Soluble Tim3 detection by time‐resolved fluorescence immunoassay and its application in membranous nephropathy

Abstract Background We aimed to develop a time‐resolved fluorescence immunoassay (TRFIA) for detecting soluble T‐cell immunoglobulin and mucin domain 3 (sTim3) in serum samples and to demonstrate a preliminary application of this method in membranous nephropathy (MN). Methods sTim3 TRFIA was developed, and the sTim3 concentration in the serum of patients with MN and healthy individuals was detected using a sandwich method. Results The sensitivity of the developed sTim3 TRFIA was 0.66 ng/mL, higher than that of an enzyme‐linked immunosorbent assay (ELISA) (1.11 ng/mL). The detection range was 0.66‐40 ng/mL. The intra‐assay coefficient of variation (CV) for sTim3 was 1.64%‐4.68%, and the inter‐assay CV was 5.72%‐9.32%. The cross‐reactivity to interleukin 6 (IL‐6) and kidney injury molecule 1 (KIM‐1) was 0.25% and 0.04%, respectively. The average recovery was 105.26%. The sTim3 concentration in patients with MN was considerably higher than that in healthy individuals (P < .001). The sTim3 concentration in the serum of patients with MN was significantly increased from G1 to G4 based on the Jonckheere‐Terpstra test (P < .001). Thus, we used sTim3 as a diagnostic indicator for distinguishing between healthy individuals and patients with MN as well as between different stages of MN. Conclusion We successfully established TRFIA to detect sTim3 in serum. We then applied this method to patients with MN, demonstrating for the first time that TRFIA is a valid diagnostic tool to detect sTim3 in serum.


| INTRODUC TI ON
T-cell immunoglobulin and mucin domain 3 (Tim3) is a type 1 transmembrane receptor that can inhibit the activation of T helper 1 (Th1) cells after ligand binding. 1 As a member of the Tim family, Tim3 was originally discovered on activated Th1 and Th17 cells, as well as T cell 1 (Tc1), where it induced T-cell exhaustion or death after binding to galectin-9 (gal-9), 2 partly through the T-cell receptor (TCR) signaling pathway. 3 Tim3 also plays an important role in adjusting the activity of monocytes, endothelial cells, macrophages, mast cells, 4 natural killer cells, and dendritic cells. Though the overall mechanisms are still unclear, an imbalance of the Tim3 expression in innate immune cells may result in excessive or inhibited inflammatory reactions, which can contribute to an autoimmune response or evasion of viruses or tumor cells. 2 Tim3 has a soluble form (sTim3) and a membrane-bound form. sTim3 can be obtained by shearing with the help of enzymes. 5 Möller-Hackbarth et al 6 identified a disintegrin and metalloprotease 10 (ADAM10) and ADAM17 as major sheddases of Tim3. It was also confirmed that another source of sTim3 is spleen cells. sTim3 mRNA is only found in spleen cells, and the activation of spleen cells leads to upregulation of sTim3. 7 Xiao et al 8 found that sTim3 and soluble programmed cell death protein 1 (sPD-1) can enhance the cell-mediated immune response induced by the simian immunodeficiency virus (SIV) vaccine in mice. High expression of membrane-bound Tim3 was observed in common autoimmune diseases, like IgA nephropathy 9 and systemic lupus erythematosus. These studies demonstrate that Tim3 is involved in pathogenesis; thus, Tim3 may be a potential therapeutic target for these diseases. 10 Membranous nephropathy (MN) is a serious autoimmune disease and one of the main causes of adult nephrotic syndrome, which can lead to endstage renal failure. 11 In addition, the role of sTim3 in MN has not been thoroughly studied.
In previous research, an enzyme-linked immunosorbent assay (ELISA) was used to detect sTim3 in patient serum. 5,12 No other detection methods for sTim3 have been reported except for ELISA.
Because ELISA uses enzymes to label the antibody or antigen, it greatly affects the biological activity of the labeled compounds. In addition, many false-positive results are obtained due to the low sensitivity of ELISA. Insufficient stability and detection range are also common disadvantages of this method. 13 Time-resolved fluorescence immunoassay (TRFIA) is an alternative and non-isotope immunoassay that uses lanthanide elements to label antigens or antibodies. It has the advantages of high sensitivity, good stability, and wide detection range compared with other detection methods.
Therefore, this successful method of detection is widely used in the biomedical field. [14][15][16] In this research, we aimed to establish a valid method to detect sTim3 and to evaluate clinical value of the method in detecting sTim3 in MN.

| Serum samples
The sera of 47 healthy individuals and 54 patients with MN were obtained from Wuxi People's Hospital affiliated to Nanjing Medical University. Blood samples were centrifuged at 1200 gfor 5 minutes to prepare the serum, and the serum samples were stored at −20°C until use. The serum samples did not undergo hemolysis or lipemia during the preparation process. The inclusion criteria for healthy subjects were that they did not have a history of kidney disease, autoimmune disorders, or infections with viruses such as the human immunodeficiency virus, hepatitis B virus, and hepatitis C virus. All registered subjects provided written, informed consent to participate in this research. The research program was approved by Wuxi People's Hospital affiliated to Nanjing Medical University.

| Coating of the microwell plates
The original concentration of the Tim3 antibody (Tim3-R024, capture antibody) was 1 mg/mL. First, it was diluted with coating buffer to 3 μg/mL. Then, 100 μL of 3 μg/mL Tim3 antibody was added into each well of a 96-well microtiter plate, and the plate was sealed and incubated overnight at 4°C. Next, the plate was washed once, after which each well was blocked with 150 μL of blocking buffer containing 3 g/L BSA at room temperature for 2 hours. Finally, the blocking buffer was decanted, and the plates were dried under vacuum, followed by storage in a sealed plastic bag with a desiccant at −20°C.

| Antibody labeling
Tim3 antibody labeling was conducted based on the manufacturer's instructions. First, 300 μL of the Tim3 antibody (Tim3-MM04, detection antibody) was transferred to an ultrafiltration tube and centrifuged at 7100 g for 8 minutes. Then, 200 μL of labeling buffer was added, and the tube was centrifuged at 7100 g for 8 minutes; this step was repeated seven times. Next, 50 μL of labeling buffer was added to the ultrafiltration tube, and the tube was centrifuged at 1000 g for 1 minute; this step was repeated twice. The antibody was mixed with the 60 μg of diethylenetriaminetriacetic acid (DTTA)-Eu 3+ and incubated overnight at 30°C with shaking. The labeled antibody was purified by Sephadex-G50, and the purified labeled antibody was stored in elution buffer containing 0.2% BSA.
Finally, the antibody was collected and stored at −20°C.

| Assay protocol
Tim3 standards (100 ng/mL) were diluted with standard buffer containing 0.5% BSA to final concentrations of 0, 5, 10, 20, and 40 ng/ mL. Then, TRFIA of sTim3 was conducted using a two-step, noncompetitive "sandwich-type" technique. A two-step method was used because the serum contained EDTA. First, 20 μL each of the standard and serum samples was added into a 96-well microtiter plate. Next, 80 μL of assay buffer was added into each well of the coated microtiter plate. The plate was incubated at 37°C with gentle shaking for 1 hour. After the incubation, the plate was washed twice with washing buffer, and 100 μL of 200-fold diluted Eu 3+ -labeled Tim3-MM04 solution was then added. Finally, the plate was incubated again at 37°C with gentle shaking for 2 hours. After six washes, 100 μL of enhancement solution was added. The plate was gently shaken for 3 minutes, and the fluorescence was then measured. The sTim3 concentration of the serum sample was calculated based on the Tim3 standard curve.

| Sensitivity and linearity
The standard curve of Tim3 consisted of different concentrations of standards (0, 5, 10, 20, and 40 ng/mL). Linear regression analysis was performed, and the equation and the coefficient of determination, R 2 , are shown as indicators of linearity. The background fluorescence intensity of 10 replicates of the 0 ng/mL standard was measured, and the sensitivity was calculated as the mean + 2SD. 18

| Precision
The precision of the method was assessed using three serum samples from patients with MN that measured low, medium, and high sTim3 concentrations. To obtain intra-assay and inter-assay precision, 10 independent experiments were conducted for each of the three samples.

| Specificity
The specificity was assessed using kidney injury molecule-1 (KIM-1) and interleukin 6 (IL-6) as potential interfering antigens. The presence of cross-reactions was determined by detecting the final fluorescence value in the presence of the antigen. Ten independent experiments were performed for each antigen.

| Recovery
The recovery was assessed by adding 100 ng/mL Tim3 standards to three serum samples with known concentrations of sTim3 at a ratio

| Method comparison
To assess the correlation between TRFIA and ELISA, the sTim3 con-

| Statistical analysis
The data were expressed as the mean ± SD. SPSS16.0 (Chicago, IL, USA) was used to analyze the data, and Prism 7.00 (San Diego, CA, USA) was used to prepare the graphs. Spearman correlation coefficients were used to assess whether the sTim3 concentration correlated with various parameters. sTim3 concentrations and various clinical parameters at different GFRs were analyzed using the Jonckheere-Terpstra test. P values <.05 were considered statistically significant.

| Assay evaluation
The sensitivity and linearity results are presented in Figure 1. The detection range for sTim3 was 0.66-40 ng/mL. The R 2 of the Tim3 standard curve was .9998, indicating a good linear relationship between the sTim3 concentration and the fluorescence intensity. The sensitivity of TRFIA was 0.66 ng/mL, which was higher than that of ELISA (1.11 ng/mL). The precision was measured by the coefficient of variation (CV). The intra-assay CV for sTim3 varied between 1.64% and 4.68%, and the inter-assay CV varied between 5.72% and 9.32% (Table 1). All CV values were less than 10%, suggesting that the precision of the experiment was acceptable. Table 2 shows that the cross-reactivities of IL-6 and KIM-1 were 0.25% and 0.04%, respectively. The extremely low cross-reactivity observed showed that TRFIA had high specificity for sTim3. As shown in Table 3, the average recovery was 105.26%. The results showed that the experiment was free from interferences in serum.

| Correlation with ELISA
Thirty-six clinical serum samples were analyzed by ELISA and the newly established TRFIA (Figure 2). The sTim3 concentrations detected by the two methods were highly similar, as indicated by the Note: Data were presented as mean ± SD. sTim3 concentration and various clinical parameters in different GFR stages were analyzed by Jonckheere-Terpstra test. P < .05 was considered significant.
F I G U R E 3 sTim3 concentration in different GFR stages of patients with MN (P < .001), Jonckheere-Terpstra test was used to analyze P value, and P < .05 was considered significant F I G U R E 4 sTim3 concentration in patients with MN (n = 54) and normal control (n = 47, P < .001), t test was used to analyze P value, P < .05 was considered significant linear fit with the equation y = 0.9687x + 1.345 and R 2 of .9892 (P < .001). The results demonstrated that TRFIA is a reliable technique for the detection of sTim3 in clinical serum samples.

| Clinical application
We investigated the sTim3 concentration in various stages of MN.
Age, systolic blood pressure, urea, creatinine, urinary protein, and osmotic pressure were noticeably increased from G1 to G4, based on the Jonckheere-Terpstra test ( Table 4). The sTim3 concentration in the serum of patients with MN was also significantly increased from G1 to G4 based on the Jonckheere-Terpstra test (P < .001, Figure 3 and Table 4).
As shown in Figure 4, the sTim3 concentration in patients was 38.5 ± 34.8 ng/mL (n = 54), which was significantly higher than that in healthy people (11.1 ± 3.3 ng/mL, n = 47, P < .001). Receiver operating characteristic (ROC) analysis of the sTim3 concentration in patients with MN is presented in Figure 5 and  the roles of Tim3 and sTim3 in pathogenesis have not been fully revealed.

| D ISCUSS I ON
Currently, commercial kits for detecting sTim3 are limited to the ELISA format. Although ELISAs are very popular detection methods in the biomedical field, the ELISA platform has many disadvantages.
For example, enzyme-labeled compounds are unstable, and labeled enzymes are susceptible to inactivation by the environment. 35 In addition, the labeling process is complicated and requires highly skilled personnel. Furthermore, the detection process is complex and time-consuming. Compared with ELISA, TRFIA has minimal effect on biological activity because of its atomic labeling. Moreover, TRFIA has higher sensitivity than other methods, with a detection limit of 10 −18 mol/L, which is markedly improved compared with that of ELISA (10 −9 mol/L) and chemiluminescence (10 −15 mol/L). 13 In the present study, TRFIA also showed higher sensitivity than ELISA (0.66 ng/mL infections. Therefore, we predict that TRFIA will be applied to the detection of additional immune checkpoint proteins in serum.
In conclusion, we successfully established TRFIA for the detection of sTim3. We also found that the sTim3 concentration was significantly correlated with renal function indicators in patients with MN, which can distinguish patients with MN and help better diagnosis.