| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
Department of Clinical Chemistry, University Hospital Maastricht, Maastricht, The Netherlands
aaddress correspondence to this author at: Department of Clinical Chemistry, University Hospital Maastricht, PO Box 5800, Maastricht, 6202 AZ The Netherlands; fax 31-43-387-4692, e-mail Dieijen{at}klinchem.azm.nl
Purification and identification of small, low-abundance proteins from complex samples such as serum for use in Western blotting experiments has long been a challenge. Unfortunately, commercially available methods are not always suitable for small sample sizes and low-abundance proteins such as cardiac troponin T (cTnT). In this report we present a highly sensitive immunoprecipitation method we have developed for extracting and concentrating low-abundance proteins from serum. We compared the performance of our method with that of the commercially available ImmunoPure® Protein-A IgG orientation Kit (Pierce), using antibodies directed against cTnT and serum of a patient with acute myocardial infarction (AMI). Both methods are based on the binding of antibodies to Sepharose beads and subsequent cross-linking with a diimido ester.
The formation of amidines between imido esters and amines from protein side chains was first described in 1962 by Hunter et al. (1), who suggested this for cross-linking purposes. Levy et al. (2) described the use of diimido esters for covalent coupling of antibodies to immobilized antigens. Later studies reported similar methods for antibody immobilization with cross-linkers such as dimethyl pimelimidate and dimethyl suberimidate (3)(4).
These methods are based on the fact that an amine is able to react with an imido ester only if it is deprotonated, which occurs at pH > pKa. At these pH values, the nucleophilic attack of an R-NH2 on one of the two outer backbone carbon atoms of dimethyl pimelimidate produces a tetrahedral intermediate that splits into the amidine link and methanol. Hunter et al. (1) showed that the right choice of pH even allows a distinction to be made between 
- and 
-amines. They showed that a pH between 7 and 8 favors reaction with -amines, whereas a pH between 9 and 10 favors reaction with -amines. At a pH between 8 and 9, both reactions occur at a significant rate. In a protein, only the N-terminal ending of a peptide chain contains an -amine. The remaining amines are -amines from the amino acids lysine or arginine. The different pKa values for lysine show that with increasing pH, the -amine is deprotonated first (pKa = 9.0) with the -amine deprotonated subsequently (pKa = 10.5). The -amine of arginine (pKa = 12.5) is still protonated at pH values between 8 and 9, which leaves, in addition to the N-terminal -amine, only the -amine from lysine residues available for reaction. Because the variable regions of mouse IgG are generally almost devoid of lysine residues [Kabat Database (5)], binding will predominantly be at constant regions, minimizing interference of the cross-linker with the antigen-binding capacity of the antibody at pH 8.6.
To evaluate both methods, we used serum samples that were obtained from a patient with AMI. In addition to a total protein concentration of 50 g/L, these samples contained a relatively small amount of cTnT. Sample 1 was taken 16 h after onset of symptoms (cTnT = 33.44 µg/L), and sample 2 was taken 238 h after onset of symptoms (cTnT = 5.53 µg/L). cTnT concentrations were measured on the Elecsys 2010 (third-generation cTnT test; Roche).
All reactions were performed at room temperature, and phosphate-buffered saline (PBS) was used at pH 7.0 unless stated otherwise. We washed 100 mg of protein A-Sepharose (Pharmacia Biotech) once with PBS and blocked it with 1 g/L bovine serum albumin (Sigma) in PBS for 60 min. A mixture of anti-cTnT antibodies [9G6, 7F4, 1C11 (all from Research Diagnostics Inc.), and 4C5 (from BioSpacific/Fortron); 70 µg of each antibody] in PBS containing 1 g/L bovine serum albumin was mixed with the Sepharose beads and rotated for 1 h. After binding of the antibodies, the beads were washed twice with excessive PBS. Antibodies were cross-linked to protein A-Sepharose by the addition of 200 mmol/L triethanolamine in PBS to which 20 mmol/L dimethyl pimelimidate (Sigma) was added directly before use (final pH 8.6). After rotation for 30 min, the beads were washed with 200 mmol/L triethanolamine in PBS. In contrast to the method by Schneider et al. (4) and the modified method of Sisson and Castor (6), we repeated the cross-linking and washing twice to improve cross-linking efficiency. The remaining reactive amino groups were quenched by addition of 50 mmol/L ethanolamine in PBS for 60 min. Non-cross-linked antibodies were removed by incubation twice (20 min each time) with 1.0 mol/L glycine-HCl (pH 3.0) at 56 °C. Cross-linked beads were stored at 4 °C in PBS containing 0.2 mL/L Tween-20 (PBST) and 0.2 g/L sodium azide until use. The Sepharose beads were loaded in the column of the ImmunoPure Protein-A IgG Orientation Kit and were cross-linked with the same mixture of anti-cTnT monoclonal antibodies (285 µg of each antibody), according to the manufacturer’s instructions.
The immunoprecipitation of troponin T (39 kDa) required 8 mg (dry weight) of antibody-coupled Sepharose beads per sample for our method. To each tube, we added 250 µL of serum, 100 µL of 6 mol/L urea, and 150 µL wash buffer (PBST); we then added 7 g/L nonfat dry milk (Protifar Plus; Nutricia). Beads were rotated for 90 min and then washed twice with PBST. Elution was performed by heating (56 °C) for 20 min in 100 µL of 1.0 mol/L glycine (pH 3.0). The sample was centrifuged, and the supernatant was kept for analysis.
For the ImmunoPure Protein-A IgG Orientation Kit, 750 µL of serum (diluted in 300 µL of 6 mol/L urea) and 750 µL wash buffer were added to the column. After incubation for 90 min, the column was washed twice with PBS. Elution was performed with four 1-mL volumes of 1.0 mol/L glycine-HCl (pH 3.0). The fractions were tested for cTnT, and on the basis of these results, fraction 3 was collected for analysis.
We prepared samples for electrophoresis by adding 20 µL of sample buffer (40 mmol/L Tris, 33 g/L SDS, 500 mL/L glycerol, and bromphenol blue) to 80 µL of eluate. We applied 20 µL of this mixture to a 4–15% linear gradient Tris-HCl polyacrylamide precast gel (Bio-Rad). The Precision Plus Protein Standard (Bio-Rad) was used as the molecular mass marker. After stacking for 15 min at 100 V and running for 40 min at 150 V, the gel was blotted on a nitrocellulose membrane (Bio-Rad) at 4 °C at 100 V for 60 min. Each membrane was blocked for 60 min in PBS containing 33 g/L nonfat dry milk. The primary anti-cTnT antibody (4C5) was added at a 1:1000 dilution in wash buffer, and the membrane was incubated overnight at 4 °C. Control experiments with a different anti-cTnT antibody (9G6; mapped epitope, residues 1–60) were also performed (see Fig. 1
in the Data Supplement that accompanies the online version of this Technical Brief at http://www.clinchem.org/content/vol51/issue1/). The membrane was washed three times in wash buffer; the secondary antibody (peroxidase-labeled goat-anti-mouse; Dako) was then added as a 1:5000 dilution in wash buffer and incubated for 60 min at 4 °C. The membrane was washed four times with wash buffer and finally once in PBS. Membranes were developed with Enhanced Chemiluminescence Buffer and captured on Kodak X-Omat Blue film (both from Perkin-Elmer Life Sciences).
|
We checked various conditions, e.g., type of beads and reaction temperature, to determine their optimums. Protein A-Sepharose was preferred over protein G-Sepharose because the latter showed more background signal in the Western blots. Temperatures (37, 56, 65, and 100 °C) for the elution of cTnT from the beads showed optimal results at 56 °C. Finally, the difference between goat anti-mouse and rabbit anti-mouse secondary horseradish peroxidase-labeled antibodies was investigated. The former gave slightly stronger band intensities.
Before interpreting the obtained Western blots, we verified several aspects of the assay. We first precipitated a positive control to check whether the protein of interest was modified or degraded during the process. The positive control showed only one band at 39 kDa (Fig. 1
). We used a negative control serum to check for binding of proteins that would give nonspecific bands: no bands were visible. A comparison between the results of the two methods showed that the method using the column (Fig. 1
, lanes 3 and 5) gave a lower signal. Although (much) longer exposure times revealed comparable protein bands, indicating similar purifying properties, the background noise was becoming too high (data not shown).
Using diimido esters as a cross-linker between antibodies and protein A-Sepharose, we have developed a powerful and highly sensitive immunoprecipitation method for extracting and concentrating low-abundance proteins from human serum with minimal interference from high concentrations of total protein and human immunoglobulins. One important property of our method is that the amount of Sepharose beads can easily be adjusted to the amount of sample available. This has two advantages: (a) To prevent contamination, it is best to allow each sample a fresh batch of Sepharose beads. This can easily be done with our method, whereas with columns, this could be more expensive (in terms of the number of columns and antibodies needed) and requires larger sample volumes. (b) Because of the low-volume elution, proteins remain present at high concentrations. This method is therefore suitable for extracting and concentrating low-abundance (µg/L) proteins from small sample quantities without the risk of cross-contamination. For example, using this highly sensitive method, we were the first to be able to detect cTnT fragments in serum samples at concentrations below the 0.010 µg/L detection limit of the third-generation cTnT assay (7)(8) (see Fig. 2 in the online Data Supplement). Moreover, this method allows us to further investigate cTnT fragmentation and clearance in vivo.
Acknowledgments
We greatly acknowledge Vincent Kleijnen for technical assistance.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  M. N. Fahie-Wilson, E. M. Hall, and E. J. Lamb The authors of the article cited above respond: Clin. Chem., December 1, 2006; 52(12): 2307 - 2309. [Full Text] [PDF]
|
|
|
|
|
|
M. N. Fahie-Wilson, D. J. Carmichael, M. P. Delaney, P. E. Stevens, E. M. Hall, and E. J. Lamb Cardiac Troponin T Circulates in the Free, Intact Form in Patients with Kidney Failure Clin. Chem., March 1, 2006; 52(3): 414 - 420. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
