Perspective: Actin-bundling Protein, Fascin-1, as a Target of Osteoarthritis Progression

Osteoarthritis: A link with actin reorganization

Osteoarthritis (OA) is a degenerative joint disease characterized by progressive articular cartilage degeneration. In end-stage OA, cartilage is completely degraded, leading to bone-on-bone contact, loss of tissue function, and pain that often necessitates total joint replacement. The cellular mechanisms that contribute to OA pathogenesis are not entirely known. A greater understanding of the cellular mechanisms of disease pathogenesis may lead to novel therapeutics that could slow OA progression and prevent the need for joint replacement.

Evidence suggests that actin reorganization is critical to OA progression. In animal models of post-traumatic OA (PTOA), RNA sequencing data has determined that actin cytoskeleton-associated molecules are the class of molecules most altered in PTOA(Gardiner et al., 2015). Within the past few years, several papers have sought to further elucidate the role actin could play in PTOA with the examination of critical molecules involved in actin network regulation (Chan et al., 2023; Lee et al., 2021; Yang et al., 2024). A recent study by Yang et al. (2024) published in “Bone Research” provides new insight that targeting the actin-binding protein, fascin (FSCN), may aid in preventing chondrocyte dedifferentiation in OA, thus providing a new potential therapeutic target against disease progression. To gain additional perspective on determining FSCN’s role in OA, we corresponded with study authors, Drs. Kai Li and Xiaochun Bai.

Alterations in actin organization drive dedifferentiation

The focus of research in the Li and Bai groups is on understanding the mechanistic underpinnings regulating OA pathogenesis to develop therapeutic targets. They aim to understand how chondrocytes are altered in OA progression and the mechanisms by which diseased chondrocytes undergo dedifferentiation, senescence, and apoptosis. The major aim of their study in “Bone Research” is to identify key molecules implicit in chondrocyte dedifferentiation that occurs in OA, which led them to study FSCN.

In OA, cartilage degeneration has been associated with chondrocyte dedifferentiation (Charlier et al., 2019; Schulze-Tanzil, 2009). Dedifferentiated chondrocytes contribute to the development of OA by altering the cellular phenotype and disrupting cartilage homeostasis. In chondrocyte dedifferentiation, chondrocytes reduce the expression of cartilage-specific extracellular matrix components (Type 2 collagen (Col2) and aggrecan) and increase the expression of fibroblast matrix (Type 1 and 3 collagens (Col1, Col2)) expression. This results in the production of fibrocartilage, which is biomechanically inferior to articular cartilage (Aigner, Bertling, Stoss, Weseloh, & von der Mark, 1993; Mgrditchian et al., 2023; Roberts et al., 2003; Yang et al., 2024). Also, dedifferentiated cells have been reported to elevate the production of catabolic enzymes such as matrix metalloproteinases (MMPs), specifically MMP-3 and MMP-13, which degrade the ECM and exacerbate OA progression (Charlier et al., 2019; Sliogeryte, Botto, Lee, & Knight, 2016).

Limited in vivo studies have visualized alterations to the actin cytoskeleton during dedifferentiation in OA. Much of the understanding of actin reorganization during dedifferentiation has been derived from in vitro studies, whereby chondrocytes are cultured in monolayer on stiff polystyrene/glass. Similar to OA, in culture, chondrocytes dedifferentiate and produce fibrocartilage at the expense of cartilage matrix expression (Abbott & Holtzer, 1966; Nameroff & Holtzer, 1967; Parreno, Nabavi Niaki, et al., 2017). This dedifferentiation in culture is enhanced by inflammatory cytokines (Behrendt et al., 2016; Guilak, Nims, Dicks, Wu, & Meulenbelt, 2018; Loeser, Goldring, Scanzello, & Goldring, 2012). Cytoskeletal changes are concomitant with changes to molecular expression during in vitro dedifferentiation, including an increase in actin expression as well as elevated actin polymerization status (low G/F-actin) (Parreno, Nabavi Niaki, et al., 2017). In this process, cortical F-actin in chondrocytes reorganize to form F-actin stress fibers (Benya, Padilla, & Nimni, 1978; Mallein-Gerin, Garrone, & van der Rest, 1991; Parreno, Nabavi Niaki, et al., 2017). The formation of stress fibers by chondrocytes regulates the aforementioned dedifferentiated gene changes.

FSCN as a critical regulator of chondrocyte dedifferentiation in OA

The Li and Bai group was motivated to study key actin-associated molecules involved in dedifferentiation in OA due to the parallels between in vitro and in vivo chondrocyte dedifferentiation. Mass spectrometry data revealed that actin-associated FSCN1 was the most upregulated gene in damaged human OA cartilage samples. They also found that FSCN1 was upregulated in dedifferentiated, cultured chondrocytes.

To further examine the role of FSCN1, the group used an in vivo surgical OA model - destabilization of the medial meniscus (DMM). In DMM, the medial meniscotibial ligament, which anchors the medial meniscus to the tibial plateau, is transected. This destabilization causes the medial meniscus to shift, increasing localized mechanical stress on the knee joint. Over time, this results in cartilage degradation (Glasson, Blanchet, & Morris, 2007). Similar to mass spectrometry of human OA chondrocytes and dedifferentiated passaged cells, the authors found that DMM surgery caused an upregulation of FSCN1. FSCN1 accumulation was localized to the superficial zone of the cartilage, corresponding to regions where dedifferentiated chondrocytes were observed. Thus, the accumulation of FSCN1 correlated with reduced Col2 expression and increased Col1 and -3 expression, as well as catabolic markers MMP13 expression in chondrocytes (Yang et al., 2024).

OA severity correlates with FSCN1 expression

To investigate the function of FSCN1 during OA, the group modified the expression of FSCN1 in vivo in conjunction with DMM-induced OA. They found that overexpression of FSCN1, using an adenoviral vector, exacerbated OA progression following DMM as exemplified by increased cartilage damage, higher degenerative (i.e., OARSI) scores, elevated MMP-3 and -13, and increased dedifferentiation marked by the upregulation of COL-1 and -3. In contrast, when the authors induced knockout of FSCN1 in cartilage, the degradation following DMM was repressed; cartilage proteoglycan content and COL2 expression were preserved, with reduced COL3, MMP3, and MMP13 staining. The expression of FSCN1 positively correlates with dedifferentiation and OA severity, and low levels of FSCN1 are chondroprotective.

Similar to the effects of FSCN KO, inhibiting FSCN in cartilage by using intra-articular injection of FSCN inhibitor, NP-G2-044, suppressed the loss of proteoglycans, COL2 and reduced the proportion of MMP-3 and MMP-13 positive cells (Yang et al., 2024). There was also a reduced presence of fibrocartilage as indicated by COL3 staining.

The regulation of dedifferentiation in OA by targeting F-actin

It is believed that FSCN regulates OA pathogenesis through modulating actin organization. To characterize the effects of FSCN in regulating OA pathogenesis, the investigators complemented in vivo studies with mechanistic in vitro cell culture studies. They demonstrated that treatment with ILβ increased the amount of F-actin stress fibers and decreased G/F actin (Yang et al., 2024). These F-actin changes correlate with a reduction in Col2 and an increase in Col3. However, targeting FSCN using shRNA or inhibition, they determined that loss of inhibition of FSCN mitigated this response by reducing stress fiber and F-actin, leading to elevated levels of Col2 and reduced Col3 as well as reduced MMP-3 and -13(Yang et al., 2024).

The downstream mechanisms regulating gene expression by FSCN-mediated F-actin reorganization and/or polymerization have yet to be completely determined. They demonstrated that actin signaling can intersect with TGFβ signaling. However, there may be even more direct links. In our correspondence, Li and Bai commented on actin as a potential physical transducer of forces connecting the extracellular environment to the nucleus (Ingber, 2003; Luo, Xu, Lele, Kumar, & Ingber, 2008). In support of this in chondrocytes, classical work by Guilak has shown that the nuclear deformation in response to compression is repressed if F-actin is perturbed using cytochalasin D (Guilak, 1995). One possibility is that IL1β increases FSCN, causing an increase in F-actin bundles that attach to the nucleus, which may amplify nuclear deformation in response to cells’ loading. This suggests that the interplay between biomechanical and biochemical forces may converge on the actin cytoskeleton. However, Langelier et al. have contended the existence of actin hardwire to the nucleus as there was no evidence for this in cartilage cross sections (Langelier, Suetterlin, Hoemann, Aebi, & Buschmann, 2000). Nevertheless, in other cell types (i.e., epithelial cells), nuclear invaginations and tunnels encasing the actin have been described (Jorgens et al., 2017). Further investigation into the existence of actin-based connections in the chondrocyte nucleus needs to be revisited.

Beyond direct physical F-actin connections, FSCN-mediated effects on actin polymerization could regulate gene expression in OA. In chondrocytes, actin reorganization has been shown to regulate transcription through actin-associated transcription factors, such as myocardin-related transcription factor-a (MRTF) and Yes-associated protein and transcriptional co-activator with PDZ-binding motif (YAP/TAZ). In primary chondrocytes, YAP/TAZ has been shown to regulate the chondrogenic expression of SOX9 and ACAN (Goto et al., 2018; Hallstrom et al., 2023). However, as cells dedifferentiate, YAP-TAZ expression is lost while the expression of MRTF is increased(Parreno, Raju, Wu, & Kandel, 2017). This coincides with enhanced fibrocartilage expression (i.e., COL1)(Parreno et al., 2014). FSCN may mediate the expression of fibrocartilage genes through MRTF through its effects on actin polymerization. MRTF contains a G-actin-binding motif and also functions as a co-activator for the transcription factor, serum response factor (SRF) (Kuwahara, Barrientos, Pipes, Li, & Olson, 2005; Miranda, Lichner, Szaszi, & Kapus, 2021; Mouilleron, Guettler, Langer, Treisman, & McDonald, 2008; Parreno et al., 2014). We suspect that as G-actin polymerizes to F-actin, FSCN assembles F-actin into bundles, forming stable stress fibers. In this process, MRTF is liberated from G-actin and enters the nucleus (Figure 1). Within the nucleus, SRF interacts with the CArG box sequence within the serum response element in the promoter regions of fibrocartilage genes(Huet et al., 2005). While FSCN inhibition or knockout has been shown to increase G/F-actin, whether MRTF is altered remains to be elucidated. Nevertheless, FSCN may positively regulate MRTF to regulate fibrocartilage matrix gene expression.

Figure 1. FSCN may target chondrocyte dedifferentiation in OA through several actin-based signaling mechanisms.

The polymerization of G-actin to F-actin and subsequent stress fiber formatting facilitated by FCSN1 may liberate G-actin binding transcription factor, MRTF, resulting in MRTF nuclear translocation. An increase in nuclear MRTF upregulates fibrocartilage mRNA levels. The downstream mechanisms causing a reduction of chondrogenic expression and upregulation of MMPs by actin polymerization in OA remains to be determined.

Actin may use several other downstream mechanisms to regulate gene expression. F-actin signaling may intersect with other signaling pathways (i.e., TGFβ signaling), directly connect to the nucleus, or alter gene expression by binding to transcription factors. Actin polymerization has been shown to regulate chondrogenic and fibrocartilage genes via YAP/TAZ or MRTF, respectively. The mechanisms by which actin regulates matrix-degrading enzyme expression (i.e., MMPs) remain to be determined. In other cell types, however, actin associates with the transcription factor, Cysteine-Rich Protein 2 (CRP2). CRP2 has been shown to regulate the expression of MMP-9 and MMP-13, which are crucial for extracellular matrix remodeling and cancer cell invasion in breast cancer cells (Hallstrom et al., 2023; Li, Yang, Qin, & Yang, 2021; Lu et al., 2024; Mgrditchian et al., 2023; Zhao et al., 2022). Further studies should aim to investigate if such pathways are used in chondrocytes during OA. Furthermore, much of the work has been on isolated cells that do not retain the primary chondrocyte phenotype. Further studies are also required to examine actin-based signaling pathways in native cartilage, in vivo, and in the context of OA.

FSCN1 inhibition as a developing therapeutic target against OA

Targeting the actin cytoskeleton to prevent OA has remained largely untapped. However, it is regarded as a targetable molecule against other diseases. The FSCN inhibitor NP-G2-044 may serve as a promising therapeutic and is currently in clinical trials for various cancers (lung, urologic) and ocular diseases (ocular neovascularization) as well as for tumor metastasis (Bai et al., 2023; Frank Yung-Chin Tsai, 2024; Wang et al., 2021; N. Zhang et al., 2023; Z. H. Zhang et al., 2024). Thus, FSCN inhibitors are already FDA-approved and Bai and Li are hopeful that FSCN inhibitors may also advance into human clinical trials for OA.

While there is potential for FSCN-targeted therapies against OA, Bai and Li are cognizant that a greater mechanistic understanding of FSCN in OA is required. There is interest in determining the relationship between actin and cartilage metabolism during OA. The authors are also interested in assessing the contribution of FSCN to different stages of cartilage degeneration in OA. Gleaning the temporal regulation of actin reorganization by FSCN during OA will provide insight into whether FSCN-based therapies are preventive and/or regenerative. While Bai and Li speculate that FSCN may play a limited role in late-stage OA, targeting FSCN1 could help prevent OA progression. Under consideration is how FSCN1-based therapeutics would be administered. Would repeated intra-articular injections be required? Also, since loss of F-actin can lead to cell death (Celeste Morley, Sun, & Bierer, 2003; Desouza-Armstrong, Gunning, & Stehn, 2017), is there a Goldilocks dosage or ways to mitigate any deleterious effects of actin perturbation?

In addition to PTOA, there have been recent reports on the effect of FSCN1 on fibroblast-like synoviocytes in rheumatoid arthritis (Chen et al., 2024). Thus, the Bai and Li group feel that targeting FSCN1 could also be beneficial in mitigating RA pathogenesis (Chen et al., 2024; Liu et al., 2021). Actin reorganization into stress fibers also contributes to disease pathogenesis in other tissue types (Inguito et al., 2022; Parreno, Amadeo, Kwon, & Fowler, 2020). A greater understanding on the regulation of FSCN in these tissue types may prove that targeting FSCN may be broadly applicable.

Actin plays a critical role in regulating chondrocyte phenotype. In addition to FSCN, other actin-regulatory molecules may serve as potential therapeutic targets in PTOA. Other actin regulators may regulate chondrocyte dedifferentiation including RhoGTPases, Tropomyosins, Transgelin, Adseverin, and Filamin, which could serve as alternate candidates for intervention. (Chan et al., 2023; Chan, Parreno, Glogauer, Wang, & Kandel, 2019; Schofield et al., 2023; Zheng, Baek, Karsenty, & Justice, 2007). Intriguingly, inhibition of the RhoGTPase, CDC42, or the Tropomyosins (TPMs), TPM3.1, is a means to suppress dedifferentiation in cultured chondrocytes. However, the effects of these factors need to be investigated in an in vivo OA model. In vivo, it was shown that loss of barbed end capping and severing molecule, adseverin, enhances late-stage PTOA (Chan et al., 2023). Whether adseverin overexpression is chondroprotective and feasible to target in humans remains to be determined. Indeed, the growing understanding on the regulation of chondrocyte phenotype in healthy and disease states provides new insights into potential targets against OA. The future holds promise for actin-based therapeutics against OA.