Regulating lipid metabolism in osteoarthritis: a complex area with important future therapeutic potential

Abstract

Background

Osteoarthritis (OA), which is characterized by pain, inflammation and pathological changes, is associated with abnormal lipid metabolism. Extensive studies have been conducted on the potential functions of lipids including cholesterol, fatty acids (FAs) and adipokines.

Materials and Methods

By searching and screening the literature included in the PubMed and Web of Science databases from 1 January 2019 to 1 January 2024, providing an overview of research conducted on lipid metabolism and OA in the last 5 years.

Results

In addition to adiponectin, several studies on the effects of lipid metabolism on OA have been consistent and complementary. Total cholesterol, triglycerides, low-density lipoprotein cholesterol, adipsin, leptin, resistin, saturated FAs, monounsaturated FAs, FA-binding protein 4 and the ratios of the FAs hexadecenoylcarnitine (C16:1) to dodecanoylcarnitine and C16:1 to tetradecanoylcarnitine induced mostly deleterious effects, whereas high-density lipoprotein cholesterol and apolipoprotein A/B/D had a positive impact on the health of joints. The situation for polyunsaturated FAs may be more complicated, as omega-3 increases the genetic susceptibility to OA, whereas omega-6 does the opposite.

Conclusion:

Alterations in lipid or adipokine levels and the resulting pathological changes in cartilage and other tissues (such as bone and synovium) ultimately affect joint pain, inflammation and cartilage degradation. Lipid or adipokine regulation has potential as a future direction for the treatment of OA, this potential avenue of OA treatment requires high-quality randomized controlled trials of combined lipid regulation therapy, and more in-depth in vivo and in vitro studies to confirm the underlying mechanism.

1. Introduction

Osteoarthritis (OA) is the most prevalent type of arthritis, affecting one in three individuals aged > 65 years, and is a major cause of disability in older persons. Over the past few decades, OA has become a major global public health concern. Moreover, it is estimated that 428.00 of 100, 000 people will develop OA by 2030 [1]. Obesity is a risk factor for OA and an increase in obesity is linked to an increased incidence of OA. Spending on OA accounts for an estimated 1–2.5% of the gross national product of the United States, Canada, the United Kingdom, France and Australia [2]. OA results in significant financial and medical costs, indirect expenditures due to early retirement or job loss [3], and the psychological burden of chronic pain [4]. Total joint replacement (TJR) surgery is considered the most successful treatment for individuals with severe osteoarthritis; however, up to one-third of patients experience unfavourable long-term postoperative pain [5]. The vast majority of medical and healthcare costs are related to the revision of complications from TJR [6]. Therefore, strategies to prevent the development or slow the progression of OA and reduce the need for TJR are urgently required. To achieve this, potential risk factors for OA must be identified to allow early intervention for individuals at a high risk of OA and provide possible therapeutic targets.

As dyslipidemia is highly associated with several prevalent and serious conditions, it is regarded as an important indicator of human health. Reduced levels of circulating high-density lipoprotein cholesterol (HDL-C) and/or low-density lipoprotein cholesterol (LDL-C), along with elevated levels of circulating triglycerides (TG), total cholesterol (TC), chylomicrons and very-low-density lipoprotein cholesterol (VLDL-C) are hallmarks of dyslipidemia caused by abnormal lipid metabolism [7]. Several reasons have been suggested to explain the annual increase in the incidence of hyperlipidemia. However, diets high in fat, particularly those high in cholesterol and animal fat, are considered the primary cause [8]. High LDL-C levels were the fifth most important risk factor for death in the Global Burden of Disease Study from 1990 to 2017 [9] and were associated with 4.3 million deaths in 2017 [10]. A Survey and Mendelian randomization study showed that lipid biomarkers are important markers that threaten the health of patients with OA [11].

Lipoproteins are specialized transport particles that carry lipids in the blood, and the two primary forms of lipoproteins seen in plasma are low-density apolipoprotein (Apo) B-containing lipoproteins and non-apolipoprotein B-containing high-density lipoproteins (HDLs). Lipid homeostasis involves the absorption, production, metabolism and excretion of cholesterol. Lipid metabolism occurs via two main pathways: endogenous and exogenous. In the endogenous pathway, cholesterol is synthesized by the liver and binds to lipoproteins for transport through the bloodstream as VLDL-C. In the exogenous pathway, dietary fat is absorbed by the intestinal mucosa to create chylomicrons. VLDL-C and chylomicrons are broken down into fatty acids (FAs) in tissues by lipoprotein lipase, which are either used in the muscles for energy or stored in fat cells [12]. Reverse cholesterol transport carries the remaining chylomicrons back to the liver for processing and excretion by HDLs or low-density lipoproteins (LDLs). The involvement of hepatic TG lipase, cell surface syndrome (a proteoglycan), LDL receptor (LDLR) and LDLR-related protein (LRP) in this process is not fully understood [13]. In the liver, cholesterol is broken down into bile acids, with the bile either recycled in the intestine or excreted in faeces [14]. Therefore, improving hepatic cholesterol uptake and removal by increasing reverse cholesterol transport should alleviate hyperlipidemia.

Obesity is caused by adipocyte proliferation and hypertrophy when lipid homeostasis is compromised [15]. Obesity-related metabolic illnesses are caused by adipose tissue malfunction resulting from endoplasmic reticulum (ER) stress, mitochondrial dysfunction and alterations in adipose tissue composition [16]. Furthermore, high FA and cholesterol levels increase the risk of oxidation of unsaturated fats, making them more susceptible to attack by reactive oxygen species or polyunsaturated FA (PUFA) free radicals, causing inflammation [17]. Obesity is associated with macrophage infiltration into adipose tissue and the subsequent skewing of the phenotype from anti-inflammatory M2 to pro-inflammatory M1 [18]. Adipokines released by macrophages and monocytes (resistin) [19] and adipose tissue (leptin, chemokines and the inflammatory adipokines tumour necrosis factor-α [TNF-α] and interleukin [IL]-6) [20] contribute to dyslipidemia. In contrast, adiponectin, visfatin and secreted frizzled-related protein 5 are examples of anti-inflammatory adipokines linked to elevated plasma HDL-C levels, decreased ApoAI catabolism, and lowered plasma TG levels [21–23].

Numerous adipokines have been identified since leptin, the first adipokine discovered in 1994, affects inflammation through various endocrine, paracrine and/or autocrine pathways. They are crucial in the pathogenesis of OA [24]. Changes in lipid metabolism, abnormal cholesterol metabolism, and lipid accumulation in chondrocytes can lead to OA [25–27].

Several studies have examined the roles of leptin, adiponectin and other factors in OA and have demonstrated a close relationship between lipid metabolism and OA. However, no study has provided a detailed summary of the relevant data on this relationship. Thus, this review provides a comprehensive analysis of clinical studies, animal models and cellular processes of important lipid and adipokine alterations in the serum, synovial fluid (SF), cartilage, synovial tissue and cells. This study summarizes alterations in the levels of lipids or adipokines in patients with OA and provides an additional summary of in vitro and in vivo research that examined the impact of lipids or adipokines on OA and potential molecular pathways.

2. Materials and methods

2.1. Aim of the review

This study aimed to review studies published in the last 5 years on the performance and possible role of lipid metabolism in the development of OA. The potential effects of lipids and adipokines on OA, as well as possible therapeutic interventions, were considered, with evidence obtained from clinical, animal and cellular studies.

2.2. Literature search

The following search phrases were used to search the PubMed and Web of Science databases: ‘osteoarthritis’, ‘hyperlipidemia’, ‘lipid metabolism’, ‘triglyceride’, ‘total cholesterol’, ‘dyslipidemia’, ‘low-density lipoprotein cholesterol’, ‘high-density lipoprotein cholesterol’, ‘adiponectin’, ‘resistin’, ‘leptin’, ‘visfatin’, ‘lipoprotein’, ‘fatty acid’ and ‘bile acid’. Only complete English-language articles from 1 January 2019, to 1 January 2024, were included in the search.

2.3. Inclusion and exclusion criteria

The inclusion criteria were as follows: (1) studies published in peer-reviewed journals evaluating the role of lipid metabolism in OA; (2) case-control, cellular and animal studies measuring lipid or adipokine levels in samples from patients with OA (plasma, serum, SF, or cartilage) and healthy controls; and (3) studies published in English.

The exclusion criteria were as follows: (1) non-original papers, such as abstracts and letters to editors; (2) articles not written in English; (3) duplicate studies; (4) reviews and veterinary research on dogs, cats and horses; and (5) articles with insufficient data. Using the inclusion and exclusion criteria, two writers (C-XL and D-X) screened the titles, abstracts and full texts. Conflicts were resolved through conversations and advice from the third author (L-J). This study did not require permission from the institutional review board because it was a review article.

The study protocol was not registered prospectively. The initial search yielded 426 relevant articles. After further screening of the titles and abstracts and removing duplicate publications, 187 studies remained, the full texts of which were reviewed. Of these, 62 articles were excluded because they did not meet the inclusion criteria and 12 additional duplicate articles were identified and removed. Therefore, 113 articles were included in the current review, of which 43 were clinical studies, 40 were animal studies and 36 were cell studies, strictly following the literature screening of the PRISMA guidelines[28] (Figure 1).

Figure 1.

Flow diagram of the study selection process. *Consider, if feasible to do so, reporting the number of records identified from each database or register searched (rather than the total number across all databases/registers); **if automation tools were used, indicate how many records were excluded by a human and how many were excluded by automation tools

3. Current knowledge of lipid metabolism in OA

3.1. Evidence from clinical studies

Numerous studies have investigated the epidemiological association between lipid metabolism and OA; these studies are summarized in Table 1. Elevated TG, TC and LDL-C levels increased the risk of OA [29–31]. High serum TC and LDL levels are associated with increased pain, disability and cartilage degeneration. Synovial membrane HDL-C and ApoA1 levels are inversely associated with cartilage injury and symptom severity, whereas plasma HDL-C levels are associated with osteophyte growth [32–35]. Furthermore, Li et al. reported that patients with knee OA (KOA) had significantly reduced serum ApoD levels, which were negatively correlated with Kellgren–Lawrence grades, visual analogue scale (VAS) scores and TNF-α levels, suggesting the potential use of ApoD for KOA diagnosis [36]. The differentiation potential of monocytes makes them an attractive therapeutic target in many pathological states [37]. The monocyte-to-HDL-C ratio is significantly higher in patients with KOA than in healthy controls and is a risk factor for the radiological severity of OA [38]. The adipsin-to-monocyte chemoattractant protein-1 ratio has been linked to cartilage volume loss in the lateral compartment in patients with KOA [39].

Table 1.

Clinical studies on the association between lipid metabolism and OA.

First author (year)	Type of study	Participants	Sample	OA diagnosis	Lipid or adipokine levels	Correlations
Zhang (2021) [32]	Case-control	184 OA receiving arthroplasty or debridement and 180 controls	Serum; SF	KL; WOMAC	Higher TC and LDL-C levels; lower HDL-C and ApoA1 levels	SF, HDL-C, and ApoA1 levels were negatively correlated with cartilage damage and radiological severity
Wu (2018) [63]	Cross-sectional	164 symptomatic KOA	Serum	ACR	Higher leptin levels; lower adiponectin levels	Leptin and adiponectin levels were significantly and negatively associated with BMD
Li (2023) [41]	MR	All Arthritis Research UK OA Genetics patients and controls	Plasma	KL ≥ 2	Lower omega-3 to omega-6 fatty acid ratio	Optimal unsaturated fatty acids levels should be maintained in patients with genetic susceptibility
Wen (2023) [29]	MR	GWAS database of large samples	/	/	Higher TC, TG, and LDL-C levels	TC, TG, and LDL-C were risk factors for OA
Jansen (2023) [33]	Prospective cohort	682 females from the Rotterdam Study	Plasma	MRI	Lower HDL-C levels	HDL-C levels affected osteophytes in the tibiofemoral and patellofemoral compartments
Herrero-Manley (2023) [34]	Cross-sectional	48 early OA and 48 controls	Serum	LP	Higher TC and LDL levels	Increased pain intensity and disability
Mai (2023) [45]	Cross-sectional	164 KOA and 78 controls	Plasma	ACR	Higher leptin levels	Increased BMI, waist-to-hip ratio, and VAS scores
Cao (2023) [38]	Cohort	323 OA and 283 controls	Plasma	ACR; KL	Higher MHR	Useful marker for diagnosis and monitoring of OA
Tudorachi (2022) [30]	Cross-sectional	56 OA and 29 controls	Plasma	KL	Higher TC and TG levels	Increased incidence and severity of KOA
Li (2022) [36]	Cross-sectional	113 KOA and 97 controls	Serum	ACR	Lower ApoD levels	Negatively correlated with severity
Xie (2022) [45]	Prospective Cohort	344 OA	Serum; cartilage	MRI	Increased C16:1 to C14 and C16:1 to C12 ratios	Promoted patellar cartilage loss
Meng (2022) [59]	MR	39427 OA and 378169 controls	/	KL ≥ 2	Genetic predisposition to 1-SD increments of APOB and LDL	Reduced the risk of KOA, HOA, and KHOA
Gløersen (2022) [61]	Cross-sectional	281 hand OA	Plasma; serum	ACR	Higher leptin levels	Increased pain severity
Schadler (2022) [49]	Cross-sectional	70 OA	Serum	Lequesne index	Higher resistin levels	Positively associated with OA severity
Lambova (2021) [50]	Cross-sectional	73 OA and 11 controls	Serum	KL ≥ 2	Higher leptin and resistin levels	More severe radiological classification
Zhu (2022) [53]	Cohort	200 OA	Cartilage; serum	ACR	Higher leptin, adipsin, and resistin levels	Positively associated with WOMAC scores and increased cartilage defects
Schadler (2021) [49]	Cross-sectional	33 OA	Serum	Symptoms	Higher FABP4 levels	Negatively associated with cartilage thickness
Fan (2021) [51]	MR	10083 OA and 40425 controls	Serum	Symptoms	Higher adiponectin, leptin, and resistin levels	Increased risk of KOA
Mustonen (2019) [40]	Cross-sectional	17 OA	SF	Symptoms	Higher FAs/PUFA levels	Affected joint lubrication, synovial inflammation, pannus formation, and cartilage degradation
Sellam (2021) [54]	Cross-sectional	878 lower limb OA	Serum	ACR; KL ≥ 2; WOMAC; OAKHQOL	Higher LAR levels	Associated with pain level, independent of radiological severity in HOA and/or KOA
Duan (2020) [55]	Cross-sectional	22 obese KOA and 10 healthy weight KOA	SF; serum; IPFP tissue	Symptoms	Higher leptin levels	Significantly correlated with IL-6 levels
Van de Vyver (2020) [43]	Cross-sectional	23 OA and 6 control	SF	Symptoms	Lower omega-6 to omega-3 ratio; higher MUFAs levels	FA profile differed between OA and control
Meessen (2020) [44]	Cross-sectional	1564 OA and 2125 controls	Plasma	KL; OARSI	Longer-chain FAs	Strongest and most specific association with end-stage knee/hip OA
Shmeik (2021) [35]	Cross-sectional	20 OA	Plasma; serum	KL; MRI	Higher TC and LDL levels	Cartilage degeneration
Min (2021) [47]	Cross-sectional	148 KOA and 101 non-KOA	Serum	ACR; KL	Higher leptin levels	Intrinsically related to osteopontin and sclerostin
Scotece (2020) [56]	Cross-sectional	100 OA	Cartilage; SF; blood	ACR	Higher adipsin, adiponectin, leptin, and resistin levels	Novel adipokine associated with OA
Orellana (2021) [52]	Cross-sectional	115 OA	SF; serum	KL; WOMAC	Higher adiponectin and leptin levels	Associated with clinical severity and local inflammatory markers in KOA
Loef (2020) [62]	Cohort	5328 OA	Plasma	ACR	Higher SFA, and PUFA levels	Positively associated with hand OA and KOA in males
Kroon (2019) [60]	Cross-sectional	6408 OA	Serum	ACR	Higher leptin levels	Greater correlation with KOA
Funck-Brentano (2019) [71]	MR	384838 OA	Serum	In-hospital	No difference	No causality
Martel-Pelletier (2019) [39]	Cohort	583 OA	Serum	Symptoms	Higher adipsin to MCP-1 ratio	Cartilage degeneration
Hindy (2019) [40]	MR	75778 OA	Blood	Medical records	Lower LDL-C levels	Reduced risk of OA diagnosis and joint replacement
Naqvi (2019) [48]	Cross-sectional	280 OA and 308 controls	Blood	ACR; KL ≥ 2	−420 mutant genotype of resistin	Increased susceptibility to OA
Mustonen (2019) [40]	Cross-sectional	10 RA, 10 OA, and 6 controls	SF	/	Lower omega-6 PUFA levels	Limited inflammation and cartilage destruction in KOA
Sanchez-Santos (2019) [58]	Longitudinal cohort	952 OA	Serum	KL ≥ 2	Higher TG levels; lower HDL-C levels	Interphalangeal joint pain promoted hand OA, not KOA
Baghban (2021) [64]	Double-blind	76 obese females with KOA	Serum	WOMAC	No difference	No causality
Cioroianu (2022) [65]	RC	62 KOA receiving colchicine or physiotherapy	Blood	KL ≥ 2	Higher TC and TG levels	Associated with high WOMAC and VAS scores
Hsu 2022 [66]	RC	66 obese patients with KOA	Blood	KL ≤ 3	Lower TC and TG levels	Improved blood biochemistry and lower-limb function.
Park (2021) [69]	RC	75 KOA	Blood	KL	Lower resistin levels	Reduced inflammatory cytokine levels, increased muscle strength
Rondanelli (2020) [70]	Double-blind	48 obese patients with moderate KOA	Blood	KL:1-3	Lower TC levels	Improved inflammation, knee function, metabolic profile, and body composition
Strath (2020) [67]	RC	21 adults aged 65–75 years with KOA	Blood	ACR	Lower leptin levels	Relieved pain; may be an opioid alternative
Kasperczak (2019) [68]	Case-control	22 female and 13 male patients with OA	Blood	/	Higher HDL levels	Improved lipid metabolism

ACR: American College of Rheumatology criteria; Apo: apolipoprotein; BMD: bone mineral density; BMI: body mass index; C12: dodecanoylcarnitine; C14: tetradecanoylcarnitine; C16:1: hexadecenoylcarnitine; FA: fatty acid; FABP4: fatty acid-binding protein 4; GWAS: genome wide association study; HDL-C: high-density lipoprotein cholesterol; HOA: hip osteoarthritis; IL: interleukin; IPFP: infrapatellar fat pad; KHOA: knee and hip osteoarthritis; KL: Kellgren-Lawrence; KOA: knee osteoarthritis; LAR: leptin to adiponectin ratio; LDL-C: low-density lipoprotein cholesterol; LP: Luyten’s proposal; MCP-1: monocyte chemoattractant protein 1; MHR: monocyte-to-high density lipoprotein-cholesterol ratio; MR: Mendelian randomization; MRI: magnetic resonance imaging; MUFA: monounsaturated fatty acid; OA: osteoarthritis; OAKHQOL: Osteoarthritis Knee and Hip Quality of Life; OARSI: Osteoarthritis Research Society International; PUFA: polyunsaturated fatty acid; RA: rheumatoid arthritis; RC: randomized controlled; SD: standard deviation; SF: synovial fluid; SFA: saturated fatty acid; TC: total cholesterol; TG: triglycerides; VAS: visual analogue scale; WOMAC: Western Ontario McMaster University Osteoarthritis Index.

Mustonen et al. detected SF-derived lipid mediators and Fas in patients with arthritis and observed a significant decrease in PUFA levels. Additionally, Fas affect cartilage architecture and inflammation more severely in OA than in rheumatoid arthritis [40]. Furthermore, Li et al. reported that increased levels of omega-6 Fas were associated with a lower incidence of hip OA (HOA) and KOA [41], consistent with two previous reports [42, 43] that concluded that the FA-to-PUFA ratio and increased levels of monounsaturated Fas (MUFAs) were associated with OA inflammation and cartilage destruction. This may be caused by changes in FA profiles in the blood, SF and joint tissues as well as possible effects on PUFA-derived lipid mediators. According to Meessen et al. and Xie et al. OA is linked to longer-chain Fas, with patellar cartilage loss caused by long-chain FA β-oxidation. This could be related to the higher ratios of hexadecenoylcarnitine (C16:1) to tetradecanoylcarnitine (C14) and C16:1 to dodecanoylcarnitine (C12) [44, 45]. In contrast, oral administration of 1000 mg L-carnitine for 12 weeks in patients with KOA improved the BMI, although it had no significant effect on the lipid profile or WOMAC score [64].

According to several studies, leptin and resistin levels are higher in patients with OA than in controls. Additionally, leptin has been observed to positively correlate with the waist-to-hip ratio, body mass index (BMI) and VAS scores, all of which can be utilized as indicators of OA [46], and with the cartilage integrity markers osteopontin and sclerostin [47]. Genetic variants of resistin have been shown to increase KOA susceptibility and are associated with the Lequesne index, which is a measure of OA severity [48, 49]; Both leptin and resistin levels have been correlated with the radiological stage of OA [50]. Other adipokines have been associated with OA, and Fan et al. identified high adiponectin levels as a risk factor for KOA [51]. Orellana et al. reported that the clinical severity of KOA is more strongly associated with adiponectin than with leptin [52]. Studies by Zhu et al. and Sellam et al. showed that resistin was linked to cartilage defects and that the leptin-to-adiponectin ratio was positively correlated with Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) scores and higher pain intensity in patients with HOA and/or KOA [53–55]. Pain may be caused by a significant correlation between leptin and IL-6, which indicates that the role of adiponectin in pain is not consistent with the above findings and may play a positive role. Retinol binding protein 4 (RBP4), a member of the lipocalin family of proteins, has been identified as a novel adipokine produced within joints affected by OA and is linked to elevated levels of matrix metalloproteinases (MMPs) and adipokines [56]. The levels of FA-binding protein 4 (FABP4), which is mostly expressed in the adipose tissue and plays a role in the intracellular transport of Fas, have been correlated with cartilage thickness and BMI in patients with KOA [57].

Previous studies have reported different lipid and adipokine levels in different joints. For example, Sanchex-Santos et al. reported that TG and HDL-C levels were associated with OA in the interphalangeal joint, but not in the knee [58]. Meng et al. hypothesized that the genetic propensity for increased LDL and ApoB levels lowers the risk of developing KOA and/or HOA [59]. Leptin has been shown to have a higher proportion of intermediaries in KOA than in hand OA [60] and to have a greater influence on hand pain than on lower extremity pain through its effects on inflammatory mediators [61]. In addition, there are sex differences in the influence of adipokines on joints. The correlation between obesity, leptin levels and OA is stronger in females than in males, and saturated FA (SFA) and PUFA levels are positively correlated with hand and knee OA in males, but not in females [62]. Other studies have reported an association between adipokines and bone mineral density (BMD) in patients with OA. Decreases in the BMD of the femur, femoral neck and femoral axis were positively and negatively correlated with serum leptin and adiponectin levels, respectively. No correlation was observed between the serum resistin levels and BMD [63].

Physical therapy, low-fat diet, and diet control plus elastic band exercise have been shown to significantly decrease knee VAS and WOMAC scores, TC, TG and leptin levels, joint load and inflammatory and oxidative stress, although no significant improvement in cartilage thickness was observed [65–67, 142]. The Arthritis, Diet and Activity Promotion Trial (ADAPT) put particular emphasis on low-fat diets and low-load blood flow restriction exercises; changes in inflammatory mediators are more important than changes in BMI [143–144]. Spa therapy affected only HDL-C levels [68]. Patients with KOA treated with exercise combined with electromyostimulation showed a significant reduction in body weight, reduced levels of resistin, IL-6, TNF-α and C-reactive protein (CRP) and improved Knee Injury and Osteoarthritis Outcome Score scores [69]. Treatment of patients with non-animal-derived chondroitin sulphate resulted in significant improvements in BMI, WOMAC scores, CRP levels and knee homeostasis [70].

In contrast, a 2019 study reported no association between knee, hip, or hand OA and lipid metabolic factors; instead, OA depended on the genetic determination of BMI and BMD [71]. However, these findings may be due to the complexity of the data and the absence of advanced research methods such as imaging grading. Numerous clinical studies have investigated the relationship between lipid metabolism and OA. Generally, lipids and adipokines have been shown to influence OA; however, not all data were statistically significant. Based on this evidence, we conclude that lipids, lipid derivatives and adipokines affect the occurrence and development of OA from various perspectives, including pain, osteophyte formation, radiological severity and changes in inflammatory and catabolic markers in synovial and cartilaginous tissues. Different factors can affect OA, either positively or negatively. Limited evidence exists that diet and exercise can improve OA, and only a few studies have demonstrated the role of therapeutic agents in regulating lipid biosynthesis and metabolic pathways in patients with OA and abnormal lipid metabolism. This study provides a theoretical basis for the development of novel clinical treatments.

3.2. Evidence from animal studies

During the past 5 years a number of animal studies have helped to elucidate the role of lipid metabolism in OA pathogenesis, most of which reported an association between lipids or adipokines and OA outcomes. Of the 33 published studies, 11, 6 and 5 involved SD, C57BL/6J and Wistar rats, respectively. The sex of the animals was not reported in six publications [74–80]. The animals were aged between 7 days and 12 weeks at the start of the study, with most being 8 weeks old. Moreover, there were 3 to 26 animals in each group, although one study [77] did not provide this information. The study period ranged from 9 to 52 weeks. The details of the animal studies and their main findings are summarized in Table 2.

Table 2.

Animal studies on the association between lipid metabolism and OA.

First author (year)	Animal	OA scoring	What was studied	Results
Liang (2023) [74]	30 SD rats	/	Efficacy and mechanism of Longbie capsules	Reduced subchondral bone loss by regulating lipid metabolism
Abughazaleh (2023) [75]	39 SD rats	Mankin; OARSI	Metabolic effects of aerobic training	Increased protection against knee damage
Tamborena Malheiros (2023) [86]	72 male and female Wistar rats	Subjective behavior scale	Effects of obesity on inflammation and pain	Obesity causes OA, inflammation, and pain
van Gemert (2023) [103]	28 female E3L.CETP mice	OARSI	Effects of intensive cholesterol-lowering treatments	Reduced joint inflammation but not end-stage pathology
Lambova (2021) [50]	50 female ICR mice	Four-degree severity gradation scale	Levels of visfatin and cartilage turnover biomarkers	Early personalized and complex intervention may prevent or inhibit joint damage
Liao (2023) [98]	30 male SD rats	Krenn’s	Effects of L-carnitine on knee OA	L-carnitine may be a potential treatment strategy for knee OA
van Gemert (2023) [104]	75 female E3L.CETP mice	OARSI	Effects of interleukin-1β inhibitor combined with cholesterol-lowering therapies	Diminished synovial thickening and cartilage degeneration
Sudirman (2023) [95]	49 male SD rats	/	Effects of collagen fermented from jellyfish	Protective effects on articular cartilage
Cao (2022) [93]	20 male C57BL/6 mice	OARSI	Effects of LRP3 knockdown on OA	Upregulated syndecan-4 by activating the Ras signaling pathway
Cho (2022) [105]	6 male Wistar rats	/	Effects of Lactobacillus LA-1 on OA	Regulated the gut microenvironment
Aboudeya (2022) [82]	30 male Wistar rats	/	Effects of short-term HFD	Increased resistin levels and autophagy-related gene expression
Johnson (2022) [108]	63 male SD rats	OARSI	Effects of nano-EE on OA	Ameliorate OA at a low dosage
Wang (2022) [94]	44 newborn male C57BL/6J mice	OARSI	Intrauterine exposure factors in OA	Provided evidence for the fetogenetic origin of cartilage diseases
Ma (2022) [112]	40 B6 female mice	OARSI	Effects of Sparstolonin B on obesity-induced OA	Inhibited cartilage degeneration and subchondral bone calcification
Wang (2021) [97]	41 male New Zealand white rabbits	Mankin	Effects of UTMDSV on OA	Promotes cartilage extracellular matrix synthesis by modulating the PPARγ-mediated cholesterol efflux pathway
Huang (2021) [87]	Adult male Wistar rats	OARSI	Biomarkers of OA	Adiponectin, leptin, and MMP3 levels were significantly higher in OA
Jung (2021) [100]	50 male SD rats	OARSI	Effects of palmitoylethanolamide on OA	Reduced knee joint swelling and cartilage degradation
Hülser (2021) [88]	9 male C57BL/6J mice	OARSI	Metabolic effects of diet-induced OA	Local adipokine expression in joints was associated with OA
Jiang (2021) [89]	72 male SD rats	Mankin	Leptin signaling pathway in OA	Leptin may be one of the initiating factors of obesity-induced OA
Tan (2021) [90]	45 male C57Bl/6J mice	ACS system	Models induced by different diets	Palmitate diet promoted UPR/ER stress and cartilage lesions
Pragasam (2021) [76]	24 Wistar rats	OARSI	Temporal changes in tibial subchondral bone and cartilage	Multiple metabolic perturbations resulted in degenerative changes
Villalvilla (2020) [81]	36 male rabbits	Mankin	Effects of HFD in OA	oxLDL may be involved in the association with OA
Song (2020) [106]	Male Balb/c mice	Mankin	Effects of Lactobacillus paracasei M5	Protected against HFD-induced OA
Chen (2020) [83]	24 male SD rats	OARSI	Roles of resistin	Inhibition of resistin prevented OA
Tsai (2020) [77]	SD rats	/	Effects of visfatin	Rescued EPC angiogenesis and pathologic changes
Kimmerling (2020) [80]	fat-1 and wild-type mice	Mankin	Lipid profiles and pathological changes	Elevated levels of omega-3 FAs in fat-1 mice inhibited OA
Tsubosaka (2020) [92]	72 male C57BL/6J mice	OARSI	Effects of eicosapentaenoic acid on OA	Prevented OA progression
Paré (2020) [78]	50 Df-/- and Df+/+ mice	OARSI	Effects of adipsin deficiency	Protected the articular tissues from spontaneous OA progression
Zheng (2020) [79]	126 adult SD rats	Mankin	Effects of C1q-TNF-related protein 9	Alleviated monoiodoacetate-induced OA
Chang (2020) [109]	42 male SD rats	OARSI	Effects of Mytilus edulis water extract	Suppressed inflammation and cartilage degradation
Mustonen (2019) [73]	22 female New Zealand rabbits	/	Infrapatellar FA composition	Reduced omega-3 to omega-6 polyunsaturated FA ratio during early OA
Votava (2019) [91]	36–45 male C57BL/6J mice	Mankin	Role of dietary FA on bone and cartilage	High-fat diet decreases bone quality
Hung (2019) [85]	5 female leptin-deficient and lean +/+ mice	/	Effects of leptin deficiency on the microstructure of growth plate cartilage	Obesity may adversely affect the growth plate cartilage structure
Yamamoto (2022) [84]	LRP1 knockout mice	/	Lack of LRP1	LRP1 plays a critical role in cartilage
Belenska-Todorova (2021) [102]	40 outbred ICR female mice	OARSI	Effects of metformin and alendronate on OA	Reduced leptin and resistin levels and pathological changes
Xie (2020) [107]	30 male SD rats	Mankin	Acupuncture on OA pathogenesis in rats fed an HFD	Relieved inflammation and lipid metabolism disorder
Siriarchavatana (2019) [110]	48 female SD rats	Mankin	Effects of foods on metabolic OA	Lowered C-telopeptide of type II collagen levels
Yen (2019) [111]	36 male SD rats	Histopathology	Effects of Shiikuwasha extract	Improved obesity
Luo (2019) [96]	16 New Zealand rabbits	/	Effects of ultrasound on synovial fluid proteome	Altered protein levels in synovial fluid
Chen (2019) [99]	40 male Wistar rats	OARSI	Effects of shea nut oil extract	Symptom relief and cartilage protection

ACS: articular cartilage structure; EPC: endothelial progenitor cell; ER: endoplasmic reticulum; FA: fatty acid; HFD: high-fat diet; LRP: low-density lipoprotein receptor-related protein; MMP: matrix metalloproteinase; nano-EE: Echinacea purpurea ethanol extract nanoparticles; OA: osteoarthritis; OARSI: Osteoarthritis Research Society International; oxLDL: oxidized low-density lipoprotein; PPAR: peroxisome proliferator-activated receptor; SD: Sprague–Dawley; UPR: unfolded protein response; UTMDSV: ultrasound-targeted simvastatin-loaded microbubble destruction.

Rats fed a high-fat diet (HFD) showed increased joint degeneration, OA scores, and body weight compared to controls. Villalvilla et al. reported that cartilage catabolic gene expression and structural integrity are unaffected by HFD consumption [81]. However, this contrasts with the majority of studies. For example, Aboudeya et al. reported that consumption of an HFD significantly increased serum and cartilage resistin levels, induced cartilage breakdown and increased autophagy, thus exacerbating OA [82]. Inhibition of resistin prevents degenerative alterations in cartilage and changes in monocyte expression in a rat model of OA [83]. Clinical studies have shown that visfatin, adipsin and leptin play important roles in OA pathogenesis. Visfatin short hairpin RNA-transfected rats and homozygous adipsin-deficient mice showed elevated serum levels of adiponectin, protecting endothelial progenitor cells (EPCs) from angiogenesis and pathological changes compared to control animals [77, 78]. Conditional deletion of LRP1 shows the importance of this protein in controlling high-affinity ligands in the cartilage in vivo [84]. Hung et al. reported that leptin-deficient animals had decreased chondrocyte volume and number and concluded that leptin was necessary for healthy bone formation and maintenance [85]. In contrast, studies by Tamborena Malheiros et al., Hülser et al. and Huang et al. showed that leptin levels were related to tibia and liver scores, led to deep cartilage and subbone defects, increased TNF-α, IL-1β, IL-6, IL-8, MMP3 and leptin levels in serum and/or adipose tissue, and reduced IL-10, total vitamin D3, and C-telopeptide fragments of type II (CTX-II) levels [86–88]. In vivo mechanistic studies have suggested that leptin levels in SF increase later than those in serum, and that leptin induces Toll-like receptor 4 (TLR4) expression via the Janus kinase (JAK) 2-signal transducer and activator of transcription 3 pathway in obesity-related OA, which may represent one of the initiating factors of OA [89]. These studies differed in that Malheiros et al. and Hülser et al. suggested that adiponectin levels decreased in patients or animals with OA and were correlated with the duration of obesity. However, Huang et al. reported that adiponectin levels are increased in OA. Because PUFAs have received considerable attention, Mustonen et al. demonstrated that inflammation and cartilage deterioration in the early stages of OA are consistent with a decrease in the omega-3 to omega-6 PUFA ratio in the infrapatellar fat pad [73]. Furthermore, Votava et al. and Kimmerling et al. reported that groups fed high-fat diets had lower-quality bones. Regardless of body weight, metabolic inflammation can be reduced by converting omega-6 PUFAs into omega-3 PUFAs using omega-3 FA desaturase encoded by the Fat-1 gene [80, 91]. Additionally, Tsubosaka et al. reported that physiologically active gelatin hydrogels gradually release eicosapentaenoic acid, another omega-3 PUFA, into the joints, inhibiting cartilage tissue breakdown and improving the synovitis score [92]. Tan et al. demonstrated the negative effects of SFAs in mice fed SFA, unsaturated FA, palmitate, or oleate diets. Mice fed a palmitate diet exhibited increased expression of apoptosis markers, unfolded protein response/ER stress markers, and negative regulators of cell survival in the knee articular cartilage [90]. Regardless of diet, LRP3 deficiency exacerbates the breakdown of the cartilage extracellular matrix [93]. In rats, an increased BMI was associated with tibial cartilage erosion, fibrillation and osteophyte formation after 6 and 9 months, respectively, as verified by histology and scanning electron microscopy [76]. Wang et al. showed that persistent overfeeding of obese pregnant mice increased MMP levels and apoptosis and decreased levels of chondrocyte collagen II, a crucial cartilage regulator in male newborn mice. This decrease was linked to elevated leptin-activated mitogen-activated protein kinase (MAPK)/extracellular signal-regulated kinase (ERK) signalling [94].

Research has suggested a role in the regulation of lipid biosynthesis and metabolism in cholesterol-induced OA. For example, in an HFD rat model of KOA, aerobic training, and collagen supplementation reduced the levels of TC, TG and HDL-C; suppressed pro-inflammatory cytokines, leptin and adiponectin expression; improved KOA scores; and protected against cartilage injury [75, 95]. Ultrasound therapy can decrease ApoA1 and increase FABP4 levels in the SF of rabbits with KOA [96], and a study by Wang et al. suggested that customized aggrecan and collagen II levels in the SF were enhanced by ultrasound-targeted simvastatin-loaded microbubble destruction, which impacted the cholesterol efflux pathway [97]. Using a rat model of KOA, Liao et al. showed that L-carnitine enhanced mitochondrial function and lipid accumulation, prevented synovitis in synovial tissue, and reduced cartilage Mankin scores [98]. Chen et al. described an oral shea nut oil triterpene concentrate that exhibited normal lipid and biochemical profiles, was safe to use over an extended period, and ameliorated cartilage degradation in a rat KOA model [99]. Zheng et al. discovered that intraarticular injection of recombinant C1q-TNF-related protein 9, a novel adipokine closely related to adiponectin, alleviated monoiodoacetate-induced inflammation, oxidative stress and knee cartilage damage in rats in a dose-dependent manner by deactivating p38 MAPK and nuclear factor (NF)-κB signalling pathways [79]. Similarly, 100 mg/kg/day palmitoylethanolamide, an endogenous bioactive lipid, reduced knee joint swelling and cartilage degradation as well as serum levels of leukotriene B4, nitric oxide and prostaglandin E2 [100]. According to Lambova et al. and Belenska-Todorova et al. the combined early administration of alendronate and metformin in mice with collagenase-induced OA led to a decrease in visfatin, leptin and resistin levels and an increase in the cartilage biomarkers CTX-II and cartilage oligomeric protein (COMP), protecting against cartilage and subchondral bone damage [101, 102]; these effects were not observed following the treatment of mice with advanced disease. This may be due to the heterogeneity in the late stages of the disease, which cannot be adequately reversed. Therefore, early interventions are necessary. Van Gemert et al. showed that E3L.CETP mice treated with cholesterol-lowering medication and an IL-1β inhibitor exhibited markedly reduced synovial thickening and cartilage degeneration compared with control mice [103, 104]. In a rat model of OA, Lactobacillus LA-1 treatment increased the levels of short-chain FAs and inhibited autophagy-induced cartilage damage and synovial membrane inflammation [105]. Lactobacillus paracasei M5 treatment of mice fed an HFD was observed to reduce serum and infrapatellar fat pad leptin levels and protect articular cartilage from destruction [106].

Acupuncture has been shown to regulate lipid metabolism and the gut microbiota in rats with obesity-induced OA [107]. The potential effects of traditional Chinese medicine on OA have been gaining attention in recent years, with Shiikuwasha extract, micro-nanoencapsulated Echinacea purpurea ethanol extract, and greenshell mussel (Perna canaliculus) and blue mussel (Mytilus edulis) water extracts producing anti-inflammatory and antioxidant effects in rats with OA, including decreased leptin, TC, TG, MMP, cytokine, NF-κB and p-JAK2 levels and inducible nitric oxide synthase production, and increased collagen II and proteoglycan expression, in HFD-fed rats, thereby reducing pain and attenuating cartilage degradation [108–111]. Sparstolonin B (Ssn B) has shown anti-inflammatory effects in many conditions and an obesity-induced model of OA, increased Osteoarthritis Research Society International scores, excessive joint space narrowing, and osteosclerosis in the load-bearing area of the tibial plateau [112]. Liang et al. established an OA + osteoporosis rat model [74]. Eight weeks of treatment with Longbie capsules significantly increased the subchondral bone quality and BMD and maintained balanced lipid metabolism, potentially through forkhead box O and cyclic adenosine 3′,5′-monophosphate signalling pathways.

These findings identify abnormal lipid metabolism as one of the main drivers of OA, aggravating OA outcomes by inducing systemic pro-inflammatory mediators and affecting several OA and pathological markers. As most of these studies did not investigate the molecular mechanisms and signalling pathways underlying this correlation, more research is necessary to determine the precise processes linking lipid metabolism to the development of OA.

3.3. Evidence from cellular studies

The cellular studies performed to explore the pathogenesis and mechanism of lipid metabolism in OA are summarized in Table 3. These experiments were conducted in parallel with in vivo experiments. For example, Scotece et al. reported that RBP4 is the most prominently expressed adipokine in OA chondrocytes and is positively linked to MMPs and adipsin [56]. Villalvilla et al. reported that adipsin-oxidized LDL decreases the levels of inflammatory agents and mediators, chemokines, MMP-13 and cyclooxygenase 2, resulting in reduced inflammation and OA chondrocyte degradation [81]. Chen et al. and Tsai et al. reported that resistin/visfatin-activated protein kinase Cα, p38 and c-Jun N-terminal kinase or phosphoinositide 3-kinase (PI3K)/Akt signalling pathways, inhibiting miR-381/miR-485-5p synthesis and thereby increasing the production of vascular endothelial growth factor (VEGF) and vascular cell adhesion molecule type 1, encouraging VEGF-induced EPC angiogenesis, and increasing monocyte adherence to OA synovial fibroblasts (OA-SFs) [77, 83]. Yamamoto et al. verified the results of in vivo experiments, which showed that inhibiting the interaction between LRP1 and its ligand led to cell death [84]. Liao et al. demonstrated in vitro that L-carnitine supplementation can increase the β-oxidation of FAs, decrease fat storage and enhance mitochondrial performance by preventing activation of the AMP-activated protein kinase-acetyl-CoA carboxylase (ACC)-carnitine palmitoyltransferase I pathway, therefore reducing synovitis [98]. Ssn B was reported by Ma et al. to suppress the inflammatory response of human OA chondrocytes and extracellular matrix degradation induced by free FA palmitic acid (PA) in a concentration-dependent manner, as well as inhibiting the formation of the TLR4/myeloid differentiation protein-2 complex caused by PA and inactivation of NF-κB [112]. In addition, metformin and alendronate attenuated fibroblast differentiation [102]. These in vitro studies were not presented as independent evidence but as an additional exploration of the underlying mechanisms and showed that lipids or adipokines affect OA through various mechanisms.

Table 3.

Cellular studies on the association between lipid metabolism and OA.

First authors (year)	Cell type	Stimulant	Signaling pathway	Effect on OA
Liao (2023) [98]	Primary rat fibroblast-like synoviocytes	Lipopolysaccharide; L-carnitine	AMPK-ACC-CPTI	Pro-inflammatory; synovial protective effects
Ma (2022) [112]	Human OA chondrocytes	Palmitic acid	NF-κB	Activated NF-κB; upregulated inducible nitric oxide synthase, cyclooxygenase-2, prostaglandin E2, and nitric oxide; degraded extracellular matrix
Villalvilla (2020) [81]	OA articular chondrocytes	Oxidized low-density lipoprotein	/	Modified pro-inflammatory and catabolic gene and protein expression
Chen (2020) [83]	OA-SFs	Resistin	PKC, p38, and JNK	Resistin inhibited miR-381
Tsai (2020) [77]	OA-SFs	Visfatin short hairpin RNA	PI3K/Akt	Visfatin inhibited miR-485-5p
Sun (2023) [134]	Human chondrocytes	Palmitic acid; xanthohumol	AMPK and NF-κB	Obesity-related OA; decreased cartilage matrix degradation induced by palmitic acid
Li (2023) [72]	Human and mouse chondrocytes	Cholesterol; circARPC1B	/	Downregulated circARPC1B; preserved the extracellular matrix
Gong (2023) [141]	Human and mouse chondrocytes	Doxorubicin	PI3K/Akt	Induce chondrocyte senescence
Yang (2023) [138]	Human C28/I2 chondrocytes	Andrographolide	NF-κB	Reduced reactive oxygen species generation and activated ADAMTS4
Primrose (2023) [118]	Human chondrocytes	Leptin	/	Reduced levels of HES1; increased levels of ADAMTS5 and MMP13
Yamamoto (2022) [84]	LRP1-deficient fibroblasts and human chondrocytes	Soluble LRP1-II	/	Increased the levels of slit guidance ligand 2, ADAMTS1, and TNF-inducible gene 6 in the chondrocyte medium
Sanchez (2022) [135]	Primary OA chondrocytes	Curcuma longa; Boswellia serrata	Nrf2; Nrf1; PPARα	Anti-oxidative, anti-inflammatory, and anti-catabolic effects
Wang (2022) [114]	Primary rat condylar chondrocytes	Fluid flow shear stress	PPARγ	Inhibited adipogenesis of chondrocytes and cartilage degeneration
Jin (2022) [133]	SW1353 chondrosarcoma cells	Lipopolysaccharides; fatty acids	TLR4/NF-κB	Mimicked OA status in vitro; affected chondrocyte pyroptosis
Liu (2022) [130]	Primary mouse chondrocytes	Acetyl-CoA carboxylase inhibitor	/	Ameliorated lipid accumulation in chondrocytes
Zhao (2022) [126]	Human chondrocytes	Resistin	p38-MAPK; NF-κB	Upregulated pro-inflammatory cytokines and matrix-degrading enzymes
Wei (2022) [119]	Primary OA rat chondrocytes	Leptin	/	Significantly increased lysyl oxidase-like 3, inhibited autophagy
Mao (2021) [137]	ATDC5 cells	IL-1β; PPARγ inhibitor	/	Inflammation, apoptosis, and extracellular matrix degradation
Wang (2022) [113]	C28I2 cells	Digoxin	/	Increased the expressions of two markers of anabolism
Papageorgiou (2021) [136]	OA chondrocytes	Resveratrol	NF-κB	Induction of autophagy
Chang (2021) [125]	Human OA chondrocytes	Visfatin	p38	Decreased intracellular elasticity and viscosity
Philp (2021) [124]	Human OA chondrocytes	Visfatin	/	Cartilage matrix degradation
Cheleschi (2021) [127]	Human OA chondrocytes	Hydrostatic pressure; visfatin	/	Cartilage matrix degradation
Harasymowicz (2021) [120]	Human OA chondrocytes	Adiponectin	/	Cartilage degeneration
Law (2020) [123]	OA-SFs	Visfatin	p38; AMPK	Enhanced adhesion of monocytes to OA-SFs
Chang (2020) [115]	OA-SFs	Apelin	PI3K; ERK	Pro-inflammatory
Gu (2020) [131]	Human SW1353 chondrocytes	AGEs; GW9508	NF-κB	Decreased free fatty acid receptor 1; suppressed inflammation and NF-κB activation
Chai (2020) [129]	Human SW1353 chondrocytes	IL-1β; omentin-1	/	G1 phase cell-cycle arrest; protection of chondrocytes from senescence
Frommer (2019) [118]	Primary human and murine osteoblasts	Free fatty acids	TLR4	Subchondral bone damage in OA
Cheleschi (2019) [122]	Human OA synoviocytes	Visfatin; resistin	NF-κB	Induced apoptosis and superoxide production
Zhuo (2019) [128]	Human SW1353 chondrocytes	TNF-α; INT-777	p38; NF-κB	Abnormal loss of extracellular matrix; reduced extracellular matrix degradation
Cheleschi (2019) [121]	Human OA chondrocytes	Visfatin	NF-κB	Induce apoptosis and oxidative stress
Zhang (2019) [116]	ATDC5 cells	Leptin	/	Stimulate hypertrophic differentiation of ATDC5 cells
Belenska-Todorova (2021) [102]	Bone marrow-derived cells	Metformin and alendronate	/	Inhibited the expression of RANK and RANKL on osteoblasts and osteoclasts and fibroblast differentiation
Nogueira-Recalde (2019) [139]	T/C28a2 chondrocytes	IL-6; fenofibrate	PPARα	Induced senescence; reduced the number of senescent cells
Vaamonde-Garcia (2019) [140]	Human OA-SFs	TGF-β1; prednisolone and/or 15d-PGJ2	PPAR-γ	Pro-fibrotic; anti-fibrotic

ACC: acetyl-CoA carboxylase; ADAMTS: a disintegrin and metalloproteinase with thrombospondin motifs; AGEs: advanced glycation end products; AMPK: AMP-activated protein kinase; CPTI: carnitine palmitoyltransferase I; ERK: extracellular signal-regulated kinase; 15d-PGJ2: 15-deoxy-Δ-12,14-prostaglandin J2; IL: interleukin; JNK: c-Jun N-terminal kinase LRP: low-density lipoprotein receptor-related protein; MAPK: mitogen-activated protein kinase; MMP: matrix metalloproteinase; NF: nuclear factor; OA: osteoarthritis; OA-SF: osteoarthritis synovial fibroblast; PI3K: phosphoinositide 3-kinase; PKC: protein kinase C; PPAR: peroxisome proliferator-activated receptor; RANK: receptor activator of NF-κB; RANKL: receptor activator of NF-κB ligand; TGF-β1: transforming growth factor-beta 1; TLR: Toll-like receptor; TNF: tumour necrosis factor.

Other studies have used in vitro experiments as a basis for research rather than as an adjunct to in vivo models. Chondrocytes treated with high concentrations of leptin upregulate the expression of a disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS) 5, MMP13 and collagen X (a specific marker of hypertrophic and calcified chondrocytes) and reduce type II collagen and aggrecan levels, thereby stimulating chondrocyte hypertrophic differentiation and accelerating senescence by activating the leptin pathway via the long form of the leptin receptor [116–118]. Leptin has been shown to significantly increase the expression of lysyl oxidase-like 3, the silencing of which stimulates chondrocyte proliferation and prevents apoptosis [119]. Harasymowicz et al. reported adiponectin receptor 1 to be the most prevalent adiponectin receptor in primary OA chondrocytes and showed that adiponectin induces the expression of chemokines, pro-inflammatory factors and MMP genes in chondrocytes from patients with OA, suggesting that these factors are crucial for cartilage degeneration in OA [120]. Studies have evaluated the effects of visfatin and resistin on OA chondrocytes or SF, revealing that they increase IL-1β/6/17 and chemokine ligand 3/4 levels, induce the production of MMPs, an apoptosis inhibitor and superoxides in OA chondrocytes or SF, damage the intracellular microtubule and microfilament networks in chondrocytes, and enhance intercellular adhesion molecule type 1 expression in OA-SFs to facilitate the adhesion of monocytes via p38 or NF-κB signalling pathways [121–126]. The effects of visfatin were exacerbated by hydrostatic pressure exposure [127]. An essential component of the bile acid receptor subclass, the G protein-coupled bile acid receptor, can stimulate the development of type II collagen and agglomeran while suppressing the expression of MMPs and ADAMTSs in SW1353 cells [128]. The adipokine apelin, which is involved in various immune cell activities, has been demonstrated to activate the PI3K and ERK signalling pathways-dependently and time-dependently promote the manufacturing of IL-1β in OA-SFs, albeit more in vitro and in vivo research is required to validate this [115]. The recently discovered anti-inflammatory adipokine omentin-1 reduces the decrease in sirtuin 1 (SIRT1), inhibiting IL-1β-induced chondrocyte senescence through its role in lipid metabolism [129]. Inhibition of ACC reduces fat accumulation in chondrocytes, suggesting that it may be a potential treatment for OA associated with obesity [130]. The well-known free FA receptor 1, GW9508, prevents the degeneration of type II collagen and aggrecan by reducing the expression of MMPs and ADAMTS-4/5 [131]. Growth differentiation factor 11 (GDF11) is necessary to stop bone marrow mesenchymal stem cells from differentiating into adipocytes; the stimulation of OA chondrocytes by fluid flow shear stress elevated the expression of peroxisome proliferator-activated receptor (PPAR)γ, CCAAT enhancer binding protein alpha, FABP4, perilipin-1 and AdipoQ while decreasing GDF11 expression. The SUMOylation of PPARγ was modified by exogenous GDF11, hence mitigating the impact of fluid flow shear stress [113]. PA-stimulated OA osteoclasts secrete higher volumes of IL-8, which is inhibited by TLR4 blockade [132]. Lipopolysaccharide-induced chondrocyte pyroptosis is decreased by PUFA inhibition of the NLRP3 inflammasome via the TLR4/NF-κB signalling pathway, while SFAs, MUFAs, or omega-6 PUFAs activate these pathways and thus increase the severity of OA [133].

Xanthohumol, a bioactive prenylated chalcone, has been demonstrated to have antioxidant, anti-inflammatory and anti-obesity properties in a variety of diseases. When administered to chondrocytes along with PA, it completely inhibited PA-induced inflammation and chondrocyte extracellular matrix degradation by inhibiting the NLRP3 inflammasome and reducing mitochondrial dysfunction [134]. To upregulate antioxidant and cytoprotective genes, Curcuma longa and Boswellia serrata extracts can be administered singly or in combination. They induce anti-inflammatory and anti-FA effects [135]. In OA chondrocytes, SIRT1 was identified as an autophagy substrate and was negatively correlated with lectin-like oxidized low-density lipoprotein receptor-1 through NF-κB deacetylation. Resveratrol increases the enzymatic activity of SIRT1 and induces autophagy, which is crucial for maintaining chondrocyte homeostasis [136]. FABP4 knockdown increases cell viability and decreases inflammatory damage, oxidative stress and apoptosis in IL-1β-stimulated ATDC5 cells [137]. Yang et al. discovered a novel FABP4 inhibitor andrographolide [138]. FABP4 levels strongly correlated with decreased cartilage thickness and elevated BMI, as previously established. Clinically approved drugs have shown therapeutic effects against OA. For example, digoxin significantly increased the expression of aggrecan and COMP, protecting OA chondrocytes; however, this regulation of chondrocytes was abolished by the deletion of LRP4 [114]. Fenofibrate, a PPARα agonist that prevented cartilage degradation, enhanced autophagic flux, decreased the amount of senescent cells by apoptosis and was used to treat dyslipidemia in patients [139]. Additionally, PPARγ agonists in conjunction with prednisolone were observed to reduce gene expression of pro-fibrotic markers, alpha-smooth muscle actin protein, and type III collagen levels, inhibiting synovial fibrosis [140]. Gene therapy is a promising and novel therapeutic strategy. For example, in chondrocytes, circRREB1 mediates senescent phenotypes linked to lipid metabolism; the E3 ligase synovial 1 binding site is competitively bound by ARPC1B, which prevents vimentin from being broken down by proteases, thereby slowing OA progression caused by high cholesterol [72]. CircRREB1 mechanically improves FA synthase stability by boosting RanBP2-mediated SUMOylation and blocking acetylation-mediated ubiquitination [141].

Over the past 5 years, a large number of studies have been conducted on the mechanism underlying the influence of lipids or adipokines on the expression of inflammatory factors by OA chondrocytes and synoviocytes, and their catabolism, autophagy and apoptosis. These studies identified potentially effective intervention measures that require further in vivo studies.

4. Potential therapeutic directions

This study is the first to collate all data on the relationship between lipid metabolism and OA published in the past 5 years, and the results of these studies suggest that changes in lipid or adipokine levels affect the occurrence and development of OA to varying degrees. For example, TG, TC, LDL-C, FABP4, RBP4, FAs, MUFAs and long-chain FA β-oxidation are associated with increased inflammation, pain, cartilage degeneration and cell damage. HDL-C, ApoA1, ApoD, PUFAs, LRP3, omentin-1 and adiponectin levels negatively correlated with inflammation, cartilage tissue or cell structural integrity and osteophyte severity. In addition, leptin, visfatin, adipsin and resistin levels are associated with radiological severity and histopathological changes in OA. In particular, the effects of leptin have been reported to be associated with inflammatory pain and the arthritis index, with SF levels rising later than serum levels. Whether TG, HDL-C, LDL, ApoB, SFA, PUFA and Leptin levels are risk factors for OA depends on the site of OA occurrence. The correlation between Leptin and OA was more obvious in women, whereas SFA and PUFA levels were associated only in men. The primary methods used in clinical, animal and cellular studies and their primary findings are summarized in Figure 2. However, the following issues were identified in the process of collation: (1) the current level of evidence is relatively simple; for example, most of the included clinical studies were not prospectively registered, and few studies were conducted using a combination of three or more methods, including systems biology, omics and experiments; (2) the basic information of a few study populations or rats is incomplete, which weakens the representativeness of the study objects and reduces the rigor of the study; and (3) although a large number of studies have proven that lipid metabolism is an important influencing factor in OA, the effect of regulating lipid metabolism on OA has not been fully explored using comprehensive analysis at the molecular level. For example, statins may have cartilage-protective and anti-inflammatory effects in OA; however, this has not yet been confirmed in clinical trials. Therefore, high-quality studies are required to explore the effects and pathways that regulate lipid metabolism in OA. Future studies should consider immortal time bias, the pleiotropic effects of drugs regulating lipid metabolism, and different study designs.

Figure 2.

Main methods used in clinical, animal, and cellular studies and their primary findings. Apo: apolipoprotein; FA: fatty acid; FABP4: fatty acid-binding protein 4; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; MR: Mendelian randomization; MUFA: monounsaturated fatty acid; n-3: omega-3; n-6: omega-6; OA: osteoarthritis; OARSI: Osteoarthritis research Society International; TC: total cholesterol; TG: triglycerides.

5. Conclusions and perspectives

Clinical studies have shown that lipids and adipokines have varying degrees of beneficial or harmful effects on OA pain, inflammation, radiological severity and patient perception. These findings have been verified in animal experiments, in which lipids and adipokines were shown to affect the histopathological progression of OA in the cartilage or synovial membrane, which are tissues that are difficult to examine in clinical studies. Further exploration of the underlying mechanisms revealed that lipids and adipokines participate in the development of OA through a variety of genes and pathways (Figure 3). Most results were consistent and complementary, regardless of the method used, from the molecular level to clinical presentation. Prior to 2019, many studies focused on the area of OA lipid metabolism, but more studies were devoted to exploring the relevance of OA epidemiology, prevalence and joint replacement, mainly consisting of clinical and animal experiments. Over the past 5 years, research in the field of OA lipid metabolism has gradually deepened, including in vivo and in vitro experimental exploration of the mechanism by which lipid metabolism affects OA. However, future studies are required to further cross-validate the association between lipid metabolism and OA, explore the underlying biological mechanisms, and determine the clinical significance of this association. This will allow us to better understand the role of lipid metabolism in the development future.

Figure 3.

Summary of the potential mechanisms underlying the association between lipid metabolism and OA. ACAN: aggrecan; ACC: acetyl-CoA carboxylase; ADAMTS: a disintegrin and metalloproteinase with thrombospondin motifs; APLN: apelin; Apo: apolipoprotein; BS: Boswellia serrata; CL: Curcuma longa; COMP: cartilage oligomeric protein; ECM: extracellular matrix; FN: fenofibrate; HDL: high-density lipoprotein; ICAM: intercellular cell adhesion molecule; LDL: low-density lipoprotein; LRP: low-density lipoprotein receptor-related protein; MMP: matrix metalloproteinase; PUFA: polyunsaturated fatty acid; OA: osteoarthritis; OA-SF: osteoarthritis synovial fibroblast; oxLDL: oxidized low-density lipoprotein; PPAR: peroxisome proliferator-activated receptor; RBP4: retinol-binding protein 4; SIRT1: sirtuin-1; Ssn B: Sparstolonin B; SYVN1: synovial 1; TC: total cholesterol; TG: triglycerides; TLR: Toll-like receptor; VCAM: vascular cell adhesion molecule; VEGF: vascular endothelial growth factor; Xn: xanthohumol.

Funding Statement

This work was supported by the Key Projects of Scientific Research Projects of Higher Education Institutions in Anhui Province (Natural Sciences) (grant no. 2022AH050449), Natural Science Major Project of Universities in Anhui Province (grant no. 2023AH040112), and 2021 Open Fund of the Anhui Key Laboratory of Applied Basic and Development Research of Modern Internal Medicine of Traditional Chinese Medicine (grant no. 2021AKLMCM005).

Authors’ contributions

CXL and LJ contributed to the study design. CXL contributed to data analysis, wrote the first draft, and revised the manuscript. WGZ and SYQ contributed to the data collection. ZXH and DX supervised the project and contributed to the manuscript revision. All the authors have read and approved the final version of the manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data supporting the findings of this study are available in public databases (Web of Science and PubMed).