Multiplex Assays Bio-Plex suspension array system tech note 64 Profiling of Human, Canine, and Rat Urine Samples Using Bio-Plex Pro RBM Kidney Toxicity Assays L. Stephen, H. Schnaars, K. Marshall, H. Akey, A. Baugher 2, D. Jethwaney 2, V. Gupta 2, and W. Tan 2 Table. Bio-Plex Pro RBM multispecies kidney toxicity assays. Assays Myriad RBM Inc., 3 Main Street, Saranac Lake, NY 2983 USA 2 Protein Technology Research and Development Function Division, Bio-Rad Laboratories, Inc., 2 Alfred Nobel Drive, Hercules, CA 94547 USA Introduction Acute kidney injury (AKI) is a serious medical condition caused by a variety of bodily insults, including trauma from an accident, blood loss from a medical procedure, or exposure to nephrotoxic drugs or chemicals. In fact, drug exposure is estimated to cause up to 25% of all cases of AKI in critically ill patients (Bonventre et al. 2). From a pharmaceutical industry standpoint, drug-induced toxicity is a serious issue, sidelining 3% of therapeutics overall from preclinical lead compounds to marketed drugs (Bonventre et al. 2). The current clinical criterion for the diagnosis of AKI relies on glomerular filtration rate (GFR), blood urea nitrogen (BUN), and serum creatinine (SCr), which are neither sensitive nor specific (Slocum et al. 22). Traditional tests of this nature employ late biomarkers that are detectable only days or weeks after kidney damage has occurred. Researchers and clinicians are looking for biomarkers that are sensitive, specific, and early indicators of AKI. In addition, biomarker level monitoring should be cost effective and permit rapid reporting of results. In collaboration with Myriad RBM and the Predictive Safety Testing Consortium (PSTC), we have developed the Bio-Plex Pro RBM kidney toxicity assay panels (Table ). Based on Luminex xmap technology, these immunoassays enable robust measurement of multiple kidney toxicity/injury biomarkers in human, canine, and rat urine samples. The multiplex format allows researchers to detect key proteins that may be upregulated within hours of kidney damage, providing valuable information throughout the drug development process from lead optimization to preclinical and clinical protocol decision making. Human Kidney Toxicity Panel Panel 2 Toxicity Panel Toxicity Panel Panel 2 Method The Bio-Plex Pro RBM kidney toxicity assays employ a standard sandwich enzyme immunoassay method using a 96-well plate format. The capture antibody-coupled beads are allowed to react with a sample containing the target of interest (incubation time: 6 min). After performing a series of washes to remove unbound materials, a biotinylated detection antibody specific for a different epitope on the target is added to the beads (incubation time: 6 min). The result is the formation of a sandwich of antibodies around the specific target. The reaction mixture is detected by the addition of a reporter dye, streptavidin-phycoerythrin (SA-PE), which binds to the sandwich complexes via the biotinylated detection antibodies (incubation time: 3 min). The contents of each well are drawn up into the Bio-Plex array system, which identifies and quantifies each specific reaction based on bead color and fluorescence signal intensity (Figure ). The magnetic beads are employed to enable automation of wash steps using a Bio-Plex Pro wash station and to ensure compatibility with all Luminex-based life science research instruments. Data acquisition was performed using the Bio-Plex 2 system at low PMT setting, the Bio-Plex 3D system at standard PMT setting, or the Bio-Plex MAGPIX system using the default setting.
Human Kidney Toxicity Panel 2 Add sample Incubate 6 min Add detection antibody antibody Incubate 6 min.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.% Add reporter molecule molecule Incubate 3 min Fig.. Schematic representation of a Bio-Plex Pro RBM kidney toxicity assay workflow. The performance of these assays was evaluated according to standard verification and validation parameters, including assay sensitivity, precision, accuracy, working ranges, specificity, and linearity of dilutions. The assays were further evaluated on urine samples collected from human, canine, or rat with various kidney-related disorders. Assay Performance and Quality Assay specificity was examined by performing singledetection cross-reactivity using selected standard points and native (spiked urine) matrices. The study was conducted by testing the individual detection antibody in the presence of multiplexed antigens and capture beads. The degree of cross-reactivity is defined as the percentage of signal detected relative to the specific signal for that analyte (Table 2). The findings were in agreement with studies carried out in spiked urine samples. Table 2. Single-detection cross-reactivity based on standard point..%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.%.9%.%.%.%.%.%.%.%.%.%.%.2%.%.7%.% Assay sensitivity, defined as limit of detection (LOD), was determined by adding two standard deviations to the average of the median fluorescence intensity (MFI) for ten replicates of the standard curve blank run on three separate plates. The value was converted to concentration as interpolated from the standard curve (Table 3). 3.4%.% 6.%.%.%.%.3%.%.%.%.5%.%.%.%.%.% Toxicity Panel.%.%.%.%.%.%.%.%.%.%.%.% 23 Bio-Rad Laboratories, Inc. Bulletin 64
Table 3. Limit of detection (LOD) in. Human Kidney Toxicity Panel 2 2.698.4.52.5.7.2.83.6.75.26.53.2 Toxicity Panel..258..4.8.49.6.5.89 Toxicity Precision was calculated by testing urine-based controls in triplicate over five independent runs (Table 4). These runs were performed over a minimum of three days, with at least three different operators. The average of all coefficients of variation (%CV) of the calculated concentrations of the five runs was reported for intra-assay precision. The %CVs of all the calculated concentrations of the five runs was used to derive inter-assay precision.. Table 4. Intra- and inter-assay concentration () %CV..83.6.75.26 Toxicity.2 C.2 5.8 4.4 4.6 2.4 C2 9.6 3.5. 6.5 2.8 Inter-assay %CV C 3.8 2. 3.5 9.4 2.4 Assay accuracy (also defined as recovery) was calculated as the percentage of the observed concentration value of a spiked standard of known concentration relative to the expected value. Overall, these assays recovered within 8 2% on at least seven of the eight standard points (Table 5). The accuracy of the assays was also demonstrated by the performance of a two-level urine-based control (C and C2) shown in Table 6. The test value obtained by the end user fell within the concentration range of the low and high control concentrations. These quality controls can be used to monitor replicate analysis, show consistency in reagent dispensing technique, and determine storage condition of the assay reagents. C 3.2 8. 2.9.7 C2.8 3.5 9.9 7.5 Inter-assay %CV C 6.7.8 5.3 3.4 Toxicity Panel 2 C.6 4.5.5 4.4 C2.3 4. 3.4 3.4 Inter-assay %CV C 7.7 6.6 2. 6.5 5.2 4. 22.6 Toxicity Toxicity 7.8.3 9.9 Note: C and C2 refer to level one and two controls provided in the kits. Table 5. Assay accuracy (standard recovery). C 4.7.5.7 5.7..4 C2.8 6.2 2. 2.6.8 3.3 Inter-assay %CV C 4.6 8.3.9 3. 2.9.7 Human Kidney Toxicity Panel 2 C...5 4. 2..5 C2 4.9 2.7.3 3.9 3. 3.8 Inter-assay %CV C..5 5.2 6.2 3. 6.4 S % % 3% % % 98% S2 98% 99% 97% 98% % 4% S3 3% 2% 4% 2% % 3% S4 98% 99% 98% % 98% 98% S5 99% 99% 99% % 2% 98% S6 3% % 9% 98% 2% 4% S7 2% % 94% 2% 95% 97% S8 % 99% 6% 85% continues 23 Bio-Rad Laboratories, Inc. Bulletin 64
Human Kidney Toxicity Panel 2 S 99% % % % % % S2 5% 99% 99% 99% 99% 99% S3 2% % % % % 2% S4 95% % 99% 99% % 99% S5 6% 99% % % 98% 98% S6 5% 3% 2% 3% 3% 3% S7 86% 97% 94% 84% 96% 4% S8 96% 4% 9% S 99% % % 98% 99% S2 99% % 2% 4% % S3 6% % 98% 99% 99% S4 94% % % 99% 3% S5 9% % % 3% 98% S6 83% 98% % 98% % S7 97% % % % 99% S8 6% 2% 98% % 3% S % % % 3% S2 94% % % 98% S3 99% 99% % 2% S4 3% 2% 98% % S5 98% 97% 98% 99% S6 3% 4% 8% % S7 82% 98% 94% % S8 96% 92% 99% Toxicity Panel S % % % % S2 3% 99% % % S3 97% 2% % 2% S4 97% 99% 99% 98% S5 % % 98% % S6 3% 4% 5% 3% S7 3% 95% 98% % S8 3% % 95% % % % % 2% 98% % % Toxicity Toxicity 99% 3% 98% 3% 99% % 2% % Note: Cells with a dash mark indicate recovery outside 8 2%. Table 6. Performance characteristics of kit controls. All values are listed in. C range 38. 45. 7..5.4.43 44. 34. 5. 4.5 4.2.3 Test value 245.28 85.4 3.66 2.85 2.76.8 4. 2.5.3.6.6.2 42. 7.6 4..8.6.5 Test value 26.44 5.26 2.73...3 Human Kidney Toxicity Panel 2 C range C range 3. 4.2 3.6.2 89. 9.4 3...37 266. Test value 5.46 8.45 6.78.24 77.87.28.29.7.7 6..85.87.5.2 8. Test value.6.66.39.2 3.35 33..2 2.8.8 36. 4.7 98. 3.6 8.4 5.5 48. 4. Test value 63.85 2.29 5.92 3.57 263.84 9.3 3.9..3.5 4.8.24 2..3.94.45 5..7 Test value 7.58.2.63.3..46 C range 279.64 73.5 3.94 4.3 838.92 29.6.83 23.9 Test value 55.8 4.77 7.83 75.8 5.6 2.63.8.2 46.84 7.9.53 3.62 Test value 3.4 5.3.29 2.22 Toxicity Panel C range.33 97..6.89. 292. 4.9 2.7 Test value.69 69.7 3.5.67.2 2.4.3.3.7 7.2.. Test value.4 5.33.8.7 Toxicity 4. 2. 8.36.79.24.2 Toxicity.68.2.8.2.4.3 Note: C and C2 refer to level one and two controls provided in the kits. 23 Bio-Rad Laboratories, Inc. Bulletin 64
Assay working range is defined as the range between the lower limit of quantitation (LLOQ) and the upper limit of quantitation (ULOQ) in which an assay is both precise and accurate. LLOQ and ULOQ were defined as the point at which the CV for samples was 3% and 2% respectively, with an accuracy of 8 2%. The results are tabulated in Table 7. Table 7. Assay working range (). LLOQ 4.36.26.5.2.2. ULOQ,75,25 23 5 2 3.8 Human Kidney Toxicity Panel 2 LLOQ.36.5.4..99 ULOQ 35 35 5 3.2 84 LLOQ.22..4.7.75.2 ULOQ 64 22 4 34 2, 98 A. Serially diluted control Human. R 2 =.9975.. 2. 4. 6. 8.. Expected concentration, Rat 5. R 2 =.9997 Toxicity.. 2.5 5. LLOQ 8.3.7.6.63 ULOQ 2,76 535 42 48 Toxicity Panel LLOQ..69.. ULOQ 4.8 8 9.5 35 Toxicity. 4.8 3. Expected concentration, Canine R 2 =.9987 Linearity of dilution ensures that analytes present in concentrations above the LLOQ can be diluted and measured accurately within the assay working range. In this study, a high control was serially diluted in standard diluent (Figure 2A) to obtain at least four data points. The observed and expected analyte concentrations were plotted to derive the correlation coefficient (R 2 ) values. The study was also conducted in urine samples collected from all species (Figure 2B)... 5. 3. Expected concentration, 23 Bio-Rad Laboratories, Inc. Bulletin 64
B. Serially diluted urine Human 24. R 2 =.993.. 2. Expected concentration, Parallelism was investigated by comparing the slope of a sample curve (spiked urine, serially diluted in urine matrix) with a standard curve prepared in standard curve diluent. The purpose is to ensure the sample matrix is biologically comparable to the matrix of the standard curve. This was judged by calculating the slope difference between the two curves (Figure 3). Human Flourescence intensity 2, 6, 2, 8, 4, Slope Difference: 5.75%. Rat R 2 =.9966 Flourescence intensity 4,.,, Concentration, Rat 2, Slope Difference: 3.83%, 8, 6, 4, 2,... Expected concentration, 2,... Concentration, 3. Canine R 2 =.9975 Flourescence intensity Canine 2,, Slope Difference: 3.29% 8, 6, 4, 2, 2,... Concentration,.. 5. Expected concentration, Fig. 2. Linearity of dilution. A high control (A) and urine samples (B) were serially diluted to obtain at least four data points. The observed and expected analyte concentrations were plotted and the correlation coefficient (R 2 ) values reflect linearity in signal response. Fig. 3. Assay parallelism. Standard curve ( ) and serial dilution ( ) of a spiked urine sample. Validation with Biological Samples The kidney toxicity assays were validated primarily with urine samples. Sample dilution is unique for each of the assay panels (Table 8). Representative sample data are shown in Figure 4. The end users may also perform the same analysis in serum or plasma samples. In this case the dilution factor should be adjusted empirically for each of the targets or panels. 23 Bio-Rad Laboratories, Inc. Bulletin 64
Table 8. Recommended sample dilution factor. Human Kidney Toxicity Toxicity Toxicity Sample Types Panel Panel 2 Panel Panel 2 Panel Urine 4 5 2 5 5 Serum/Plasma User Defined 5, 4, 3, 2,, Human p =.5.7.6.5.4.3.2. Human p =.48 2 6 2 8 4 Human p =.25 Human Cystatin Human Human.8.6.4.2 p =.25 4 3 2 p =.23 4, 3, 2,, p =.43 Rat Rat Rat.8.6.4.2 p =.26 x -8,, 8, 6, 4, 2, p = 7.7 x -6.2.8.6.4.2 p =.23 x -9 Rat Rat Rat.4.2.8.6.4.2 p = 3.7 x -5 3,5 3, 2,5 2,,5, 5 p =.33 8, 4,, 8, 4, p =. 2,, 3, 5, Canine Canine Canine 2..5...3. Canine Canine Fig. 4. Sample validation with urine samples 2. collected from healthy subjects and subjects with various kidney-related disease types. N = 4 for human subjects with chronic kidney disease, diabetic nephropathy, and renal stones 6. (N = 4 for healthy), N = 3 for rats with polycystic kidney disease (N = 6 for healthy). Bio-Plex Data Pro software was used to calculate statistical difference by t-test, where a p-value <.5 indicates statistical significance. The canine data. were generated from one animal, 2, and 8 days post gentamycin treatment. 23 Bio-Rad Laboratories, Inc. Bulletin 64
Instrument alignment Bio-Plex 2, Bio-Plex MAGPIX, and Bio-Plex 3D Systems The assays were evaluated on Bio-Plex 2, Bio-Plex MAGPIX, and Bio-Plex 3D systems. Excellent agreement in sample readout was recorded on all three platforms (Figure 5). 8 7 6 5 4 3 2 5 4 3 2 Bio-Plex 2 Bio-Plex 2 Human Bio-Plex MAGPIX Rat Bio-Plex MAGPIX Bio-Plex 3D Bio-Plex 3D Conclusions The Bio-Plex Pro RBM kidney toxicity assays, developed in partnership with Myriad RBM, comprise a highly relevant set of biomarkers for early detection and characterization of kidney toxicity/injury. All seven assay panels demonstrated excellent sensitivity, broad working assay range, and high precision. We observed robust performance in human, canine, and rat urine samples across a range of physiological conditions from normal to diseased to drug-treated. The multiplex format provides an effective solution for protein biomarker detection compared to traditional ELISA, and the premixed all-in-one kits enable drug development researchers to efficiently test lead compounds in both preclinical animal models and human clinical trials. References Bonventre JV et al. (2). Next-generation biomarkers for detecting kidney toxicity. Nat Biotechnol 28, 436-44. Slocum JL et al. (22). Marking renal injury: can we move beyond serum creatinine? Transl Res 59, 277 289. The Bio-Plex suspension array system includes fluorescently labeled microspheres and instrumentation licensed to Bio-Rad Laboratories, Inc. by the Luminex Corporation. xmap and Luminex are trademarks of the Luminex Corporation. Manufactured by Myriad RBM Canine 35 3 25 2 5 5 Bio-Plex 2 Bio-Plex MAGPIX Bio-Plex 3D Fig. 5. Alignment in sample readout. Each bar represents analyte levels of normal urine samples collected from human (N = 5), rat (N = 6), and canine (N = 5), recorded by Bio-Plex 2 (n), Bio-Plex MAGPIX (n), and Bio-Plex 3D (n) systems. Bio-Rad Laboratories, Inc. Life Science Group Web site www.bio-rad.com USA 8 424 6723 Australia 6 2 994 28 Austria 877 89 Belgium 9 385 55 Brazil 55 544 5699 Canada 95 364 3435 China 86 2 669 85 Czech Republic 42 24 43 532 Denmark 44 52 Finland 9 84 22 France 47 95 69 65 Germany 89 3 884 Greece 3 2 9532 22 Hong Kong 852 2789 33 Hungary 36 459 6 India 9 24 4293 Israel 3 963 65 Italy 39 2 269 Japan 3 636 7 Korea 82 2 3473 446 Mexico 52 555 488 767 The Netherlands 38 54666 New Zealand 64 9 45 228 Norway 23 38 4 3 Poland 48 22 33 99 99 Portugal 35 2 472 77 Russia 7 495 72 4 4 Singapore 65 645 388 South Africa 27 86 246 723 Spain 34 9 59 52 Sweden 8 555 27 Switzerland 26 674 55 5 Taiwan 886 2 2578 789 Thailand 8 88 22 88 United Kingdom 2 8328 2 Bulletin 64 Rev A US/EG 3-89 73 Sig 22