Amplification of NGAL cDNA

Mammary gland tumors (MGT) have been reported to highly express NGAL [21] and therefore, in order to clone the NGAL gene sequence, total RNA was extracted from MGT by reagent TRIzol (Invitrogen) according to the manufacturer’s instructions. The first-strand cDNA was synthesized by SuperScript III reverse transcriptase (Invitrogen) using RNA (1 μg) and primers (50 μM). Subsequently, NGAL was amplified by polymerase chain reaction (PCR) with primer designed according to GenBank accession no. XM_548441 (NGAL-F: 5^′-ATGACCCAAGTTCTCCTGTG; NGAL-R: 5^′-TCACTCATCAATGCACTGGTC). The thermocycling conditions started with 94°C for 5 min followed by 35 cycles of heat denaturation at 94°C for 30 s, primer annealing at 60°C for 30 s, DNA extension at 72°C for 30 s and a final extension at 72°C for 5 min. The resulting PCR product with predicted size was isolated and cloned into PCR 2.1 TOPO vector (Invitrogen), designated c-NGAL/TOPO. The identity of the NGAL was verified by automated sequencing (Mission Biotech, Taipei, Taiwan).

Construction of a plasmid expressing recombinant NGAL

Initially, the canine NGAL gene was amplified by PCR from the c-NGAL/TOPO plasmid using the primer set F-BamH: 5'-aggatccaatgacccaagttctcctg-3' and R- Xho: 5’-ttctcgagctcatcaatgcactggtc-3‘. The PCR conditions were similar to that used for NGAL amplification from tissue, except for primer annealing at 49°C. Subsequently, the PCR product was digested with BamH I/Xho I and cloned into pET32b. The resulting plasmid was confirmed by bi-directional sequencing.

Preparation of recombinant dog NGAL

Protein expression was carried out in E. coli. strain BL21 AI (Invitrogen) following the procedures described previously [22]. Briefly, protein expression was induced with isopropyl-β-D-1-thiogalactopyranoside (IPTG, 0.8 mM) and 0.2% L-arabinose (CALBIOCHEM) at 16°C for 24 h. The bacteria pellets were resuspended in lysis buffer (500 mM NaCl, 500 mM Tris–HCl, 20 mM imidazole, and 8 M urea; pH 7.4) and this was followed by three sonication cycles of 30 sec each (550 Sonic Dismembrator; Fisher Scientific). After a second centrifugation, the cell supernatant containing the recombinant proteins was recovered and purified using chelating Sepharose Fast Flow (GE Healthcare) by following the manufacturer’s instructions. Finally, the bound protein was eluted in 4 mL elution buffer (0.05 M Tris–HCl, 0.5 M NaCl, 400 mM imidazole, and 8 M urea, pH 7.4) and then dialyzed against 1x PBS at 4°C to remove excess imidazole and urea.

Western blotting

Purified proteins were separated by 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and then electrophoretically transferred to nitrocellulose membrane. The procedures for Western blot analysis followed those in a previous report [23]. Briefly, after a blocking step in PBS containing 0.1% Tween-20 (PBST) and 5% fat-free dried milk for 1 h at room temperature, the membranes were incubated with 1:5000 diluted anti-His tag antibody (AbD Serotec) or 1:500 diluted rabbit immunized with recombinant NGAL in PBST-5% dried milk at 4°C for overnight. Next, the filter was washed in PBST, which was followed by incubation with horseradish peroxidase (HRP)-conjugated secondary antibody in PBST-2% dried milk at room temperature for 1 h. After extensive washing with PBST, the filter was developed using an enzyme-linked chemiluminescence system (ECL, Amersham, GE Healthcare) and exposed to X-ray film.

Production of antibodies against NGAL

Three BALB/c mice, aged eight weeks and three-month old New Zealand white rabbits were immunized with 50 μg NGAL/mouse (200 μg/rabbit) mixed with complete Freund's adjuvant (Sigma). After initial immunization, two boosters of the same dose were given at two-week interval. Blood was collected from the mouse submandibular vein by a lancet and from the ear vein of the rabbits. After centrifugation, the plasma was transferred to a new tube and stored at -20°C until use. The study was approved by the Institutional Animal Care and Use Committee of National University of Chung-Hsing University, permit number: 100-66.

Establishment of a sandwich ELISA for the detection of canine NGAL

Our NGAL ELISA was optimized as follows. A checkerboard titration was initially conducted using various concentrations of the recombinant canine NGAL protein (at concentrations of 0, 2.48, 7.4, 22.3, 67, 200, and 600 pg/mL) in combination with serially diluted rabbit serum (dilutions of 1:100, 1:200, 1:400, 1:800, 1:1600, and 1:3200). All combinations were repeated in triplicate. It was found that the 1: 800 dilution of rabbit serum was sufficient to give strong signal with the lower concentration of NGAL proteins and, based on this, 1:800 diluted rabbit serum was used as the capture antibody in our Sandwich ELISA. The dilution of the mouse serum, as the detector antibody, was optimized following the same strategy.

After optimization, the ELISA conditions were used to detect NGAL in clinical samples. To do this, a 96-well microplate was coated with 100 μL of rabbit sera containing anti-NGAL polyclonal antibody (the capture antibody) at 1:800 dilution in coating buffer (0.15 M sodium carbonate, 0.35 M sodium bicarbonate, pH 9.6) at 37°C for 1 h. After blocking, the test samples (20-fold diluted with PBS-T containing 5% dried milk), recombinant NGAL protein at known concentrations as positive controls, and calibrator (diluted with blocking buffer) were individually added to each well and incubated at 4°C overnight. After washing with PBS-T, each well received 100 μL of mouse anti-NGAL polyclonal antibody (detector antibody, 1:3000 diluted in PBS-T containing 5% fat-free dried milk). After incubation at 37°C for 1 h, HRP conjugated goat anti-mouse IgG antibody (5,000 fold diluted in blocking buffer) was added to the wells. After 1 h incubation, the result was visualized using tetramethylbenzidine (TMB) substrate kit (Clinical Science Laboratory, Inc.). The optical density (OD) of each well was read at 450 nm using a microplate reader (TECAN). Each sample was analyzed in triplicate, and the OD of the triplicates was averaged. Samples with an OD value three times greater than the negative control serum were considered positive. A commercial Dog NGAL ELISA kit (BioPorto Diagnostics, kit 043) was compared with our results to assess the performance of our sandwich ELISA. Linear regression was conducted to measure the correlation between the commercial ELISA kit and our sandwich ELISA.

Cases and sample collection

In this study, cases that had undergone surgery at the Veterinary Teaching Hospital, National Chung Hsing University, between 2009 and 2011 were enrolled. To obviate postoperative volume depletion and to avoid prerenal azotemia, all dogs received fluid transfusion for maintenance during and after surgery. Prior to the surgery, blood and urine samples were collected, while clinico-pathological information was simultaneously recorded. In addition, demographic information, including age, sex, body weight, and surgery type of each case, was also recorded. The anesthesia protocol was individually set for each patient. Generally, the preanesthetic agents, including opioid, midazolam/ketamine and propofol, were used for induction and isoflurane was administrated for general anesthesia. Serum and urine samples were obtained for NGAL analysis before the surgery and at frequent intervals after surgery when this was possible (12 h, 24 h, 48 h and 72 h post operation). The urine and serum supernatants were stored at -80°C until the NGAL test was conducted. For comparison, serum creatinine was simultaneously tested at the same time points.

Data categories

Dogs with an elevated serum creatinine (> 132.6 μmol/L) or relevant renal clinical signs/history such as polyuria, polydipsia, oliguria, changes in the size of the kidneys, weight loss, obstructive urinary disease and proteinuria before surgery were excluded from the AKI study. The remaining animals were included in the study, enrolled for urine NGAL detection and used to compare the data from the in-house ELISA and commercial ELISA. Corresponding to the AKIN criteria [3], the serum creatinine of each case was taken as the baseline. The cases with a serum creatinine level that became raised to 26.5 μmol/L or more from the baseline within 48 h were considered to form the postoperative ‘AKI group’. The remaining cases formed the ‘No AKI group’. In addition, the operation duration and the use of any nephrotoxic related medicine including non-steroidal anti-inflammatory drugs (NSAIDs), angiotension-convertion enzyme inhibitors (ACE-inhibitors) and chemotherapeutic medicines such as carboplatin were also recorded.

Statistical methods

SPSS 16.0 was used for analysis. For continuous variables, normally distributed variables are represented as a mean ± S.E.M and were compared by the independent Student’s t test or by one way ANOVA coupled with post hoc analysis by the Tukey’s method. Non-normal distribution variables were compared by the Mann–Whitney U test and recorded as medians (interquartile range, IQR). The χ² test or Fisher’s exact test were used for categorical variable comparison. Univariate analysis of the serum and urine NGAL values by logistic regression was applied to identify the variables that were possible associated with AKI in dogs post operation.

Preparation and verification of the NGAL antibodies

To develop an in-house NGAL ELSIA, the entire coding region (597 nucleotides; 199 amino acids) of the canine NGAL gene was cloned into the expression vector pET32a (Figure 1A) and the recombinant NGAL protein was expressed and identified. As shown in Figure 1B, the thioredoxine-NGAL was successfully expressed with an expected molecular weight of approximately 42 kDa (Figure 1B). The identity of the NGAL protein was confirmed by Western blot analysis using antibody against the histidine tag (Figure 1C) and by mass spectrometry (data not shown).

As analyzed by Western blotting, the sera from the mice and rabbits immunized with NGAL were able to recognize the recombinant NGAL protein (Figure 2A, sample 2, 4) and also specifically detected canine NGAL (25 kDa) from two renal failure dogs in which the presence of NGAL had been diagnosed by a commercial kit (Figure 2B, sample 2 and 3). This confirms the specificity of antibodies produced by the present study.

Establishing the ELISA for NGAL detection

Based on the checkerboard titration result, the condition of ELISA for the detection of canine NGAL was optimized as 1:3000 diluted mouse serum and 1:800 diluted rabbit serum, which were used as the detector antibody and the capture antibody, respectively. Following optimization, twenty clinic urine samples collected from 10 healthy dogs and 10 renal failure dogs were analyzed. To evaluate the performance of the in-house NGAL ELISA, these samples at a dilution of 1:20 were also tested by a commercial NGAL kit in parallel. Results of commercial kit indicated that the OD of the urine samples from healthy dogs ranged from 0.042 - 0.521 (equivalent to 0.93 pg/mL - 159.6 pg/mL of calibrator NGAL protein), and that higher readings (0.384 - 2.396; 111.5 pg/mL - 785.6 pg/mL of the calibrator NGAL protein) were detected from renal failure dogs (Figure 3A, left panel). Consistent results were obtained using our NGAL ELISA with a higher OD being detected from the renal failure dogs (0.306 - 1.507; 306 pg/mL - 4492 pg/mL of recombinant NGAL protein) compared to the healthy group (0.114 - 0.356; 0 pg/mL - 314.3 pg/mL of recombinant NGAL protein; Figure 3A, right panel).

Comparison of two ELISA methods

As the format, signal detection system, and assay standardization of these two ELISA systems were different in many ways. Bland-Altman plot for twenty urine samples was first performed to assess the agreement of the two methods. The mean difference between our ELISA and the commercial one is 1064.2 pg/mL, and 95% limits of agreement are -1739.4 pg/mL and 3867.8 pg/mL (Figure 3B). Linear regression and coefficient of correlation analyses were used to correlate the results of these two ELISA systems. A cross-sectional pilot study was designed to verify the standardized commercial ELISA kit against our NGAL ELISA assay. As shown in Figure 3C, NGAL concentrations in 21 urine samples and 8 calibration standards (NGAL range, 0 to 400 pg/mL) were determined by the two assays and these were found to be highly correlated (r = 0.96).

The equation obtained was Y = 0.1771X + 131.1735 with a coefficient of correlation 0.9646 and a r² = 0.9307 (p < 0.0001); this was then used to correlate the concentration of canine NGAL determined by our ELISA assay using the purified recombinant fusion NGAL protein as calibrators with that obtained by the commercial ELISA kit using the standard calibrator NGAL protein.

Detection of NGAL in acute kidney injury (AKI) before and after surgery using NGAL ELISA

Next, our ELISA was used to analyze the level of NGAL in dogs that had undergone surgery. Of the 39 cases investigated, 13 were for orthopedic surgery, 20 were for soft tissue surgery and six were for tumor mass removal. An increase in serum creatinine greater or equal to 26.5 μmol/L within 48 h post surgery was found to have occurred in 12 cases (30.7%), which were then classified into the AKI group. The remaining dogs were classified into the No AKI group.

In contrast, when the serum creatinine levels were examined, no distinct difference was found between the two groups of dogs until 24 h post surgery (Table 2). However, when the NGAL concentrations in the serum of the AKI and No AKI groups at the various times were compared, they were found not to be significant different between the two groups (Table 2). In addition, there was no significant difference between the AKI and No AKI groups for BUN.