| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Abstracts of Oak Ridge Posters |
1 Ciphergen Biosystems, 6611 Dumbarton Circle, Fremont, CA 94555
aauthor for correspondence: fax 510-505-2101, e-mail efung{at}ciphergen.com
Ovarian cancer is the fifth leading cause of death from cancer in women and is the leading cause of gynecologic death in developed countries. Although the overall 5-year survival rate of patients with ovarian cancer is only 25%, the survival rate of patients diagnosed with early-stage ovarian cancer (I/IIa) approaches 90% (1). Therefore, a noninvasive diagnostic test that can detect early-stage ovarian cancer would be highly desirable, particularly if it could be coupled to a second, noninvasive diagnostic test such as transvaginal ultrasound. In a recently completed multiinstitutional protein expression profiling study encompassing 503 women from four institutions, we developed a three-marker panel that could distinguish women with early-stage ovarian cancer from healthy controls with higher diagnostic accuracy than could CA125 (area under the curve, 92% vs 77%) (2). Purification and identification of the three markers revealed that they were a truncated form of transthyretin, full-length apolipoprotein A1, and an internal fragment of inter-
-trypsin inhibitor heavy chain 4 (ITIH4).
Transthyretin is known to be cysteinylated and glutathionylated in addition to being truncated (3). ITIH4 is cleaved at multiple sites, generating numerous cleavage products (4). Conventional immunoassays generally cannot simultaneously distinguish or quantify multiple posttranslationally modified forms of a protein or its isoforms. Mass spectrometry (MS)-based assays have the advantage of being able to do so, but are often not considered to be quantitative. Because of the chromatographic nature of the ProteinChip® array surface, surface-enhanced laser desorption/ionization time-of-flight (SELDI-TOF)-MS has the advantage of being quantitative (5). When immunocapture is coupled with SELDI-TOF-MS, a single immunoassay can quantify multiple forms of an analyte. The immunocapture can be performed directly on the ProteinChip array surface (by use of a reactive surface array) or on beads coated with antibody. In the latter case, the bound material is eluted onto ProteinChip arrays.
Several antibodies to transthyretin and apolipoprotein A1 were evaluated for their performance in SELDI-TOF-MS-based ProteinChip assays. Selection criteria included dynamic range of linear response, specificity as determined by the number of peaks detected in SELDI-TOF-MS immunoassays, and antibody stability as assessed by the quantity of intact antibody present in SELDI-TOF-MS quality-control assays of the reagents. The two commercially available antibodies selected were rabbit anti-human transthyretin antibody (cat. no. A0002; Dako Cytomation Corporation) and mouse anti-apolipoprotein A1 antibody (cat. no. 178474; EMD Biosciences, Inc.). To generate antibodies to ITIH4, a peptide corresponding to the discovered biomarker for ovarian cancer derived from ITIH4 was chemically synthesized with addition of cysteine at the NH2 terminus (SynPep, Dublin CA). The sequence is as follows: CMNFRPGVLSSRQLGLPGPPDVPDHAAYHPF. The peptide was conjugated to keyhole limpet hemocyanin and injected into rabbits according to a 69-day standard protocol used by SynPep. Rabbit bleeds were collected, and ELISA tests to determine titers were performed by SynPep. Antibodies were purified by use of protein A and affinity chromatography.
For the immunoassays, we covalently linked the antibody to AminoLink® Plus beads (Pierce Biotechnology). Beads were washed twice with 200 µL of phosphate-buffered saline (PBS), pH 7.2, containing 1 mL/L Tween. Calibrants for transthyretin were purchased from Kamiya (K-assay Prealbumin Calibrator reagents) and were diluted 1:50 in the same buffer to a final volume of 100 µL/assay well. For the ITIH4 assay, the antigen peptide stock described above was added to 100 mL/L normal human serum (cat. no. H-1388; Sigma) diluted in the same buffer to a final volume of 100 µL/assay well. In a Sigma serum sample with no added antigen peptide, no endogenous peptide was detected. A serum titration curve was generated for the apolipoprotein A1 assay by use of pooled human serum (Intergen) diluted in the same buffer to a final volume of 100 µL/assay well. The amount of apolipoprotein A1 in this particular lot of human serum was measured beforehand by ELISA in an independent laboratory and was reserved in aliquots solely for this purpose.
Reactions were performed in 96-well filter plates at room temperature for 2 h on a MicroMix shaker (DPC). At the end of the incubation, the beads were washed three times (3 min each time) by shaking with 200 µL of PBS, pH 7.2, containing 1 mL/L Tween. The captured material was eluted by three additions of 50 µL of an organic elution buffer (333 mL/L isopropanol–167 mL/L acetonitrile–1 mL/L trifluoroacetic acid–1 mL/L CHAPS), with incubation for 5 min each time with shaking. The three eluates were pooled in a 96-well V-bottomed plate by centrifugation at 671g for 1 min in a desktop Sorvall. Array binding was performed immediately, or the eluates were stored at –80 °C before additional processing. In addition, ELISA was performed on the standard serum; the reference values obtained were 257 mg/L (reference interval, 200–400 mg/L) for transthyretin and 1300 mg/L (reference interval, 880-1800 mg/L) for apolipoprotein A1. No ELISA is available for ITIH4 or its individual peptides.
All ProteinChip array incubations were performed in 96-well bioprocessors. CM10 ProteinChip arrays (Ciphergen) were preequilibrated with binding buffer (150 µL of 100 mmol/L sodium acetate, pH 4.0) twice, 5 min each time. For the transthyretin and apolipoprotein A1 immunoassays, 40 µL of the organic eluate was mixed with 60 µL of 100 mmol/L sodium acetate, pH 4.0, buffer in the wells. For the ITIH4 assay, 50 µL of the organic eluate was mixed with 50 µL of 100 mmol/L sodium acetate, pH 4.0, buffer. Array binding was performed at room temperature for 30 min with shaking. The arrays were washed three times for 5 min each with 200 µL of pH 4.0 binding buffer, with shaking. The arrays were rinsed with water once for 1 min and then briefly allowed to air dry. For the transthyretin and apolipoprotein A1 assays, 1 µL of 12.5 mg/L sinapinic acid (SPA) was added to each spot, dried, and then reapplied. For ITIH4, 1 µL of 20% saturated -cyano-4-hydroxy-cinnamic acid (CHCA) was added to each spot, dried, and then reapplied. For the external normalization purpose, 5 fmol of bovine insulin was mixed with CHCA and applied to each spot for the ITIH4 assay. SPA and CHCA were prepared in 500 mL/L acetonitrile–5 mL/L trifluoroacetic acid. Matrix was applied using the Biomek 2000 (Beckman).
The arrays were read in a PBSIIc ProteinChip reader, a time-lag focusing, linear, laser desorption/ionization time-of-flight mass spectrometer. All spectra were acquired in the positive-ion mode. Time-lag focusing delay times were set at 400 ns for ITIH4, 900 ns for transthyretin, and 1200 ns for apolipoprotein A1. Ions were extracted by use of a 2.9-kV ion extraction pulse and were accelerated to final velocity by use of 20 kV of acceleration potential. The system used a pulsed nitrogen laser at repetition rates varying from 2 to 5 pulses/s. Typical laser fluence varied from 30 to 150 µJ/mm2. An automated analytical protocol was used to control the data acquisition process in most of the sample analysis. Each spectrum was the mean of at least 50 laser shots and was externally calibrated against a mixture of known peptides or proteins. Mass accuracy was 
0.05%.
An example of the SELDI-TOF-MS assay of the three markers is shown in Fig. 1
. The SELDI-TOF-MS transthyretin immunoassay, in contrast to the traditional nephelometric method, can simultaneously quantify and distinguish four forms of transthyretin: unmodified, cysteinylated, glutathionylated, and truncated. In our experience, cysteinylated transthyretin is the largest peak of these forms, and the truncated form is the smallest (Fig. 1A
). Although the SELDI-TOF-MS immunoassay for the ITIH4 peptide showed a single peak when performed on the peptide (Fig. 1B
), immunoassays performed on human serum revealed the antigenic peptide as well as a series of smaller peptides representing sequential N-terminal truncation. Although these truncations may arise in vivo or ex vivo, depending on the stability of the peptides, the addition of protease inhibitors had no effect on the intensity or pattern of fragmentation, suggesting that these truncations resulted from in vivo protease activity. Incubation for longer periods of time decreased the overall peak heights but did not alter the pattern of fragmentation. Incubation at room temperature gave slightly lower peak heights than incubation at 4 °C, but like the length of incubation and addition of protease inhibitors, had no effect in overall pattern of fragmentation (data not shown). Results obtained with the apolipoprotein A1 assay are shown in Fig. 1C
. The bottom graph shows the titration curve for apolipoprotein A1. The x axis shows the calculated amount of apolipoprotein A1 according to the ELISA data.
|
Data from a sample calibration curve for the ITIH4 assay are shown in Table 1
, along with the CV. The limit of detection (value with a signal-to-noise ratio of 5) was 0.037 mg/L. The range of linearity for the ITIH4 was almost two orders of magnitude. Within-run reproducibility was measured by assaying the pooled Intergen human serum five times on one plate; the mean (SD) concentration of the ITIH4 fragment (3273 Da) was 1.10 (0.2) mg/L (n = 5; CV = 14%). Between-run reproducibility was measured by assaying the pooled Intergen serum 10 times in separate assays; the mean concentration of the ITIH4 fragment (3273 Da) in the Intergen serum was 1.11 mg/L (n = 16; CV = 22%). Recovery experiments were done by measuring ITIH4 peptide added to PBS-Tween buffer as well in 100 mL/L Sigma human serum (which does not contain endogenous ITIH4 peptide). We calculated the amount of the peptide added to Sigma serum, based on the calibration curve generated with the PBS/Tween buffer, and compared this value with the predicted value. The mean recovery was 82% (range, 61–102%).
|
Several groups have used proteomics approaches to identify candidate biomarkers for ovarian cancer (6)(7)(8). Some approaches do not include purification and identification of candidate markers, choosing to rely purely on the pattern derived through multivariate statistics. In other cases, a candidate marker was identified and assayed by conventional techniques. However, SELDI-TOF-MS immunoassays can provide quantification of multiple analytes. Because these assays can simultaneously quantify posttranslationally modified forms of the target analyte, this approach may be preferred when quantification of these forms is required for diagnostic accuracy (9)(10)(11)(12). These assays are currently being tested on a large series of serum samples from patients with ovarian cancer to determine which specific combination of posttranslationally modified forms of these proteins confers the highest diagnostic accuracy.
Acknowledgments
We thank an anonymous reviewer for many helpful suggestions.
References
subunit as potential serum biomarker in ovarian cancer: identification and characterization using proteomic profiling and mass spectrometry. Clin Cancer Res 2003;9:2904-2911. in renal cancer patients. Lab Invest 2004;84:845-856.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  D. de Boer, M. M. Erps, W. K.W.H. Wodzig, and M. P. van Dieijen-Visser Inadequate Attempts to Measure the Microheterogeneity of Transthyretin by Low-Resolution Mass Spectrometry Clin. Chem., July 1, 2005; 51(7): 1299 - 1300. [Full Text] [PDF]
|
|
|
|
|
|
E. T. Fung Inadequate Attempts to Measure the Microheterogeneity of Transthyretin by Low-Resolution Mass Spectrometry - Reply Clin. Chem., July 1, 2005; 51(7): 1300 - 1300. [Full Text] [PDF]
|
|
|
|
|
|
F. J. Schweigert and J. Raila Inadequate Attempts to Measure the Microheterogeneity of Transthyretin by Low-Resolution Mass Spectrometry - Reply Clin. Chem., July 1, 2005; 51(7): 1300 - 1301. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, 13.8 kDa (y = –0.0034x2 + 0.765x – 0.5336; R2 = 0.9963); 
, 13.7 kDa (y = –0.001x2 + 0.3228x – 0.2185; R2 = 0.9957); 
, 14.0 kDa (y = –0.0012x2 + 0.2895x – 0.1516; R2 = 0.9956). (B), SELDI immunoassay using polyclonal antibody to ITIH4. The spectrum (top) shows capture of ITIH4 antigen peptide added to 100 mL/L reference serum The peptide in the reference serum is undetectable. The graph (bottom) shows the calibration curve for ITIH4 peptide capture. Data were normalized against an externally added calibrant, insulin. Equation for the line: y = –3.4409x2 + 25.36x + 0.0948 (R2 = 0.9999). (C), SELDI immunoassay using monoclonal antibody to apolipoprotein A1 (ApoA1). The spectrum (top) shows capture of apolipoproteinA1 from pooled normal human serum. The graph (bottom) shows the titration curve for apolipoprotein A1. Equation for the line: y = –0.003x2 + 0.6153x – 0.0424 (R2 = 0.9959). Error bars, SD.