| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Proteomics and Protein Markers |
1 Department of Autoimmunology, Statens Serum Institut, Copenhagen, Denmark.
Departments of2
Clinical Biochemistry and3
Nephrology, Rigshospitalet, Copenhagen, Denmark.
4 Dialysis Unit, Department of Medicine, Storstrømmens Sygehus Nykøbing F, Nykøbing, Denmark.
5 Institute of Medical Anatomy, University of Copenhagen, Copenhagen, Denmark.
aAddress correspondence to this author at:, Statens Serum Institut, Bldg. 81, Room 536, Artillerivej 5, DK-2300 Copenhagen S, Denmark. Fax 45-32683876; e-mail nhe{at}ssi.dk.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: We used immunoaffinity–liquid chromatography–mass spectrometry analysis of serum samples to examine whether structurally modified ß2M is generated in the circulation. In addition, we developed an immunoassay for the quantification of a cleaved ß2M variant in biological fluids based on novel monoclonal antibodies and applied this assay to patient and control sera.
Results: A specific alteration compatible with the generation of lysine-58–cleaved and truncated ß2M (
K58-ß2M) was found in the sera of many (20%–40%) dialysis patients but not in control sera or sera from patients with cerebral amyloidosis (Alzheimer disease). Applied to patient sera, specific immunoassays revealed that dialysis, as expected, significantly lowered the total ß2M concentration, but the concentrations of K58-ß2M remained unchanged after dialysis. The results also show that patients dialyzed with less biocompatible membranes have higher serum concentrations of cleaved ß2M (mean, 8.5, 1.8, and 0.7 mg/L in cuprophane membrane-dialyzed, polysulfone membrane-dialyzed, and control sera, respectively).
Conclusions: This study for the first time demonstrates and assigns the structure of a specific ß2M variant in sera from dialysis patients. Because this variant is conformationally unstable in vitro, it may be involved in in vivo amyloidogenesis.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
33% of patients on chronic hemodialysis (CHD) develop histologically prevalent amyloidosis after 4 years of hemodialysis [dialysis-related amyloidosis (DRA)](7). The histologic prevalence of ß2M amyloidosis is similar in patients in chronic peritoneal dialysis(8), and patients with long-standing kidney disease may develop amyloidosis in the absence of dialysis treatment(9). It is still unknown why ß2M becomes amyloidogenic under CHD and kidney disease conditions or why the aggregates target joints and cartilaginous tissue. There is no simple relationship between total serum ß2M concentrations and the occurrence of DRA(10), and other conditions with chronically increased ß2M concentrations (e.g., hematologic cancers) in serum do not present with amyloidotic joint lesions. Thus, specific structural modifications of ß2M may be relevant for DRA and kidney disease-associated amyloidosis. We have recently shown that the cleavage of ß2M at lysine-58 that may take place in inflammatory sera leads to an impairment of its solubility because of increased conformational instability(11). Although most ß2M amyloid fibrils contain unmodified and fully intact ß2M(12), several structural variants, notably cleaved, truncated, and adduct-modified ß2M, have also been demonstrated in extracts of ß2M amyloid from patients(13)(14)(15)(16)(17)(18).
An important unanswered question is whether the modified species found in ß2M amyloids are modified before participating in amyloid formation in vivo or represent modifications taking place after amyloid deposition of normal ß2M. Thus, it is of interest to address the state of circulating ß2M in patients. It has long been known from electrophoretic analyses that charge-modified ß2M is present in the sera of some hemodialysis patients(19), but the exact structures of these ß2M species are undetermined, and a detailed characterization of the whole population of wild-type and modified ß2M molecules present in sera from patient on CHD has hitherto not been carried out. To address this issue, we used immunoaffinity–liquid chromatography–mass spectrometry (IA-LC-MS), combining the specificity of immunoaffinity isolation of biomolecules with the resolving power of MS(20) to search for ß2M variants in sera from patients on CHD and healthy controls. We found that several patients carry a modified form of circulating ß2M that is cleaved at and deficient in lysine-58 (K58-ß2M). To quantitatively characterize the presence of this specific variant, we subsequently developed K58-ß2M–specific monoclonal antibodies (mAbs) and immunoassays and used these tools to measure the K58-ß2M concentrations in sera from patient on CHD and from controls.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
K58-ß2M (see Fig. 2C
) were purified from urine of a nephropathy patient followed by cleavage as described previously(11)(21)(22)(23). Purified proteins were kept at 1–20 g/L in PBS (pH 7.4) at –20 °C until used. Monoclonal anti-ß2M antibody was produced at Statens Serum Institut (product no. Hyb 290-03; Antibodyshop). Polyclonal rabbit anti-human ß2M antibodies (product no. A0072) were obtained from Dako, and alkaline phosphatase-conjugated goat anti-rabbit immunoglobulin was from Sigma (cat. no. A-3812).
|
patient material
This study was approved by the Regional Ethical Committee of Copenhagen (case KF01-230/02). Blood samples from dialysis patients were obtained after informed consent and were allowed to clot at room temperature for 15 min. After centrifugation at 2000g for 5 min, sera were frozen immediately and stored at –20 °C. Sera from 2 groups of patients were examined by IA-LC-MS for ß2M modifications. Initially, 14 patients undergoing dialysis with cuprophane membranes were enrolled, and sera were obtained just before the initiation of dialysis. Additionally, 25 patients on CHD, all dialyzed with polysulfone membranes and without clinical evidence of DRA (9 women, 16 men; age range, 33–83 years; mean age, 61.28 years) were enrolled. These patients had all been undergoing dialysis treatment for <10 years. As controls, fresh samples from healthy laboratory personnel (n = 7) and from Alzheimer disease patients (n = 9) were collected and processed in the same way.
IA-LC-MS
Using an electrospray ionization time-of-flight mass spectrometer (Mariner; Applied Biosystems) coupled to a nano-HPLC system (Ultimate; Dionex), we prepared, installed, and processed immunoaffinity columns containing immobilized anti-ß2M mAb (Hyb 290-03) as described previously(20). Before analysis, sera were filtered through Ultrafree-MC centrifugal filters (pore size, 0.45 µm; Millipore). Deconvolution was based on 3 charge states of the raw spectra and performed with DataExplorer (Ver. 3.5) included in the Mariner software. The subsequent quantification of ß2M isoforms was based on the extracted ion chromatograms by use of charge states from native ß2M and K58-ß2M (5 for each of the oxidized and unoxidized variants). Ten selected representative ions within the regions between the ß2M peaks were subtracted from the ion chromatogram of the ß2M peaks to compensate for the contribution of background. MS experiments performed with known quantities of the 2 ß2M species ensured that their ionization efficiencies were comparable.
capillary electrophoresis
Capillary electrophoresis experiments were performed on a Beckman P/ACE 5010 instrument equipped with liquid sample cooling and ultraviolet detection. The electrophoresis buffer was 0.1 mol/L phosphate (0.081 mol/L Na2HPO4 and 0.019 mol/L NaH2PO4, pH 7.38), and samples were injected under pressure for 8 s, corresponding to a sample volume of 9 nL. Electrophoresis was carried out at a constant current of 90 µA. Detection took place at 200 nm, and the separation tube was an uncoated fused-silica capillary (50 µm i.d.; Polymicro Technologies or Beckman Coulter) with a total length of 47 cm (40 cm to the detector window). To test the binding specificity of mAbs, the antibodies were bound to protein G-Sepharose (Amersham) that had previously been washed with electrophoresis buffer. The mixture containing 100 µL of protein G-Sepharose slurry and 400 µL of mAb at 1 g/L in PBS was incubated for 10 min at 25 °C. After the mAb–protein G-Sepharose slurry was washed, it was incubated for 30 min at 25 °C with a mixture of 10 µL of wild-type ß2M (2 g/L) and 10 µL of K58-ß2M (2 g/L). The sample was centrifuged, and the supernatant containing the unbound molecules was analyzed by capillary electrophoresis. All samples were mixed with 0.04 g/L acetyl-Pro-Ser-Lys-Asp-OH, a marker peptide.
anti-K58-ß2M mAbs
A group of 5 female NMRI mice were preimmunized once with tuberculin and then immunized 6 times at biweekly intervals except for the second boost, which followed 4 weeks after the primary immunization. Venous blood in EDTA was obtained 10 days after each immunization. Immunizations involved intraperitoneal injections of the peptide immunogen S3-CVEHSDLSFS-OH at a dose of 25 µg/mouse in 500 µL of PBS containing 1 mg of Al(OH)3. The development of peptide-specific antibodies in mouse sera was monitored by an ELISA with ovalbumin-conjugated peptide coated at 1 µg/well in carbonate buffer (pH 9.6) and blocked with 20 g/L bovine serum albumin in PBS. The mouse plasma was incubated at a dilution of 1:10 in PBS for 1 h at room temperature, and the amount of bound mouse immunoglobulin was subsequently visualized by incubation for 1 h with a 1:2000 dilution of alkaline phosphatase-conjugated goat anti-mouse immunoglobulin antibody (Sigma-Aldrich) with p-nitrophenyl phosphate as the chromogen and measurement of absorbance at 405 nm. One mouse with a high titer response was subsequently sacrificed 3 days after intravenous boosting with 25 µg of the S3-peptide conjugate. The splenocytes from this mouse were then fused with P3X63Ag8.653 cells, and hybridomas were established by standard procedures. The hybridomas producing K58-ß2M–specific antibodies were selected by a capture ELISA in which the plates were coated with rabbit anti-ß2M (product no. A-0072; Dako) and blocked with bovine serum albumin. Half of the plate was subsequently incubated with ß2M and the other half with K58-ß2M. Incubation with hybridoma supernatants at a 1:10 dilution and visualization of immunoglobulin binding then were performed as described. In all, a total of 730 hybridoma supernatants were tested, and 4 of these cultures showed specificity toward K58-ß2M. Three cell lines (332-01, -02, and -03), which all produced antibodies reacting with K58-ß2M and not appreciably with wild-type ß2M, as judged by capillary electrophoresis and ELISA, were finally established.
elisa
K58-ß2M concentrations in sera from patients on CHD were measured quantitatively by a capture ELISA. The procedure was as follows: Wells of MaxiSorp microtiter plates (Nunc) were coated overnight at 4 °C with the anti-K58-ß2M mAb 332-01 at a concentration of 50 mg/L in PBS. The plates were blocked for 1 h at room temperature with 20 g/L bovine serum albumin in PBS and then incubated for 1 h with serum diluted 1:10 in PBS. After incubation with a 1:1000 dilution of rabbit anti-ß2M (product no. A-0072; Dako) in PBS for 1 h at room temperature, the plates were incubated with a 1:2000 dilution of alkaline phosphatase-conjugated goat anti-rabbit antibody (cat. no. A-3812; Sigma) in PBS for 1 h at room temperature. Visualization took place using p-nitrophenyl phosphate, and the absorbance in each well was measured at 405 nm. Purified K58-ß2M added in known amounts to serum from a healthy donor was used as an internal standard to calculate K58-ß2M concentrations in unknown samples. Detection limits were determined as the concentration of K58-ß2M in the applied sample (i.e., serum diluted 1:10) that gave a signal 3 times above the buffer blank. Total ß2M concentrations in sera were determined by an in-house ELISA.
statistical tests
We used the nonparametric Mann–Whitney test for comparisons of the K58-ß2M concentrations detected by ELISA in sera from the groups of healthy donors and CHD patients. P <0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). The system uses an immunoaffinity column containing immobilized anti-ß2M mAbs that capture ß2M from serum. Subsequently, the masses of the captured ß2M molecules are provided after the bound material is eluted over a desalting column directly into a mass spectrometer(20). Barring alterations of epitopes crucial for binding to the capture antibody, this approach yields a profile of the total population of both wild-type and altered ß2M molecules present in a particular sample. Specific alterations were found in ß2M in several patients on CHD. In control sera (healthy individuals and Alzheimer disease patients), only unoxidized wild-type ß2M and Met-99–oxidized(11) wild-type ß2M were detected (i.e., masses of 11 729.2 for the unoxidized wild-type and 11 745.2 for the oxidized wild-type ß2M). The oxidized subpopulation was found to a variable degree in all samples. In addition, 2 minor fractions of increased masses (approximately +98 and +213) were detected in some samples. The nature of these alterations is unknown, but they were not specifically found in CHD patients. In contrast, in 6 of the initially examined 14 CHD patients, and later in 5 of the 25 patients dialyzed with polysulfone membranes, we observed an increased amount of a ß2M variant with a decreased mass of 11 619 (and its oxidized form at 11 635 in various proportions), which was not detectable in any of the controls (Fig. 1
).
|
The identity of the circulating mass-deficient ß2M variant that was found in 20%–40% of the examined CHD patients is most likely a cleaved form of ß2M with lysine-58 removed (K58-ß2M). This species has a theoretical molecular weight of 11 619.0 and has been reported to be increased in sera from patients with inflammatory diseases or cancer(24)(25)(26)(27). For comparison, the mass spectra of purified wild-type ß2M and K58-ß2M are shown in panels A and B, respectively, of Fig. 2
, and the schematic structure of K58-ß2M is shown in Fig. 2C
. To obtain a relative measure of the amounts of K58-ß2M compared with native ß2M detected by the IA-LC-MS system, the ion chromatograms of the mass spectral analyses were quantified with subtraction of background as described in the Materials and Methods. Previous experiments ensured that purified K58-ß2M and native ß2M did not bind differentially to the immobilized antibody and that they ionized similarly(20). In practice, the lower limit of the MS-based estimate of the K58-ß2M/ß2M ratio in serum analyses is 5%. However, IA-LC-MS analysis results showed that in some CHD patients, >20% of total serum ß2M was in the K58-ß2M form (Fig. 2D
).
To get a quantitative measure of the K58-ß2M concentrations, we developed specific mAbs against K58-ß2M (see the Data Supplement that accompanies the online version of this article at http://www.clinchem.org/content/vol51/issue7/) and used them as the basis for constructing solid-phase immunoassays. Our aim was to get high-affinity IgG mAbs that would be reactive with K58-ß2M in solution. Initial attempts using purified, whole K58-ß2M as the immunogen failed to yield K58-ß2M–specific clones. We therefore used a decameric synthetic peptide corresponding to the new COOH terminus appearing in K58-ß2M (see Fig. 2C
) for the immunizations. Three clones that reacted specifically with purified K58-ß2M but not with wild-type ß2M were produced in this way. The specificity of 1 of these clones (332-01) was subsequently ascertained by capillary electrophoresis of mixtures of immobilized antibodies and target antigens and controls as shown in the online Data Supplement. In this analysis, the anti-K58-ß2M antibody reacted only with K58-ß2M, i.e., the peaks (f and s, representing different conformations of K58-ß2M, in Fig. 1 of the online Data Supplement) disappear because the K58-ß2M is bound to and precipitated with the protein G-Sepharose–immobilized antibody, whereas the peaks representing the marker peptide and wild-type ß2M are unaffected. In comparison, the anti-ß2M antibody used for IA-LC-MS reacts with both the wild-type and the K58 form of the antigen. The unrelated anti-amyloid ß-peptide antibody reacts with none of the proteins.
To be able to measure K58-ß2M specifically in biofluids in the presence of high amounts of irrelevant proteins, we developed a quantitative capture immunoassay. As capture antibody, we used the novel anti-K58-ß2M mAb developed in this study: We avoided using a polyclonal rabbit anti-ß2M antibody as capture reagent because the polyclonal antibody binds both variants of ß2M, and thus the binding of K58-ß2M may be quenched by high concentrations of wild-type ß2M, which is a typical occurrence in sera from kidney patients. The polyclonal antibody was instead used as the secondary antibody followed by a detection antibody. The various layers of the assay were optimized by use of purified K58-ß2M added to normal sera (Fig. 3
). Assay performance based on the addition of known amounts of purified K58-ß2M to normal serum is illustrated in Fig. 3
. The correlation coefficient (R2) for the calibration curve was 0.93, and the limit of detection was 30 µg/L.
|
Measurements in patient sera by this ELISA (Fig. 4
) confirmed significantly higher amounts of K58-ß2M in sera from the group of CHD patients compared with the control group (healthy donors and Alzheimer disease patients), particularly in the group of CHD patients dialyzed with cuprophane membranes. In those patients, the mean K58-ß2M concentration was 8.5 mg/L (P <0.0001 vs the control group), whereas the patients dialyzed with polysulfone had a mean concentration of 1.8 mg/L (P = 0.016). Both patient groups thus differed significantly from the control group (mean K58-ß2M concentration, 0.7 mg/L) and from each other (P <0.0001). Sera from controls and patients were sampled and stored identically, an important procedure because K58-ß2M concentrations appear to increase with prolonged storage at 4 °C and because freeze–thaw cycles appear to remove some K58-ß2M through precipitation (data not shown). The K58-ß2M concentrations were highest in patients treated with cuprophane membranes, but the polysulfone membranes were also associated with increased K58-ß2M (Fig. 4A
). The immunoassay data correlated roughly with the semiquantitative relative estimates of the K58-ß2M concentrations determined by the MS analyses (Fig. 2D
). It was not possible to show any significant alterations of K58-ß2M concentrations before and after dialysis (Fig. 4B
) irrespective of the type of membrane used, although the concentrations wild-type ß2M, as expected, clearly decreased after dialysis. Thus, there were no obvious correlations between total ß2M concentrations and K58-ß2M concentrations in these samples (Fig. 4A
). This finding appears to indicate that the amounts of K58-ß2M in the CHD sera are not a direct consequence of the high ß2M concentrations. In addition, patient age or time in dialysis had no direct correlation with K58-ß2M concentrations (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
mobility in crossed immunoelectrophoresis and was also based on the finding that intact ß2M is prone to cleavage at lysine-58, with subsequent removal of this residue, by the concerted activities of activated C1s and carboxypeptidase B(22). This process was shown to modify ß2M when purified ß2M was added to patient sera, and the modification could be inhibited by specific enzyme inhibitors(27).
The K58-ß2M variant may be generated in stored sera and in sera from patients with various diseases and may be lost on repeated freezing–thawing, but we demonstrated that in carefully sampled material, there is a significant increase of circulating K58-ß2M in 20%–40% of dialyzed patients with chronic renal disease. This finding indicates that many patients on CHD process ß2M into K58-ß2M much more readily than do healthy controls. Sampling in the presence of proteinase inhibitors will be necessary to asses whether ex vivo proteolytic activity contributes to the observed alterations. The antibody-reactive epitopes of other ß2M structural variants might theoretically be lost; therefore, their presence in sera cannot be ruled out completely because the multistep chromatography may preclude recovery of such species and because such variants would be invisible in the K58-ß2M–specific immunoassay.
The findings in this study appear to rule out the possibility that cleaved ß2M is a constant fraction of the total ß2M population in healthy individuals and therefore in absolute terms is increased in hemodialysis patients, who typically have increased serum ß2M concentrations. We found no direct correlation between total ß2M concentrations and K58-ß2M concentrations. Indeed, we found similar concentrations of K58-ß2M before and after dialysis despite clearly decreased concentrations of wild-type ß2M. The finding of increased K58-ß2M in sera from patients undergoing hemodialysis but not in healthy controls is interesting because we have shown that K58-ß2M at physiologic pH and temperature is more conformationally unstable, i.e., prone to unfolding, aggregation, and seeded amyloid formation, than is wild-type ß2M(28). The concentration of circulating K58-ß2M thus may be a measure of the biocompatibility of the dialysis process, i.e., the degree of complement activation by the dialysis procedure. However, the present studies do not demonstrate that circulating K58-ß2M is a precursor of ß2M amyloid or is involved in the initiation of ß2M amyloid formation. Such proof will require immunochemical demonstration of the presence of this variant in amyloid tissue and in vitro demonstration of direct amyloidogenicity of K58-ß2M under conditions resembling in vivo environment as well as prospective studies in dialysis patients. However, the availability of a specific K58-ß2M immunoreagent now makes these questions more easily addressable.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
K58-ß2M, cleaved ß2-microglobulin lacking lysine-58; mAb, monoclonal antibody; and PBS, phosphate-buffered saline. 
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  S. Giorgetti, S. Raimondi, S. Cassinelli, M. Bucciantini, M. Stefani, G. Gregorini, G. Albonico, R. Moratti, G. Montagna, M. Stoppini, et al. {beta}2-Microglobulin is potentially neurotoxic, but the blood brain barrier is likely to protect the brain from its toxicity Nephrol. Dial. Transplant., April 1, 2009; 24(4): 1176 - 1181. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Song, M. Patel, C. N. Rosenzweig, Y. Chan-Li, L. J. Sokoll, E. T. Fung, N.-H. Choi-Miura, M. Goggins, D. W. Chan, and Z. Zhang Quantification of Fragments of Human Serum Inter-{alpha}-Trypsin Inhibitor Heavy Chain 4 by a Surface-Enhanced Laser Desorption/Ionization-Based Immunoassay Clin. Chem., June 1, 2006; 52(6): 1045 - 1053. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. F. Winchester Novel Changes in {beta}2-Microglobulin in Dialysis Patients Clin. Chem., July 1, 2005; 51(7): 1089 - 1090. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and after (•) hemodialysis (HD) using polysulfone membranes.