Correlation between serum cathepsin S and insulin resistance in type 2 diabetes

Cathepsin S (CatS), a proteolytic enzyme, which belongs to the cysteine proteinase family, is associated with atherosclerosis, coronary heart disease, cancer and other diseases. The present study aimed to explore the correlation between serum CatS and insulin resistance (IR) in patients with type 2 diabetes. A total of 51 patients with type 2 diabetes (Group DM) were recruited for this study and 49 healthy individuals were selected as normal controls (Group NC). Blood pressure and body mass index (BMI) were recorded, and serum creatinine, CatS, glycosylated hemoglobin (HbA1c), lipid and insulin levels, and fasting plasma glucose (FPG) levels were measured in all the participants. The homeostatic model assessment index of IR (HOMA-IR) was calculated according to FPG and serum insulin levels. Serum CatS, very low density lipoprotein (VLDL) and triglyceride (TG) levels in Group DM were significantly higher compared with those in Group NC (P=0.000, 0.014 and 0.020, respectively). Significantly positive correlations were identified between CatS levels and VLDL and TG levels, respectively (P<0.05 for both); however, no significant correlations were determined between CatS levels and age, course of disease, blood pressure, cholesterol, BMI, FPG, HbAc1 and HOMA-IR (P>0.05). Further stratification analysis showed that CatS had no association with IR at different HOMA-IR and HbA1c levels. The present study demonstrated that serum CatS, which was significantly increased in patients with type 2 diabetes, had no correlation with IR. This indicates that CatS and IR are independent of each other; however, the precise mechanisms require further investigation.


Introduction
Insulin resistance (IR) is a reduction in reaction or sensitivity to insulin and was first presented by Reaven in 1988 (1). IR is considered to be the common cause of impaired glucose tolerance, diabetes, obesity, dyslipidemia and hypertensive diseases. IR syndrome is associated with multiple metabolic disorders and was renamed metabolic syndrome (MS) by Zimmet et al in 1997 (2). A previous study (3) identified that MS is closely associated with atherosclerosis (AS) and numerous factors, including cytokines, are involved in AS formation. Cathepsin S (CatS) is a cysteine protease, an important lysosomal protease and the predominant cathepsin. CatS is released from primary cultured microglia and regulated by the P2X7 receptor (4), which is associated with matrix metalloproteinases and serine protease, and involved in extracellular proteolytic degradation. In addition, CatS has important physiological functions in the extracellular environment, including degradation of the extracellular matrix (ECM), regulation of growth factors, vascular proliferation of various cytokines, and cell migration, proliferation and apoptosis. Sukhova et al (5) demonstrated that CatS expression is increased in atherosclerotic sites. Furthermore, studies have shown that CatS is important for the formation and development of AS (6)(7)(8)(9). Type 2 diabetes is a polygenic, hereditary, metabolic disease and the main pathophysiological mechanisms are IR and insufficient insulin secretion.
Since CatS, MS and AS are closely associated with each other, IR is one of the most important pathophysiological mechanisms of type 2 diabetes and AS is one of the pathological features of diabetic vascular complications. However, whether CatS is involved in the pathophysiological mechanisms of type 2 diabetes is yet to be elucidated. A previous study demonstrated that the increased expression of CatS was in parallel with an increase in proinflammatory cytokine levels during the development of autoimmune inflammatory diseases in the non-obese diabetic (NOD) mouse disease model (10). The present study aimed to compare the levels of CatS between patients with type 2 diabetes and healthy individuals, in order to investigate the changes in CatS expression and the possible correlation between CatS and IR.
cided with the 1999 World Health Organization standards for diabetes diagnosis and classification (11). Exclusion criteria comprised patients who had heart, liver or pancreatic diseases; primary or secondary renal diseases (with the exception of diabetic nephropathy); inflammatory tumor or immunity diseases; impaired glucose tolerance induced by glucocorticoid and steroidal agents; and acute complications or serious chronic complications of diabetes mellitus, including ketoacidosis, hyperosmolar coma and lactic acidosis. Fortynine healthy individuals without diabetes, infection, tumors and autoimmune diseases were chosen as the normal controls (Group NC; 25 males and 24 females). A total of 51 diabetic patients in DM group included 26 males and 25 females. The study was approved by the local research ethics committee (Anhui Provincial Hospital, Anhui, China) and written informed consent was obtained from all patients before the study commenced.
Methods. The gender, age, height, body weight, systolic pressure and diastolic pressure of the patients were recorded, and the body mass index (BMI) and the homeostatic model assessment index of IR (HOMA-IR) were calculated in all participants. Venous blood samples were collected to measure serum creatinine, cholesterol (TCH), triglyceride (TG), very low density lipoprotein (VLDL), fasting plasma glucose (FPG), insulin and glycosylated hemoglobin A1c (HbA1c) levels following >8 h of overnight fasting. Simultaneously, a serum sample was collected and preserved at -80˚C to test for the expression of CatS.
Serum CatS levels were determined using a sandwich enzyme immunoassay (ELISA; Krka, d.d., Novo mesto, Riga, Slovenia; Riga, Slovenia; Human Cathepsin S) according to the manufacturer's instructions. Serum samples were diluted 1:1 with 0.9% physiological saline, and then added to the pre-coated ELISA plate (100 µl per well). Plates were incubated Age (

Results
Comparison of the clinical data between Group NC and Group DM. As shown in Table I, a comparison of the clinical data between Group NC and Group DM indicated no significant differences in age, and serum creatinine and TCH levels. However, the CatS, FPG, VLDL and TG levels in Group DM were significantly higher than those in Group NC (P= 0.000, 0.000, 0.014 and 0.020, respectively).

Correlation between serum CatS levels and various clinical indices.
Changes in the serum levels of CatS were significantly associated with TG and VLDL levels (P<0.05 for both); however, no significant correlations were identified between serum CatS levels and age, disease course, blood pressure, TCH, BMI, FPG, Cr, HbAc1 and HOMA-IR (Table II).

Correlation between serum CatS levels and clinical data in
Group DM at different HOMA-IR levels. The third quartile

Discussion
CatS was first identified in hematopoietic cell endosomes, which belong to the non-glycosylated papain-like single-chain protease family, and demonstrates proteolytic and endopeptidase effects. Numerous studies have shown that CatS is associated with AS, coronary heart disease and cancer. However, few studies have identified a correlation between CatS and diabetes. In animal models, deletion of the CatS gene resulted in immunity to type 1 diabetes mellitus in certain NOD mice (12). The present study indicated that CatS levels were significantly higher in patients with type 2 diabetes than those in the normal controls, and this is consistent with the findings of Liu et al (13). The current study suggests that CatS may be involved in the occurrence and development of diabetes. IR is a one of the key pathophysiological mechanisms of type 2 diabetes and serum CatS levels are elevated in patients with type 2 diabetes. Therefore, it is necessary to investigate the association between serum CatS concentration and IR in patients with type 2 diabetes. The hyperinsulinemic-euglycemic clamp test is the gold standard for evaluating IR. In addition, BMI, waist circumference, waist to hip ratio, fasting insulin, fasting blood glucose, Shima Soku used the insulin action index (IAI) and HOMA-IR, may also be used as auxiliary indices for evaluating IR. The present study selected BMI and HOMA-IR as the indices of IR. Through hierarchical analysis, no significant correlations between serum CatS and BMI and HOMA-IR were detected. This suggests that HOMA-IR, serum CatS and IR are independent of each other, and that CatS may be involved in the occurrence and development of diabetes as an independent factor. Due to reductions in lipoprotein lipase activity, the TG levels readily increase and VLDL clearance decreases after eating in patients with diabetes, and the accumulation of VLDL increases. Increases in TG levels in the skeletal muscle are also considered to be one of the causes of IR (14). The present study identified that serum CatS levels were significantly associated with TG and VLDL levels when the HbA1c level was >9% or the BMI was <25 kg/m 2 , respectively. The exact mechanisms responsible for this remain unclear, but may be related to postprandial glucose. In addition, it has been indicated that CatS promotes fat formation, particularly in the early stages of adipocyte differentiation, which suggests that CatS may be involved in the pathophysiological mechanisms of lipid metabolism.
The duration of diabetes (disease course) is correlated with decreased B cell function. The present study demonstrates a significant correlation between serum CatS and diabetes course when the HbA1c level is >9% or the BMI is <25 kg/m 2 . However, the precise mechanisms require further investigation.
It has been demonstrated that patients with type 2 diabetes have a low-grade inflammatory state in vivo (15). Inflammation may cause IR and is involved in associated diseases, including diabetes, metabolic syndrome and AS. Jobs et al (16) identified that serum CatS levels were associated with C-reactive protein, which is an important indicator of inflammation. In addition, Naour et al (17) indicated that CatS may also be involved in the energy balance in adipose tissue and circulation. These studies suggest that serum CatS may be independent of IR, and involved in the occurrence and development of diabetes through certain pathways, including the inflammatory response.
A limitation of the present study is that the sample size was relatively low, which may have affected the HOMA-IR results. In addition, more precise methods for the assessment of IR, including glucose clamp experiments, could have been used. In addition, the related inflammatory indices and postprandial blood glucose levels require analysis In conclusion, the present study demonstrates that serum CatS levels are significantly increased in patients with type 2 diabetes, which reflects their inflammatory state and shows no significant correlation with IR. The precise mechanisms responsible for this require further investigation.