Quantification of NTproBNP in Rat Serum Using Immunoprecipitation

Dec 11, 2007 - arkers of heart failure. BNP hormone and the inactive. NTproBNP are predominantly secreted in the ventricles of the heart in response t...
0 downloads 0 Views 198KB Size
Download PDF
Anal. Chem. 2008, 80, 561-566

Accelerated Articles

Quantification of NTproBNP in Rat Serum Using Immunoprecipitation and LC/MS/MS: a Biomarker of Drug-Induced Cardiac Hypertrophy Michael Berna,*,† Lee Ott,† Steve Engle,† David Watson,† Philip Solter,‡ and Bradley Ackermann†

Lilly Research Laboratories, Eli Lilly and Company, Greenfield, Indiana 46140, and Department of Pathobiology, University of Illinois at UrbanasChampaign, Urbana, Illinois 61802

Brain natriuretic peptide (BNP) and N-terminal proBNP (NTproBNP) are well established in the clinic as biomarkers of heart failure. BNP hormone and the inactive NTproBNP are predominantly secreted in the ventricles of the heart in response to pressure overload and, consequently, are being investigated as markers of druginduced cardiac hypertrophy in rat to support drug development. In the work presented here, an immunoaffinity-based LC/MS/MS assay was developed and validated to measure a selective tryptic fragment of NTproBNP in rat serum. The assay covers the range of 13-329 pg/mL of the tryptic fragment LLELIR, corresponding to 0.1-2.5 ng/mL intact NTproBNP. A stable isotopelabeled version of NTproBNP containing the tryptic fragment LLELI[13C615N1]R was prepared by solid-phase peptide synthesis and was used as an internal standard to minimize assay variability. Due to endogenous NTproBNP present in rat serum, human serum was used as the control matrix, and parallelism between rat and human serum was established by standard addition. Assay accuracy (% RE) and precision (% CV) were measured at three concentrations on each of 4 days and did not exceed 4.2 and 14.5%, respectively. Additionally, study data are presented from the application of this assay in which rats demonstrated a significant increase in NTproBNP serum concentration following administration of an agent known to induce cardiac hypertrophy. In this study, the relationship between serum NTproBNP and cardiac hypertrophy was corroborated by increases in heart weight and magnetic resonance imaging of the test subjects’ left ventricle. To our knowledge, this represents 10.1021/ac702311m CCC: $40.75 Published on Web 01/08/2008

© 2008 American Chemical Society

the first reported assay for NTproBNP in preclinical species for the assessment of cardiac hypertrophy.

B-Type natriuretic peptide (brain natriuretic peptide, BNP) and an associated cleavage product of proBNP (see Figure 1), N-terminal natriuretic peptide (NTproBNP), are currently used in the clinic in the diagnosis and treatment of patients with heart failure (HF).1 The neurohormone BNP and NTproBNP are primarily secreted from the cardiac ventricles in response to pressure overload, and current evidence clearly supports the use of these peptides in three clinical settings: patients with acute dyspnoea, prior to discharging patients hospitalized with acute HF, and in the long-term management of patients with HF.1 Natriuretic peptides are believed to mitigate adverse cardiovascular remodeling through the promotion of vasodilation and natriuresis, and circulating levels of these peptides correlate with the magnitude and pattern of left ventricular hypertrophy.2 There is a long history of using biomarkers as a noninvasive method of diagnosing and treating cardiovascular disease.3 Consequently, in an effort to decrease the attrition rate during drug development due to drug-induced cardiotoxicity, we are developing a panel of biomarkers that have proven clinical utility and can be applied to veterinary species during toxicology * To whom correspondence should be addressed. Phone: 317-277-6279. E-mail: [email protected]. † Eli Lilly and Company. ‡ University of Illinois at UrbanasChampaign. (1) Mueller, C.; Breidthardt, T.; Laule-Kilian, K.; Christ, M.; Perruchoud, A. P. Swiss Med. Weekly 2007, 137, 4-12. (2) Konstam, M. A. JAMA, J. Am. Med. Assoc. 2007, 297, 212-214. (3) Herrmann, J.; Volbracht, L.; Haude, M.; Eggebrecht, H.; Malyar, N.; Mann, K.; Erbel, R. Med. Klin. 2001, 96, 144-156.

Analytical Chemistry, Vol. 80, No. 3, February 1, 2008 561

Figure 1. Metabolic processing of preproBNP. This results in the release of a signal peptide and proBNP, which is subsequently secreted by cardiac myocytes. Further processing of proBNP results in the release of the neurohormone BNP and the inactive cleavage product NTproBNP.

assessment. These include myosin light chain 1 (Myl3),4 fatty acid binding protein 3 (Fabp3),5 and cardiac troponin I (cTnI),6 which have demonstrated their utility in evaluating cardiac and musculoskeletal necrosis. In the work presented here, we evaluate NTproBNP as a biomarker of drug-induced cardiac hypertrophy, a safety concern that has led to late-stage development failures of drug candidates such as PPARγ and dual PPARRγ agonists.7 ELISA has historically been the method of choice for protein quantification owing to its high throughput and sensitivity; however, LC/MS/MS is playing an increasingly important role in this venture due to its unparalleled selectivity, minimal method development time, and cost.8-11 Our strategy takes advantage of the method development speed and selectivity of LC/MS/MS, in combination with immunoaffinity capture sample enrichment, to rapidly validate biomarker biology, triage candidate markers for ELISA development, and subsequently validate ELISA selectivity. Using this approach, the proteins of interest are isolated from serum by immunoprecipitation (IP) and surrogate peptides are produced through proteolysis, usually with trypsin. The tryptic peptides are ultimately measured using stable isotope dilution and LC/MS/MS. (4) Berna, M. J.; Zhen, Y.; Watson, D. E.; Hale, J. E.; Ackermann, B. L. Anal. Chem. 2007, 79, 4199-4205. (5) Zhen, Y.; Berna, M. J.; Pritt, M. L.; Jin, Z.; Watson, D. E.; Ackermann, B. L.; Hale, J. E. Proteomics: Clin. Appl. In press. (6) Zhen, E.; Berna, M. J.; Jin, Z.; Watson, D. E.; Ackermann, B. L.; Hale, J. E. The 55th ASMS Conference on Mass Spectrometry, Indianapolis, IN, June 3-7, 2007. (7) Rubenstrunk, A.; Hanf, R.; Hum, D. W.; Fruchart, J.-C.; Staels, B. Biochim. Biophys. Acta. In press. (8) Barnidge, D. R.; Goodmanson, M. K.; Klee, G. G.; Muddiman, D. C. J. Proteome Res. 2004, 3, 644-652. (9) Gerber, S. A.; Rush, J.; Stemman, O.; Kirschner, M. W.; Gygi, S. P. Proc. Natl. Acad. Sci. U.S.A. 2003, 12, 6940-6945. (10) Anderson, N. L.; Anderson, N. G.; Haines, L. R.; Hardie, D. B.; Olafsom, T. W.; Pearson, T. W. J. Proteome Res. 2004, 3, 235-244. (11) Berna, M.; Schmalz, C.; Duffin, K.; Mitchell, P.; Chambers, M.; Ackermann, B. Anal. Biochem. 2006, 356, 235-243.

562

Analytical Chemistry, Vol. 80, No. 3, February 1, 2008

In the work here, an LC/MS/MS assay was developed and validated to measure NTproBNP in rat serum. NTproBNP was isolated from serum using IP, and following enzymatic digestion with trypsin, a 6-mer peptide was quantified using multiple reaction monitoring detection. A practical illustration of the assay is presented in which elevated levels of NTproBNP are detected following a 14-day administration of an agent known to cause cardiac hypertrophy. EXPERIMENTAL SECTION Chemicals and Reagents. Formic acid (88%), HPLC acetonitrile, and HPLC methanol were obtained from MallinckrodtBaker (Paris, KY), and HPLC water was acquired from Fisher Scientific (Fair Lawn, NJ). Bovine serum albumin (1% w/v, pH 7), EDTA (0.5 M, pH 8), sodium chloride (5 M), Triton X-100, ammonium biocarbonate, and Tris-HCl buffer (1 M, pH 8) were obtained from Sigma-Aldrich (St. Louis, MO). The protein-A agarose gel used for IP was from Pierce (Rockford, IL), and the rabbit polyclonal antibodies (2AG, 3AG) to rat proBNP were obtained from Invitrogen (Carlsbad, CA). Mass spectrometry grade Trypsin Gold was purchased from Promega (Madison, WI), while control rat and human serum were from Bioreclamation (Long Island, NY). Stable isotope-labeled NTproBNP (SIL), containing the tryptic fragment (LLELI[13C615N1]R), and a 6-mer synthetic peptide to the NTproBNP tryptic fragment (LLELIR) were synthesized by Midwest BioTech (Fishers, IN); the 6-mer peptide was used for method development purposes only. Rat NTproBNP (H-HPLGSPSQSPEQSTMQKLLELIREKSEEMAQRQLSKDQGPTKELLKRVLR-OH) was expressed at Eli Lilly and Co. (Indianapolis, IN) and characterized by LC/MS, amino acid analysis (Purdue University, West Lafayette, IN), SDSPAGE, and BCA protein assay (Pierce). Buffer 1 was prepared by diluting the solutions obtained from Sigma-Aldrich to a final concentration of 50 mM Tris-HCl, 150 mM NaCl, 1 mM EDTA, and 0.1% bovine serum albumin and by

Figure 2. Product ion spectrum of the NTproBNP tryptic fragment LLELIR presented indicating predominant y and b ions. The spectrum was obtained at low collision energy to verify the identity of the tryptic fragment; however, the product ion at m/z 530.3 used for SRM was the base peak when the collision energy was optimized for quantification.

adjusting the pH to 7.5. Buffer 2 was prepared in the same fashion as buffer 1 but did not contain bovine serum albumin and had 0.01% Triton X-100. The digest buffer was 50 mM ammonium bicarbonate (pH 8.5) containing 10% methanol (v/v). The rabbit polyclonal antibodies (Ab, 2AG, and 3AG) were mixed 1:1 (w/w) and immobilized to the protein-A gel according to the Pierce package insert (product 45215) at a concentration of 4 µg of Ab/12.5 µL of gel slurry (50% slurry by volume). Cardiac Hypertrophy Study Design. Two groups of eight Sprague-Dawley rats were used in this study. The first group received a once-daily dose (50 mg/kg) of an agent known to induce cardiac hypertrophy by oral gavage for a period of 14 days. The second group received a once-daily dose of vehicle using the same protocol outlined above. On day 15, blood samples (∼4 mL) were collected from all animals via the abdominal aorta, a protease inhibitor was added, and the samples were allowed to coagulate at room temperature for between 30 min and 1 h. Following coagulation, the samples were centrifuged and serum was collected on dry ice prior to storage at -70 °C. The animals were euthanized by carbon dioxide followed by cervical dislocation. Following euthanasia, the left ventricle was collected from each subject and weighed, and gross pathology observations were noted. One day prior to dosing, and on days 0 (postdose), 4, and 5, magnetic resonance imaging (MRI) images of the ventricles of all subjects were collected. Sample Preparation. To facilitate LC/MS/MS and optimize sensitivity, a tryptic fragment (LLELIR) selective for NTproBNP was used as a surrogate measure of NTproBNP serum concentration, taking advantage of their stoichiometric relationship. NTproBNP was used to prepare standard samples, which were analyzed in duplicate at 0.10, 0.25, 0.50, 0.75, 1.0, and 2.5 ng/mL (resulting LLELIR concentrations ranging from 13 to 329 pg/mL). The 2.5 ng/mL standard was prepared by spiking 36.4 µL of a 137 ng/mL solution into 2 mL of control human serum. The remaining standards were prepared through dilution of the 2.5 ng/mL standard with control human serum. Validation (QC) samples were prepared to evaluate accuracy and precision at 0.10 (LLOQ), 0.50 (MID), and 2.5 (ULOQ) ng/mL using the same procedure used to prepare the standard samples.

Double blank (no SIL or NTproBNP) and blank (no NTproBNP) samples were analyzed with each set of standard samples to verify the absence of peaks that might interfere with the detection of NTproBNP and SIL. Solvent blank samples containing digest buffer were also placed after the highest standard sample to evaluate system carryover. Immunoprecipitation and Digestion. On day one of the IP procedure, 1 ng of SIL and 100 µL of each sample were transferred to a Handee spin column (Pierce, 69705). The samples were diluted with 300 µL of buffer 1 followed by the addition of 12.5 µL of protein-A gel (4 µg of Ab). The samples were rocked at ambient temperature for 2 h followed by overnight at 4 °C. Following incubation, the gel was washed three times with 500 µL of buffer 2. The gel was re-suspended with 100 µL of digest buffer, and 1 µg of trypsin was added. The samples were allowed to digest overnight while rocking at 37 °C. The digest buffer was separated from the gel using Handee spin columns, and 50 µL was injected for analysis. Mass Spectrometric Conditions. Mass spectrometric detection was accomplished using a Finnigan TSQ Quantum Ultra (San Jose, CA) operated in positive ESI mode. The following mass spectrometric settings were used: scan width 0.1 amu, scan time 0.1 s, Q1/Q3 widths 0.7/0.9 amu, chrom filter 5 s, Q2 CID (Ar) 1.5 mTorr, spray voltage 3800, sheath gas (N2) 50 psi, aux gas (N2) 15 psi, capillary temperature 350 °C, quad MS/MS bias -1.6 V, source CID 5 V, and MS acquisition time 4.25 min. Following digestion, selected reaction monitoring (SRM) was used to detect LLELIR and LLELI7R. The SRM transition for each analyte was as follows: LLELIR, m/z 378.8 f 530.3 (CE 14 V, tube lens 117 V); LLELI7R, m/z 382.2 f 537.3 (CE 14 V, tube lens 117 V). To maximize selectivity, the [M + 2H]2+ precursor ions and the highest intensity, singly charged product ions were selected for the SRM transitions. The product ion spectrum of LLELIR is presented in Figure 2. Chromatographic Conditions. Chromatographic separation was performed using a Shimadzu HPLC system (Wood Dale, IL) that consisted of four LC10AD pumps, two SCL-10Avp controllers, and an HTS PAL Leap autosampler (Carrboro, NC). Sample desalting and cleanup steps were performed online using column Analytical Chemistry, Vol. 80, No. 3, February 1, 2008

563

Table 1. Intra- and Interday Validation Statistics for NTproBNP Spiked into Human Serum validation sample concn (ng/mL) day

statistic

0.10

0.50

1

mean (ng/mL) accuracy (% RE) precision (% CV) n

0.11 10.0 17.2 5

0.51 2.0 5.6 5

2.6 2.4 8.1 5

2

mean (ng/mL) accuracy (% RE) precision (% CV) n

0.11 12.0 8.1 5

0.54 8.2 15.1 5

2.4 -4.2 9.0 5

3

mean (ng/mL) accuracy (% RE) precision (% CV) n

0.097 -3.4 15.0 5

0.57 14.9 11.6 5

2.8 11.8 7.2 5

4

mean (ng/mL) accuracy (% RE) precision (% CV) n

0.10 -1.2 17.7 4

0.43 15.1 10.8 5

2.3 -6.7 5.2 5

all

mean (ng/mL) accuracy (% RE) precision (% CV) n

0.10 4.2 14.5 19

0.51 2.5 10.8 20

2.5 0.0 7.4 20

Figure 3. Column-switching diagram presented in the inject position.

switching and a ThermoScientific Sprite Aquasil C18 column (20 × 2 mm, 5 µm). The following mobile phases (MP) were used: MP A and C were 1000:1 HPLC water/88% formic acid (v/v), MP B was acetonitrile, and MP D was 800:200:1 methanol/acetonitrile/ 88% formic acid (v/v/v). A CapCell-Pak C18 MG (1 × 35 mm, 5 µm) analytical column was used following the trap column for chromatographic separation (Phenomenex). The LC separation employed column switching (Figure 3), which occurred in a three-step process: Step one, 50-µL sample aliquots were injected onto the trap column (ambient temperature) using 5% MP B in A delivered at 0.5 mL/min. The effluent was delivered to waste. Step two, at 1.0 min, the trap and analytical columns were placed in series, and a MP C/D linear gradient at 0.18 mL/min was used to elute LLELIR and LLELI7R into the mass spectrometer. Step 3, at 2.8 min, the trap and analytical columns were placed in parallel (starting conditions) for re-equilibration and the next injection. The analytical column was held at 55 °C, and its gradient profile was (min/% MP D in C) 0.0/10, 1.0/switch valve, 1.0/10, 2.8/switch valve, 4.0/55, 4.0/10, 4.25/10. Validation Experiments. The accuracy and precision of the assay to measure NTproBNP was evaluated on each of four days by analyzing five replicates of three concentrations (0.10, 0.50, and 2.5 ng/mL). The validation samples were bracketed by duplicate standard curves that spanned the dynamic range of 0.10-2.5 ng/mL NTproBNP (13-329 pg/mL LLELIR). Assay intra- and interday accuracy (percent relative error, % RE) and precision (percent coefficient of variation, % CV) were calculated and are presented in Table 1. Assay selectivity was evaluated during the validation through the analysis of blank samples (no NTproBNP, spiked SIL) prepared from three different lots of pooled human serum (10 patients/pool); the presence of interfering peaks in rat serum could not be measured due to endogenous NTproBNP. In addition, analyte carryover was evaluated by analyzing solvent blanks following the highest standard sample. The percent absolute carryover and percent relative carryover (to the LLOQ) were calculated and are presented in the Results and Discussion. Because of endogenous NTproBNP in rat serum, human serum was used as the control matrix during the validation and study sample analysis. This approach was validated using standard addition in which control rat serum was supplemented with 0, 0.4, 0.8, and 1.2 ng/mL NTproBNP. The endogenous rat serum level was measured using a standard curve prepared in human serum (human serum does not contain rat NTproBNP). This measurement was used to assign nominal concentrations to the supplemented rat serum samples, which were analyzed in triplicate along with duplicate standard curves in human serum. Finally, the percent difference between the nominal (endogenous NTproBNP + supplemented) and the measured concentrations was 564 Analytical Chemistry, Vol. 80, No. 3, February 1, 2008

2.5

used to establish that standards prepared in human serum could be used to accurately and precisely predict rat serum concentrations. NTproBNP and SIL stock solutions were prepared in 100:0.1 HPLC water/88% formic acid (v/v). These solutions were stored at ∼4 °C when not in use. The stability of the stock solutions was evaluated by comparing the results obtained from replicate injections made over time to an initial mean value obtained following the preparation of the solutions. Room-temperature matrix stability was evaluated in rat serum over periods of 4 and 24 h. Rat serum was supplemented with 0.4 ng/mL NTproBNP (endogenous plus 0.4 ng/mL), and three 100µL aliquots from each pool were incubated for 4 or 24 h at room temperature. Following incubation, the stability samples were immunoprecipitated, digested, and analyzed by LC/MS/MS. Room-temperature stability was evaluated by comparing the incubated samples to freshly prepared (T ) 0) samples. Longterm freezer and freeze/thaw stability were not evaluated during the validation, but will be evaluated during study sample analysis if the need for long-term freezer storage arises. Extract stability was evaluated during each analysis by bracketing the validation or study samples with duplicate standard curves. Any degradation of the analyte over the course of the analysis would cause the slopes of the two standard curves to diverge. Calibration curves were obtained by plotting the peak area ratio of LLELIR to its internal standard versus concentration. A weighted (1/concentration2) least-squares regression was used to obtain a linear equation over the range of the calibration graph. The origin was not used in the standard curve calculations. RESULTS AND DISCUSSION Method Development. Clinical evidence suggests the possibility of multiple structural forms of circulating BNP and that these forms may present a selectivity issue with point-of-care and

RIA-based assays.12 These data, taken with intraindividual BNP variability that has been reported to be 28.4% over the course of 1 week (21.1% for NTproBNP),13 led to the conclusion that an immunoaffinity-based LC/MS/MS assay for NTproBNP may be advantageous in evaluating cardiac hypertrophy in rat. This choice was further supported by a report of BNP instability and high clearance in man relative to NTproBNP,2 and by investigative studies in rat using a commercially available ELISA assay for BNP. The ELISA results were contradicted by preproBNP mRNA transcript levels and by increases in heart weight (unpublished data), further suggesting issues with selectivity or stability. During assay development, we had difficulty optimizing digestion conditions using trypsin. Our typical protocol, which involves using a trypsin/protein ratio of 1:20 followed by overnight incubation at 37 °C, resulted in less than 10% of the expected fragment LLELIR being generated. It was empirically determined that a trypsin/protein ratio of 4000:1 was required for optimum digestion efficiency. This can be explained by the presence of a glutamic acid residue on the C-terminal side of the arginine in the tryptic fragment LLELIR(E). According to the Promega package insert, the activity of trypsin is decreased when an acidic residue is present on either side of a susceptible bond. Given the large excess of enzyme required for digestion, Promega Trypsin Gold was chosen to avoid increasing the chemical background from autolysis. The recovery of NTproBNP was affected by multiple factors including immunoprecipitation, digestion, and ionization suppression. The absolute recovery, which includes all of these factors, was estimated by comparing the signal obtained from an extracted standard at the lower limit of quantitation (100 pg/mL NTproBNP) to a neat injection of the synthetic peptide LLELIR. The absolute recovery was determined to be ∼63%. Assay Validation. The assay as described was linear over the range of 0.10-2.5 ng/mL on NTproBNP, which corresponds to 13-329 pg/mL of the tryptic fragment. The results of the validation with respect to intraday and interday accuracy and precision are presented in Table. 1. The interday accuracy (% RE) and precision (% CV) of the assay did not exceed 4.2 and 14.5%, respectively, over the range of validation concentrations. Representative extracted ion chromatograms obtained from the analysis of a standard sample at the LLOQ and of a blank sample are presented in Figures 4 and 5. The presence of NTproBNP in serum obtained from healthy rats complicated the preparation of standard samples and the overall validation of the assay. Consequently, human serum was used as the control matrix during the validation and for the analysis of rat study samples; the rat form of NTproBNP is absent in human serum. To validate this approach, parallelism was established between human and rat serum (i.e., the same increase in response per unit increase in concentration of NTproBNP was obtained using both human and rat serum). This was accomplished by supplementing rat serum with NTproBNP and evaluating the precision and accuracy of replicate analyses at four (12) Hawkridge, A. M.; Heublein, D. M.; Bergen, H. R., 3rd; Cataliotti, A.; Burnett, J. C., Jr.; Muddiman, D. C. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 1744217447. (13) O’Hanlon, R.; O’Shea, P.; Ledwidge, M.; O’Loughlin, C.; Lange, S.; Conlon, C.; Phelan, D.; Cunningham, S.; McDonald, K. J. Card. Failure 2007, 13, 50-55.

Figure 4. Extracted ion chromatogram obtained from the analysis of human serum spiked with NTproBNP at the LLOQ (0.1 ng/mL). The tryptic peptide LLELIR (A) was monitored at the SRM transition m/z 378.8 f 530.3 and the stable isotope-labeled version LLELI7R (B) at m/z 382.2 f 537.3.

Figure 5. Extracted ion chromatogram obtained from the analysis of a blank human serum sample. The retention time of LLELIR (A) is indicated by an arrow. The blank was spiked with internal standard (B).

concentrations relative to standards prepared in human serum. The precision and accuracy results obtained from the analysis of rat serum QC samples are presented in Table 2. Key measures of assay selectivity were investigated during the validation. The absence of background in human serum that could interfere with the analyte or internal standard was demonstrated during each analysis using blank (( internal standard) samples. Three different lots of pooled human serum were evaluated, and the background present was determined to be less than 16% of the LLOQ in all cases; the signal-to-noise ratio of the LLOQ was typically 10:1. Analyte carryover succeeding the highest standard was 0.006%, which represents only 0.2% of the LLOQ. Endogenous NTproBNP in rat serum varies significantly. The endogenous level measured in the rat serum used to evaluate accuracy and precision was 110 pg/mL, while a commercial source had concentrations in the 1 ng/mL range, which could potentially be explained by differences in diet or blood collection method. The basal levels measured in the rat hypertrophy study (see Figure 6) ranged from 55 to 132 pg/mL. Analytical Chemistry, Vol. 80, No. 3, February 1, 2008

565

Table 2. Intraday Validation Statistics for NtproBNP Spiked into Rat Serum rat serum supplemented with NTproBNP (ng/mL)a statistic

0 (basal)

0.4

0.8

1.2

mean (ng/mL) accuracy (%RE) precision (%CV) n

0.11 nab 11.5 3

0.55 -20.0 9.6 3

0.84 -13.1 13.6 3

1.2 -8.8 1.8 3

a Control rat serum was supplemented with 0.4, 0.8, and 1.2 ng/mL NTproBNP. These samples were interpolated from a calibration graph obtained from the analysis of human serum standards to demonstrate the validity of using human serum as the control matrix for rat serum analyses. b Accuracy cannot be calculated for the basal level because the nominal concentration is unknown (the measured concentration was estimated to be 0.11 ng/mL).

Figure 6. NTproBNP concentrations in rat serum correlate with heart weight increase following administration of a compound known to induce cardiac hypertrophy (p < 0.0001).

Analyte stability in solution and in rat serum was studied during method development. The synthetic peptide LLELIR was found to be stable in 50:50 water/acetonitrile (v/v) for at least 4 months, and NTproBNP was found to be stable in 0.1% formic acid for at least 3 months, when the solutions were stored at 4 °C. Roomtemperature stability was evaluated in rat serum, and NTproBNP was found to be stable for at least 4 h. It should be noted that NTproBNP failed 24-h room-temperature stability criteria when a 67% decrease in response was observed relative to freshly prepared samples. Consequently, precautions should be taken to minimize the time that study samples remain at room temperature. Extract stability was demonstrated through the use of bracketing (front/ back) standard curves for all analyses. Significant differences

566 Analytical Chemistry, Vol. 80, No. 3, February 1, 2008

between the curves were not observed as determined by backcalculated residual errors, which indicates that the analyte is stable in the injection solvent for at least the duration of an analytical run. Cardiac Hypertrophy Study. Following 14 days of a oncedaily (50 mg/kg) dose of an agent known to induce cardiac hypertrophy, the treatment group demonstrated a significant increase in NTproBNP (0.332 ( 0.03 ng/mL) relative to the control group (0.107 ( 0.009 ng/mL); see Figure 6 (p < 0.0001). The significance of this finding with respect to cardiac hypertrophy was corroborated by a 34% increase in heart weight in the treatment group and by MRI of the left ventricles of both groups. Currently, we are conducting a study with a design similar to the one presented here, but with blood collection at 1, 3, 7, 10, and 15 days postdose. Preliminary results indicate that NTproBNP levels can predict cardiac hypertrophy at three days postadministration (unpublished data). CONCLUSION Currently, there is an unmet need for a biomarker to predict drug-induced cardiac hypertrophy in veterinary species. The natriuretic peptides are well established in the clinic as biochemical markers of heart failure, and we are currently evaluating their use for the aforementioned purpose in rat. Due to the potential for spurious results when using immunoaffinity-based approaches to measure natriuretic peptides, and for high variability of BNP levels due to issues with clearance and stability, an LC/MS/MS method was developed for NTproBNP. The purpose of this assay is 3-fold: (1) support the verification of NTproBNP as a marker for cardiac hypertrophy in rat, (2) help guide the investment in the costly development of an ELISA assay, and (3) assist in the evaluation of ELISA selectivity. Following validation, the assay was used to predict drug-induced cardiac hypertrophy following administration of an agent known to cause this condition, which has provided the impetus to begin the development of an ELISA assay. Research is currently ongoing to transfer the centrifuge tube, protein A/G gel-based immunoprecipitation to a protein A/G ELISA plate format, followed by the use of microwave-assisted proteolytic digestion, to dramatically increase assay throughput. Additionally, the use of µLC and nanoESI is being investigated to improve assay sensitivity. These measures are being taken in an attempt to decrease the gap in sensitivity and throughput between IA-LC/MS and ELISA techniques.

Received for review December 11, 2007. AC702311M

November

8,

2007.

Accepted