Study On The Role Of Hyperlipidemia In Knee Osteoarthritis And The Possible Inflammation Mechanism

Knee osteoarthritis(KOA)is the most common degenerative arthritis that results in knee pain,swelling,stiffness,and activity limitation.Osteoarthritis caused 16.28 million of years lived with disability(YLDs)worldwide in 2016 and ranked as the leading cause for lower limb disability.KOA takes up a lot of health resources,induces loss of ability to work,reduces quality of life,and finally results in huge direct and indirect economic burdens worldwide.According to the latest Global Burden of Disease Study,more than 300 million people worldwide have been affected by osteoarthritis,and 14.7 million new cases have been recorded.Along with the aging population and changing lifestyles,the burdens from KOA is dramatically increasing.The mechanisms underlying KOA remain insufficiently clear.Previous researches have demonstrated that aging,obesity,and occupational activities were the risk factors for KOA.In 2017,the Nature Reviews Rheumatology put forward that osteoarthritis was characterized by varied phenotypes while all the phenotypes shared metabolic alterations.Recently,numerous studies have investigated the associations of diabetes or metabolic syndrome with KOA.Hyperlipidemia manifested as abnormally elevated blood lipids or lipoproteins levels,is a common metabolic disorder.Hyperlipidemia might play a role in the development and progression of KOA.Adipocytes share a common mesenchymal stem cell precursor with musculoskeletal cells.When adipocytes proliferation is up-regulated by increased lipids levels,proliferation of musculoskeletal cells could be competitively inhabited.Elevated levels of adipocytokine like leptin up-regulates the expression of matrix-metalloproteinase-13(MMP13),which will promote cartilage catabolism.The elevated levels of oxidized low-density lipoprotein(ox-LDL)and free fatty acid(FFA)following hyperlipidemia can stimulate chondrocytes to release inflammatory cytokines,which are associated with the onset and advance of KOA.A few published studies have found that KOA cases had hyperlipidemia or elevated blood FFA levels.However,the association between hyperlipidemia and KOA and the possible mechanism remain unclear.In order to estimate the association between hyperlipidemia and KOA,we developed the present study with 13906 participants from the Dongfeng-Tongji cohort.The objectives of our study were:(1)To describe the distribution of KOA among middle-aged and older adults and to explore the risk factors;(2)To analyze the crosssectional association between hyperlipidemia and the prevalence of KOA with baseline information collected in 2013;to estimate the longitudinal association between hyperlipidemia and the incidence of KOA with follow-up information collected in 2018;(3)To explore the inflammatory mechanism for the relationship between hyperlipidemia and KOA in mice models.Part One: The Distribution of Knee Osteoarthritis among Middleaged and Older Adults and the Risk FactorsObjectives: To evaluate the prevalence and distribution of KOA among middleaged and older adults,and to explore the possible risk factors.Methods: The participants came from the Dongfeng-tongji cohort,specifically,all of the participants who underwent physical examinations in the Dongfeng Central Hospital in 2013.Face-to-face investigation and physical examination were performed for each participant to collected information regarding demographic characteristics,occupation history,lifestyles,and health status.Knee pain was defined as pain on most days for at least 1 month during one year,or persistent pain within a week,or morning stiffness.KOA was diagnosed based on both knee pain and X-ray radiograph changes.Central obesity was defined as a waist ?90 cm for male or a waist ?80 cm for female,based on the International Diabetes Federation criteria.According to the most intense job the individual had taken,the workload was divided into mild,moderate,heavy,or severe heavy.Based on the job held for the longest duration,the work posture was grouped into sitting,standing,squatting or kneeling,or bending if such posture was maintained for more than half of one work shift(i.e.≥4h for most cases)according to the job content description.Shift work was identified as working with a schedule involving unusual working hours as opposed to the normal daytime work schedule,i.e.from 8:00 to 17:00,for at least 1 year.Chi-squared tests were conducted to compare the prevalence of KOA or knee pain in different groups.Logistic regression models were used to evaluate the associations between potential risk factors and KOA or knee pain,and odds ratios(ORs)and 95% confidence intervals(CIs)were calculated.Results: The present study included 13906 participants with a mean age of 64.7±8.2 years old.A total of 927 KOA cases and 5420 knee pain cases were defined,the prevalence of KOA and knee pain was 6.7% and 39.0%,respectively.The prevalence of KOA and knee pain were much higher in women(8.5% and 46.5%,respectively)than in men(4.4% and 30.0%,respectively).Along with the elevated age,the prevalence of both KOA and knee pain significantly increased.The prevalence of KOA among males aged <65,65-69,≥70 years were 3.6%,4.7%,and 5.0%,respectively;the corresponding prevalence for knee pain were 28.3%,30.0%,and 31.7%,respectively.The prevalence of KOA among females aged <55,55-64,≥65 years were 5.4%,8.0%,and 11.9%,respectively;the corresponding prevalence for knee pain were 39.7%,47.0%,and 52.0%,respectively.Compared with participates with primary education or below(7.5% and 42.1%,respectively),the prevalence of KOA and knee pain significantly decreased in those with university education or above(5.2% and 34.7%,respectively).Central obese individuals showed significant higher prevalence of both KOA(8.8%)and knee pain(44.7%)than their unaffected colleagues(4.8% and 34.1%,respectively).The lowest prevalence of KOA and knee pain were detected among non-drinking males(3.7% and 28.1%,respectively),the prevalence significantly increased among male drinkers(5.0% and 31.8%,respectively)and male ex-drinkers(5.5% and 30.6%,respectively).While females presented no difference in the prevalence of KOA and knee pain regarding different drinking status.Participants with regular jogging,cycling,or swimming showed lower prevalence of KOA and knee pain than those without.With the increase of workload,the prevalence of KOA and knee pain significantly increased.Compared with participants in sitting work position(6.5% and 36.7%,respectively),the prevalence of KOA and knee pain were much higher among those in standing(7.1% and 40.7%,respectively)or bending position(6.6% and 41.1%,respectively).Shift workers showed significant higher prevalence of KOA(7.0%)and knee pain(41.6%)than daytime workers(6.4% and 37.3%,respectively).The results from multivariable logistic regression analyses suggested that,female gender(OR 2.40,95% CI 1.90-3.03),elevated age(1.04,1.03-1.05),central obesity(1.47,1.26-1.71),and alcohol drinking(1.26,1.09-1.55)were risk factors for KOA.While female gender(2.34,2.08-2.62),aging(1.02,1.01-1.03),central obesity(1.30,1.21-1.41),alcohol drinking(1.22,1.10-1.35),elevated workload(ORs for moderate,heavy,and severe heavy workload were 0.99,1.17,and 1.22),and shift work(1.17,1.08-1.28)were risk factors for knee pain.Conclusions: Females,older adults,obese individuals,drinkers,heavy labor workers,and shift workers were more susceptible to KOA and knee pain.Female gender,aging,central obesity,alcohol consumption were potential risk factors for KOA and knee pain,and heavy workload and shift work were potential risk factors for knee pain.Part Two: The Cross-Sectional and Longitudinal Associations between Hyperlipidemia and Knee OsteoarthritisObjectives: To evaluate the effect of hyperlipidemia on KOA and knee pain in both cross-sectional and longitudinal study.Methods: The study population in part one were included and were followed-up 5 years later.Questionnaire investigations and physical examinations were conducted in both cross-sectional and longitudinal researches.Hyperlipidemia was diagnosed with fasting serum triglyceride(TG)≥1.70 mmol/L,or fasting serum total cholesterol(TC)≥5.2 mmol/L,or clinically diagnosed hyperlipidemia.In the cross-sectional analyses,logistic regression models were used to calculate the ORs and 95% CIs for prevalent KOA and knee pain associated with hyperlipidemia.Each continuous lipid was included in the logistic model to estimate the linear association between each lipid levels and the risk of KOA or knee pain.Furthermore,logistic model with cubic smoothing spline was used to test the nonlinearity between each lipid levels and prevalent KOA or knee pain.Stratified analyses by potential confounding factors were further conducted.Relative excess risk due to interactions(RERIs)were used to estimate the additive interactions between hyperlipidemia and other factors on KOA and knee pain.In the longitudinal analyses,COX regression models were used to calculate the hazard ratios(HRs)for incident KOA and knee pain associated with hyperlipidemia.Continuous lipid levels and cubic smoothing splines were introduced into the COX models to assess the linear and nonlinear associations between each lipid and the risks of incident KOA and knee pain,respectively.Stratified analyses and interaction analyses were also conducted in COX models.Results: A total of 13906 participants were included in the cross-sectional study.The ORs for KOA and knee pain among hyperlipidemia cases were 1.35(95% CI 1.16-1.56)and 1.31(95% CI 1.21-1.43),respectively,compared with their unaffected colleagues.Compared with participants without hyperlipidemia and no lipid-lowering drugs use,the ORs for KOA and knee pain in those with hyperlipidemia but no lipidlowering drugs use were 1.24(1.01-1.52)and 1.28(1.14-1.42),respectively;the ORs in those with hyperlipidemia and lipid-lowering drugs use were 1.46(1.21-1.76)and 1.36(1.22-1.52),respectively;but the ORs were insignificant among those without hyperlipidemia but use lipid-lowering drugs(KOA: 1.09,0.77-1.55;knee pain: 1.04,0.86-1.26).Each unit increase in TG was associated with 8% and 4% increases in the risks of prevalent KOA and knee pain,respectively.Each unit increase in TC and low density lipoprotein cholesterol(LDLC)was associated with 9% and 7% increases in the risk of prevalent knee pain,respectively.High density lipoprotein cholesterol(HDLC)was non-linearly associated with the decreased risk of prevalent knee pain,with the non-linear P trend <0.001.Further stratified analyses indicated that the associations between hyperlipidemia and the risks of prevalent KOA and knee pain were significant in each subgroup,regardless of females or males,young or old individuals,obese or non-obese individuals,drinkers or non-drinkers,participants with different workload,shift workers or daytime workers.But the associations between hyperlipidemia and KOA and knee pain were more significant in women than in men.The results of interaction analyses demonstrated that hyperlipidemia showed additive interactions with female gender on KOA and knee pain,with the RERIs of 0.81(0.11-1.50)and 0.75(0.37-1.13),respectively.After excluding participants with KOA or knee pain at baseline,a total of 7901 and 6613 participants were included in the longitudinal analyses for incident KOA and knee pain,respectively.Participants with hyperlipidemia showed 1.52 times(95% CI 1.18-1.96)the risk of incident KOA and 1.21 times(95% CI 1.08-1.36)the risk of incident knee pain,compared with those without.When compared with participants without hyperlipidemia and no lipid-lowing drugs use,the significantly increased risks of incident KOA(HR 1.74,95% CI 1.27-2.40)and knee pain(HR 1.21,95% CI 1.03-1.41)were detected among those with hyperlipidemia and lipid-lowing drugs use,but no significant increases were found among those without hyperlipidemia but use lipidlowering drugs(HR for KOA: 1.68(0.95-2.98),HR for knee pain: 1.08(0.80-1.45)).The incidence of KOA and knee pain slightly increased with the elevated TG and LDLC levels,while slightly deceased with the elevated HDLC levels.Each unit increase in TC was associated with a 6% increase in the risk of incident knee pain.In the stratified analyses,the incidence of KOA and knee pain were significantly higher among hyperlipidemia cases than their unaffected colleagues,for both females and males,both young and old individuals,both obese and non-obese participants,both drinkers and non-drinkers,participants with different work loads,both shift workers and daytime workers.The associations between hyperlipidemia and the risks of incident KOA and knee pain were more significant in females than in males.The interaction analyses indicated that hyperlipidemia and female gender showed slightly additive interactions on incident KOA and knee pain,with the RERIs of 0.75 and 0.25,respectively.Conclusions: Hyperlipidemia significantly increased the risk of prevalent and incident KOA and knee pain among middle-aged and older adults.The cross-sectional and longitudinal dose-response relationships were detected between TC and the risks of KOA and knee pain.Gender,age,lifestyles,or occupation activities did not change the association between hyperlipidemia and KOA or knee pain.Part Three: The Role of Nalp3 Inflammasome in Hyperlipidemia Induced Knee Osteoarthritis in MiceObjectives: To explore the possible role of Nalp3 inflammasome in hyperlipidemia induced KOA in mice.Methods: Two experiments were performed.Experiment 1,in which mice were fed with a high cholesterol diet(HCD)for 6 months to prepare hyperlipidemia induced KOA model.A total of 48 C57BL/6 mice were randomized into HCD group,normal diet(ND)group,HCD with wheel running group,and ND with wheel running group.The fasting serum TG,TC and LDLC levels were measured;the total knee joints were harvested for Safranin-O and fast green staining to assess osteoarthritis progression;and the cartilage and synovium were harvested for measurement of MMP13 m RNA relative levels.Experiment 2,in which mice were intraperitoneal injected with caspase-1 inhibitor Ac-YVAD-CMK to prepare Nalp3 inflammasome blocking model.A total of 48 C57BL/6 mice were randomized into HCD group,ND group,HCD with Ac-YVADCMK group,and ND with Ac-YVAD-CMK.The fasting serum lipids levels,serum IL-1β levels,and cartilage and synovium m RNAs relative levels were measured.The knee joint tissue sections were made for Safranin-O and fast green staining and immunohistochemistry.Results: The highest TG,TC,and LDLC levels were detected in HCD group(mean concentration: 1.12,5.62 and 1.04 mmol/L),followed by HCD with wheel running group(1.11,4.50 and 0.68 mmol/L),and the lowest levels were found in NC with wheel running group(0.89,2.29 and 0.17 mmol/L).Positive dose-response relationship between lipids levels and the progress of KOA was detected.The histological slides suggested that HCD group presented superficial cartilage matrix loss and chondrocytes death,HCD with wheel running group showed cartilage surface discontinuity,and the cartilage in ND group and ND with wheel running group remained intact.The OARSI scores were highest in HCD group(3.6),were secondhighest in HCD with wheel running group(1.0),and were lowest in NC with wheel running group(0.1).Those results indicated successful preparation of hyperlipidemia induced KOA mice model.Compared with ND group,HCD group showed significantly increased serum IL-1β levels(226.4 vs 77.2 pg/ml,P=0.05),cartilage and synovium caspase-1 m RNA relative levels(1.66 vs 1.08,P=0.01),cartilage and synovium IL-1β m RNA relative levels(2.07 vs 1.08,P=0.006),and cartilage caspase-1 and IL-1β protein levels.It suggested that Nalp3 inflammasome might play a role in hyperlipidemia induced KOA.Blocking caspase-1 did not stop lipids increasing,but significantly reduced the IL-1β levels.Compared with HCD group,HCD with Ac-YVAD-CMK group showed significantly decreased serum IL-1β levels(103.6 vs 226.4 pg/ml,P=0.05),cartilage and synovium IL-1β m RNA relative levels(1.42 vs 2.07,P=0.09),cartilage and synovium IL-1β protein levels,and significantly decreased OARSI scores(0.3 vs 3.6,P=0.003).The results suggested that blocking caspase-1 slowed down the progression of KOA induced by hyperlipidemia.Conclusions: Hyperlipidemia enhanced the stimulation of Nalp3 inflammasome levels in knee tissues.Blocking the function of Nalp3 inflammasome could to some point protect cartilage from degradation.It might be possible that Nalp3 inflammasome positively regulate the KOA damage induced by hyperlipidemia.