A quantitative detection of Cardiotrophin‐1 in chronic heart failure by chemiluminescence immunoassay

Abstract Background Cardiotrophin‐1 (CT‐1) is a cytokine that could induce cardiomyocytes hypertrophy and dysfunction. Plasma CT‐1 might serve as a cardiac biomarker both in diagnosis, staging, and prognostic assessment of heart failure. Methods In this study, a one‐step paramagnetic particles‐based chemiluminescence immunoassay (MPs‐CILA) for rapid and sensitive detection of plasma CT‐1 was established. Plasma samples were directly incubated with biotin‐labeled anti‐CT‐1 antibody (bio‐Ab) and acridine ester labeled anti‐CT‐1 antibody (AE‐Ab) to form sandwiched complex. The sandwiched CT‐1 was then captured by streptavidin modified paramagnetic particles (MPs‐SA) for rapid separation and signal generation. Results The proposed MPs‐CLIA presents a laudable linear relationship ranging from 7.8 pg/mL to 200 ng/mL with a detection limit of 1.0 pg/mL. The recoveries of spiked human plasma samples at low (10pg/mL), medium (100 pg/mL), and high (800 pg/mL) levels of CT‐1 were 96%, 104%, and 110% respectively. The intra‐analysis coefficient variation (CVs) of the 3 samples was 8.92%, 6.69%, and 3.54%, respectively. And the inter‐analysis coefficient variation (CVs) was 9.25%, 10.9%, and 4.3%, respectively. These results strongly indicate high sensitivity, wide linear range, acceptable precision, and applicable reproducibility of the proposed method to detect plasma level of CT‐1. Finally, Plasma CT‐1 from 140 subjects with or without chronic heart failure was analyzed to assess the clinical application of MPs‐CILA. Conclusions Noteworthily, the MPs‐CLIA method is highly automated such that it is suitable for high‐throughput detection of CT‐1 in clinical inspection.


| INTRODUC TI ON
Chronic heart failure (CHF) is a global public health concern greatly impacting patient's quality of life, and inflicting significant morbidity and mortality. 1 As CHF is often characterized by obscure symptoms, associating with a variety of pathophysiological and biochemical diseases, its early diagnosis is challenging. 2,3 BNP or pro-BNP has identified as an objective indicator for the diagnosis of heart failure and is currently recommended by the European Society of Cardiology (ESC) guidelines for CHF diagnosis. 4 However, BNP/pro-BNP significantly varies regarding non-cardiac factors, such as age, gender, race, renal function result in low specificity for CHF diagnosis. [5][6][7][8][9] Cardiotrophin-1 (CT-1) is a member of interleukin-6 superfamily with molecular mass of 21.5KD. 10 CT-1 protein is abundantly expressed in heart tissue, its overexpression is mainly stimulated by ventricular stretch/pressure, 11 which promotes cardiac hypertrophy 12 and myocardial fibrosis 13,14 via binding to gp130/LIF complex, 15,16 eventually participates the progression of chronic heart failure. 13 It is predominantly secreted by myocardium cell through coronary sinus into the peripheral circulation. 17 A plethora of studies present that CT-1 level in heart and plasma were significantly elevated in CHF patients. [18][19][20] Tsutamoto et al 21 have demonstrated that plasma levels of CT-1 increased with the severity of CHF, and high plasma levels of CT-1 associated with high mortality in CHF patients. Moreover, compared with BNP/pro-BNP, CT-1 is independent of age, gender, BMI, and renal function. 18 Considering the importance of CT-1 to the presence and severity of CHF, developing an automated detection platform for accurately and rapidly quantifying CT-1 at low concentration possesses a great value.
Thus far, various analytical methods have been reported to be used detect CT-1. In 1999, Talwar et al 22 developed a competitive immunoluminometric assay, firstly quantitative assessment of plasma CT-1 in humans. The method requires a pre-analysis step as long as 24 hours, and due to the extra extraction step, its specificity is lower and it is more prone to errors. Some methods proposed later were abandoned due to complicated operation process, long detection period, and high detection cost. 17,21,23 Currently, chemiluminescence immunoassay is increasingly used in ultramicro-analysis of biological substances due to the advantages of being extreme sensitivity, high specificity, good reproductivity, and simplicity, which has basically replaced radioactive immunoassay and enzyme immunoassay technology. 24 In this study, we constructed a paramagnetic particles-based chemiluminescence immunoassay (MPs-CLIA) for rapid determination of cardiotrophin-1 in plasma. We adopt streptavidin-coated paramagnetic particles (MPs-SA) in combination with the capture antibody (biotin-labeled anti-CT-1 antibody, bio-Ab) to separate the immune complexes from the complex matrix, which considerably improves the efficiency of the separation and washing steps, and provides a simplified procedure. [25][26][27] After separation by magnetic separation column, the immune-complexes were recovered for the quantitative CT-1. In the immunoassay scheme, chemiluminescence signal produced by an anti-human free CT-1 antibody labeled with acridinium ester (AE-Ab) was directly proportional to the amount of CT-1 in a sample. The chemiluminescent acridinium ester labels have excellent chemiluminescent properties such as low background signal, high detection sensitivity and no need for a catalyst and thus efficiently simplify the detection procedure. 28 2 | MATERIAL AND ME THODS

| Apparatus and reagents
A photomultiplier instrument for chemiluminescence signal detection was developed by our laboratory. Electrochemiluminescent Ltd (USA). Hydrogen peroxide/sodium hydroxide reagent was prepared by ourselves. Washing buffer was made by PBS containing 0.1% Tween-20 and 0.1% BSA. All Double distilled water was prepared using water purified with an ultrapure water system (Zhejiang university second affiliated hospital). All of the other chemicals were standard commercial products of analytical-reagent grade.

| Patient recruitment
To verify the clinical applicability of our method, 100 patients hospitalized with chronic CHF (72 men and 28 women) aged between 22 and 84 years (mean 60) were recruited at our institutions and quantified CT-1 from these samples by MPs-CLIA ( Table 1). The Collected plasma using EDTA or heparin as an anticoagulant, plasma was centrifuged for 10 minutes at 3000 × g, store samples in aliquot at −80°C for later use. Avoid repeated freeze/thaw cycles.

| Preparation of Biotin-labeled anti-human free CT-1 antibody (bio-Ab)
Preparation of bio-Ab was conducted through the reaction between N-hydroxysuccinimide biotin and anti-human free CT-1 antibody. As The purified bio-Abs solution containing 0.02% sodium azide and 0.25% BSA was stored at −20°C.

| Preparation of acridinium ester labeled antihuman free CT-1 antibody (AE-Ab)
Briefly, anti-human free CT-1 antibody solution was diluted to 0.16 mg/mL by 0.1 M PBS (PH:7.4). Then, the antibody solution was reacted with 10 μL acridinium ester solution (0.5mM) previously dissolved in anhydrous dimethyl formamide (DMF). The reaction solution was mixed well at room temperature for 30min. The AE-Ab solution was also purified by sephadex G-75 Chromatography column. The purified AE-Abs solution containing 0.02% sodium azide and 0.25% BSA was stored at −20℃.

| MPs-CLIA procedure
MPs-CLIA was based on double antibodies, which specifically recognizes CT-1. As is shown in Figure 1, a "one-step" reaction, 50 μL bio-Ab (capture antibody) (1.6 μg/ mL) in 5% BSA-PBS, 50 μL AE -Ab was measured with a photomultiplier instrument to quantify CT-1 concentration. Figure 1 shows a schematic diagram of the experimental procedure.

| Statistical analysis
Continuous variables are presented as mean ± standard deviation (SD) or median ± inter-quartile range (IQR), and significant differences were

| Linear curve
Recombinant human CT-1 in the form of freeze-dried powder was dissolved with sterile PBS conducted by the manufacturer protocol. The serially diluted calibration samples in human plasma matrix (7.8, 15.6, 31.25, 62.5, 125, 500, 1000 pg/mL) were measured using MPs-CLIA.
Under the optimal conditions of exploration, standard curve obtained was RLU = 32.646C CT1 + 174.35 with linear range of 7.8pg/mL-200ng/ mL and a good determination coefficient of 0.9996 (Figure 3). This range of concentrations quantification is far greater than the pathological level of CT-1, which enough provides clinically useful and sensitive information without pre-dilution of specimens.

| The limit of detection
The minimum detection limit is among the most important characteristics of applying a method to a clinical sample and obtaining a minimum number of false negatives. The calculated limit of detection (LOD) was determined from assay of 10 replicates of the zero calibrator with three times and calculated from the experimental result. LOD can be expressed as LOD = 3Sa/b where Sa is the standard deviation of the response and b is the slope of the calibration curve. 30 The detection limit was calculated as 1.0 pg/mL. Several immunoassay methods for plasma level of CT-1 were summarized in Table 2. Compared with the previous reports, MPs-CLIA presents relatively wider linear range, a lower detection limits and shorter detection time.

| Analysis of the accuracy
To evaluate the accuracy of MPs-CLIA, recovery experiment was performed by spiking various amounts of CT-1 (10 pg/mL, 100 pg/mL, 800 pg/mL) into pooled plasma samples and analyzed immediately.

| Precision
The precision of the MP-CLIA was estimated based on repeated measurement of healthy human pooled plasma spiked with calibrator CT-1. Three different concentrations of CT-1 calibrator (10, 100, and 800 pg/mL) were prepared under the same conditions. The intraanalysis precision was calculated by analyzing each concentration 10 times per run in one day (n = 10). Similarly, these samples were analyzed 10 times for three days (n = 30) to obtain the inter-analysis.
As shown in Table 4, the intra-analysis CVs of the three samples enrolled were 8.92%, 6.69%, and 3.54%, respectively, and inter-analysis CVs were 9.25%, 10.9%, and 4.3%, respectively. The CVs are acceptable for the very low concentration of CT-1 measured, at the pg/mL order. Such reproducibility is highly acceptable and in favor of the MPs-CLIA assay.

| Plasma CT-1 detection in clinical sample
Plasma CT-1 levels from 100 CHF patients and 40 subjects were determined to validate the clinical application of MP-CLIA. It shows that the CT-1 concentrations were significantly higher in CHF patients (median: 70.43pg/mL) compared with healthy individuals (median: 40.70 pg/mL) (P < .05, Figure 4A). ROC curve was further performed to evaluate the diagnostic value of CT-1, which yielded an area under the curve of 0.66 ( Figure 4B). In the cutoff of 52.06pg/mL, the sensitivity and specificity are calculated as 63.5% and 65%, respectively. The relationship between levels of plasma CT-1 and pro-BNP was studied. As is shown in Figure 5, regression analysis revealed significant association between CT-1 and pro-BNP (r = 0.339, P < .001

| CON CLUS ION
In conclusion, we constructed a double-antibody "sandwich"based chemiluminescent immunoassay for rapidly determine plasma level of CT-1. Compared to the current reported methods for CT-1 level analysis, the MPs-CLIA assay greatly reduces the detection time by as short as 1 hour, whereas the widely applied enzyme-linked immunosorbent assay (ELISA) requires more times of 4-5 hours. [31][32][33] And the assay is highly automated, avoiding the requirement of complicated operating procedures. Furthermore, systematic validation analysis revealed that the novel method has sufficient precision, accuracy, and reliability for quantitative CT-1. This developed chemiluminescence immunoassay will be a powerful tool to further explore more clinical value of CT-1.

ACK N OWLED G EM ENTS
We thank the Clinical Laboratory and Vasculocardiology Department from the Second Affiliated Hospital of Zhejiang University School of Medicine for essential support.

AUTH O R CO NTR I B UTI O N S
YP involved in protocol development, performed laboratory measurements, statistical analysis, and manuscript writing. XW involved in statistical analysis. YD and HW performed laboratory measurements. WL and PY collected clinical samples and data. ZT involved in protocol development and revised manuscript. All authors read and approved the final manuscript.

E TH I C S A PPROVA L A N D CO N S E NT TO PA RTI CI PATE
This study was approved by the Ethics Committee of the Second Affiliated Hospital, Zhejiang University School of Medicine. Written informed consent was obtained from all study participants before commencement of the study.