggravating intra-articular inflammation. The interactions between the SLS and synovial inflammatory cells further explain the underlying mechanisms. Cell-based therapies to restore lymphatic function exhibited more effective therapeutic effects compared to traditional pharmacological treatments, but further modifications are needed to strengthen their effectiveness and adapted to clinical applications. Optimizing bioavailability, safety, and delivery strategies of these cell-based therapies may take another five to ten years. If all goes well, the first approved cell-based therapy to alleviate OA through restoring lymphatic function may emerge by the mid-2030s.

Abbreviations

OA

Osteoarthritis

SLS

Synovial lymphatic system

SF

Synovial fluid

LEC

Lymphatic endothelial cell

LYVE-1

Lymphatic vessel endothelial receptor-1

PDPN

Podoplanin

PROX1

Prospero homobox 1

VEGF

Vascular endothelial growth factor

LMC

Lymphatic muscle cells

LN

Lymph node

ICG-NIR

Indocyanine green-near infrared

LEPC

Lymphatic endothelial progenitor cell

Tregs

Regulatory T cells

MC

Mast cell

DC

Dendritic cell

NEU

Neutrophil

MSC

Mesenchymal stem cell

Authors’ contributions

Wang Zeng: Conceptualization, Methodology, Validation, Investigation, Resources, Writing - Original Draft; Jiangyu Xiang: Validation, Investigation, Visualization; Yang Liu: Validation, Investigation, Funding acquisition; Shirong Chen: Conceptualization, Investigation, Writing - Review & Editing, Supervision; Hao Wang: Conceptualization, Methodology, Validation, Resources, Writing - Original Draft, Funding acquisition.

Funding

This work was supported by Natural Science Foundation of Chongqing (Grant No.CSTB2023NSCQ-BHX0096) and (Grant No.CSTB2022NSCQ-MSX0052).

Data availability

All data generated or analysed during this study are included in this published article.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declared no conflicts of interest relevant to this article.