| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Case Report |
Light Chain Dimers in Serum of a Patient with Multiple Myeloma
1 Division of Clinical Biochemistry and Immunology,
2 Department of Biochemistry and Molecular Biology,
3 Department of Molecular Pharmacology and Experimental Therapeutics,
4 Mayo Proteomics Research Center, and
5 Division of Hematology, Department of Internal Medicine, Mayo Clinic, Scottsdale, AZ 85259.
6 Division of Hematology, Department of Internal Medicine, Mayo Clinic, Rochester, MN 55905.
aAddress correspondence to this author at: Division of Clinical Biochemistry and Immunology, 920 Hilton, Mayo Clinic, 200 1st St. SW, Rochester, MN 55905. Fax 507-266-4088; e-mail kyle.robert{at}mayo.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionCase ReportMaterials and MethodsResultsDiscussionReferences |
|---|
or 
type in their urine. Usually, these light chains exist as monomeric or dimeric forms, but rarely, larger molecules, such as tetramers, have been reported in the serum.
Methods and Results: We report the presence of trimeric complexes of light chain dimers in a patient who was diagnosed with a free light chain multiple myeloma 2 years earlier and subsequently underwent a stem cell transplant. Recently, the patient presented with a large serum M-spike (23 g/L) by protein electrophoresis. The spike consisted of monoclonal light chains without a heavy chain. The urine contained only 8 mg of light chain in a 24-h specimen. Quantitative analysis of the serum and urinary free light chains (FLCs) indicated the probability of larger aggregates of FLCs. Size-exclusion chromatography, electrophoresis, analytical ultracentrifugation, and mass spectrometric studies of the serum revealed almost exclusively the presence of trimolecular aggregates of light chain dimers without other multimeric species.
Conclusion: Monoclonal light chains may present as hexameric aggregates that cannot be cleared by renal excretion.
|
Introduction |
|---|
|
TopAbstractIntroductionCase ReportMaterials and MethodsResultsDiscussionReferences |
|---|
light chains is associated with an increase in the conversion of monomers to dimers and tetramers (7). One of the first reports of BJP tetramers was from a patient with MM who had a light chain with a molecular mass of 84 kDa, composed of two noncovalently linked dimers (8). Since then, there have been other reports of multimeric light chains from patients with MM. In addition, the urine of healthy individuals contains 10–20% tetrameric polyclonal light chains (9).
We report here the identification of multimeric light chain complexes presenting as a large serum M-spike composed entirely of monoclonal FLCs without heavy chain in a patient with MM. This is the first study, to the best of our knowledge, to report trimeric aggregates of light chain dimers. The vast majority of patients with BJ proteinemia have either no serum M-spike or a very modest one.
|
Case Report |
|---|
|
TopAbstractIntroductionCase ReportMaterials and MethodsResultsDiscussionReferences |
|---|
light chain; however, IFE of a 24-h urine sample was negative for a monoclonal protein. The patient underwent a stem cell harvest and on August 30, 2000, received melphalan, 200 mg/m2, followed by a stem cell infusion on September 1, 2000. He engrafted on day 9 after transplant. IFE of the serum showed no monoclonal protein. The patient had recurrent symptoms 12 months after the transplant and was seen in October 2001 with increased bone pain. The hemoglobin was 104 g/L. A bone marrow aspirate revealed 90% plasma cells, and a metastatic bone survey revealed osteopenia with compression fractures of several mid-thoracic and lumbar vertebrae. Serum electrophoresis 1 year after transplant showed a 
component of 25.4 g/L (M-spike of 22 g/L). A 24-h urine specimen contained 327 mg of protein with an M-spike of 28.5 mg/24 h. IFE of the urine showed a small protein, whereas IFE of the serum revealed a monoclonal light chain and no reactivity with IgG, -A, -M, -D, and -E. The patient was treated with doxorubicin (adriamycin), vincristine, and dexamethasone. Because a bone marrow harvest was unsuccessful, he was treated with thalidomide and dexamethasone.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionCase ReportMaterials and MethodsResultsDiscussionReferences |
|---|
size-exclusion chromatography and gel electrophoretic analysis
We diluted 3 µL of patient serum sample to 20 µL with phosphate-buffered saline (10 mmol/L sodium phosphate, 9 g/L NaCl, pH 7.4) and loaded the sample on a Superdex 200 column (10 x 300 mm; Amersham Biosciences) for fractionation by size-exclusion chromatography (SEC) using the Biologic Workstation (Bio-Rad); 0.5-mL fractions were collected. We used gel filtration calibrators from Bio-Rad (thyroglobulin, 670 kDa; -globulin, 158 kDa; ovalbumin, 43 kDa; myoglobin, 17 kDa; and vitamin B12, 1430 Da) to calibrate the column to generate a calibration curve for molecular mass estimation.
Five-microliter aliquots of selected fractions from the SEC were used for protein separation by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) under reducing conditions with 12% Tris-glycine HCl gels (Criterion precast gel; Bio-Rad) and molecular mass markers (Bio-Rad). We prepared reduced samples by boiling the protein in sample buffer with ß-mercaptoethanol (2-ME) and nonreduced samples by boiling in the same buffer without 2-ME. Gels were stained with BioSafe colloidal blue stain (Bio-Rad). Protein chromatography, N-terminal sequencing, gel electrophoresis, and mass spectrometry were performed in the Mayo Clinic Proteomics Research Center.
analytical ultracentrifugation
Sedimentation equilibrium.
Sedimentation experiments on the light chain complex purified by SEC were performed using the Beckman Optima XL-1 analytical ultracentrifuge. We used BioSpin 6 chromatography columns (Bio-Rad) to equilibrate the proteins analyzed in 20 mmol/L HEPES, pH 7.8, 100 mmol/L KCl, with or without 2.0 mmol/L dithiothreitol (DTT). We analyzed 300-µL protein samples at 
500 mg/L in double-sector cells with sapphire windows against a reference composed of the BioSpin column equilibration buffer. To visualize the protein at the bottom of the cell, we added 20 µL of the immiscible fluorocarbon, FC-43, to each sample. Samples were analyzed at 30 000g (20 000 rpm) in an An60Ti rotor at 20 °C. Data were obtained using both the absorbance and interference optical systems at a wavelength of 280 nm as scans of A280 (protein concentration; c) vs radial distance (r). Equilibrium was achieved when scans taken 3 h apart after an initial 24–48 h of centrifugation were superimposable.
Molecular masses were determined from all equilibrium experiments by finding the best fit of the primary data, using the nonlinear, least-squares Levenberg–Marquardt fitting routine, which minimizes the 
2 residuals between the function and the data. For the proteins analyzed here, the monomer–polymer model was used and is represented below:
 | (1) |
is the angular velocity; 
is the partial specific volume of the protein; and 
is the density of the buffer. The number of monomers in the polymer assembly is n, and b is the baseline absorbance.
mass spectrometric analysis
All mass spectrometry analyses were performed on a Finnigan-MAT 900 mass spectrometer with a position and time-resolved ion counter detection system. The light chain fraction obtained by SEC was analyzed by infusion at 0.3 µL/min and a concentration of 1 µmol/L in the positive ion mode, using microelectrospray ionization mass spectrometry. The protein stored in phosphate-buffered saline was exchanged by gel filtration into 20 mmol/L ammonium acetate buffer, pH 6.8, using a Bio-Rad P-6 spin column. Sulfur hexafluoride (SF6) gas was used in all experiments to decrease coronal discharge and enhance the signal-to-noise ratio. Scans were collected at a resolution of 1200 over a mass range of 1000–7000 atomic mass units at 10 s/decade. Multiple scans were collected and collated, and the multiply charged spectra were transformed into the molecular mass scale by use of algorithms supplied by Finnigan-MAT. The buffered protein was denatured from the native state by the addition of 300 mL/L acetonitrile and 10 mL/L acetic acid (to reduce the pH to 4.8).
|
Results |
|---|
|
TopAbstractIntroductionCase ReportMaterials and MethodsResultsDiscussionReferences |
|---|
The data presented were obtained from a patient who was diagnosed with MM in May 2000. Laboratory analysis performed 5 months later revealed the presence of a very small free monoclonal protein in the serum and light chains in the urine. The patient was reevaluated in 2001 after a stem cell transplantation when an increase in the size of the monoclonal protein was noted. The most remarkable feature of the recent evaluation was the presence of an unusually large M-protein in the serum by protein electrophoresis (23 g/L; Fig. 1A
) with small amounts of BJP in the urine (8 mg/24 h; Fig. 1B
). In addition, the lack of a heavy chain by IFE led to additional quantitative analysis of serum FLCs by nephelometry [Refs. (10)(11); Abraham RS, Katzmann JA, Clark RJ, Kyle RA, Gertz MA, manuscript in preparation] with the FREELITE reagents, which revealed 0.102 g/L free light chain at a dilution of 1:20 (40 µL of sample plus 210 µL of diluent; Table 1
). To ensure that this result was an accurate estimation of the FLC amount and not a reflection of antigen excess, leading to an erroneous interpretation, additional dilutions of the patient serum sample were made manually (Table 1
). When we diluted the serum sample almost 8000-fold from the original 1:20 dilution, the free concentration was 344 g/L, indicating that the large amounts of FLC had produced a falsely low estimation of the monoclonal protein. This is the first instance, however, in our experience where the serum FLC concentration was significantly higher than the serum M-spike.
|
|
We evaluated a cohort of 38 MM patients for serum M-spike and FLC concentrations. The median serum M-spike concentration was 30.5 g/L (range, 8.4–65 g/L). The median serum FLC concentrations for and were 0.0073 g/L (range, 0.00007–0.456 g/L) and 0.013 g/L (range, 0.00002–5.5 g/L), respectively. In this group of patients with intact monoclonal immunoglobulin, the ratio of serum FLCs to serum monoclonal immunoglobulin was much smaller than the current patient, and there was no direct correlation of the serum M-spike value with serum FLC concentrations.
The absence of correspondingly large amounts of BJP in the urine suggested the possibility that light chains were forming multimers and retarding renal excretion. To investigate the assumption that the free light chains were forming multimers in serum, we performed SEC (Fig. 2A
). Using predetermined calibrators, we observed that the light chain had a mass of 140 kDa (fractions 17–21 eluting at 27–31 min, respectively), whereas albumin had a mass of 70 kDa (Fig. 2A
). The fractions obtained from the SEC were subjected to N-terminal sequencing to verify the identity of the immunoglobulin light chain and human albumin (data not shown). We also evaluated selected fractions from the 140-kDa peak by SDS-PAGE protein electrophoresis and mass spectrometry. The samples were electrophoresed under reducing conditions to visualize the monomeric units of light chain [Fig. 2B
(i)].
|
Five fractions obtained from the SEC were run on SDS-PAGE along with the appropriate molecular mass markers. Although all five fractions (fractions 17–21) showed a band at 25 kDa, the band was most dominant in the first three fractions [fractions 17–19; Fig. 2B
(i)]. Because the theoretical molecular mass for a monomeric subunit of light chain is 25 kDa, it was clearly evident that the SEC fractions eluting at 140 kDa were reduced into individual subunits with a molecular mass of 25 kDa, corresponding to monomers of light chain species. The molecular mass of the multimeric complex suggested aggregates that were larger than tetramers.
Because light chains preferentially exist as covalent dimers, protein electrophoresis was performed under nonreducing conditions to visualize the dimer [Fig. 2B
(ii)]. The light chain fraction (fraction 18) obtained by SEC was run under nonreducing and reducing conditions on the same gel. The nonreduced sample was prepared by boiling in sample buffer without 2-ME, whereas the reduced sample included 2-ME. We observed dimers with a molecular mass of 40 kDa under nonreducing conditions and monomers of 25 kDa in the reduced sample [Fig. 2B
(ii)].
The conformation of proteins influences migration patterns during gel electrophoresis. It is plausible, therefore, that because light chains exist as covalent dimers, making them more globular in structure, they would migrate faster under nonreducing conditions, giving a smaller apparent molecular mass (40 kDa) than expected (50 kDa) based on the electrophoretic mobility of reduced light chain monomers (25 kDa).
To substantiate the finding that the light chains were aggregating into large multimers, we performed analytical ultracentrifugation as described. The sedimentation equilibrium analysis of the serum sample, performed in the presence or absence of DTT, revealed a molecular mass of 128 kDa (Table 2
). The data shown are the average of four data sets. The sedimentation data for the light chain without DTT are shown in Fig. 3
. Absorbance readings at 280 nm were taken at different radial positions across the cell. The molecular mass of the protein at equilibrium was calculated by use of an exponential derivation of the Lamm equation. A nonlinear least-squares fit (Levenberg–Marquardt) was done to fit the data to a monomer–polymer model. The absorbance values ranged from 0 to 1.5, with residuals that were ±0.03 (Fig. 3
), suggesting minimal fluctuation of the data points.
|
|
Although electrophoresis and ultracentrifugation analysis provide molecular mass estimates, a more accurate assessment of molecular mass is possible by mass spectrometry methods, which also allow definitive classification of the multimeric form based on its constituent subunit structures. Mass spectrometric studies were done on the Finnigan-MAT mass spectrometer, using the SEC-purified light chain fraction (fraction 18). The sample was rebuffered with ammonium acetate to provide optimal conditions for ionization as well as to maintain the native structure of the protein without additional modifications. On deconvoluting the mass spectra, we observed that the molecular mass of the purified light chain from serum was 136 kDa (Fig. 4
), which closely matched the data obtained by gel electrophoresis (140 kDa) and ultracentrifugation (128 kDa). To determine the molecular mass of the subunit structure, the protein was denatured by acidic pH conditions and the organic solvent acetonitrile. Mass spectra data obtained on the denatured protein showed a molecular mass of 45 kDa (Fig. 4
, inset). Data obtained from running purified light chain calibrators on the mass spectrometer (Abraham RS, Bergen HR III, Bradwell AR, Timm M, Fonseca R, manuscript in preparation) also showed dimers having a mass of 45 kDa and monomer subunits with a mass of 21–23 kDa. The molecular mass of 136 kDa for the native protein, along with the molecular mass of 45 kDa for the dimer, suggested that FLC protein in this individual was forming homotrimeric aggregates of the dimer. Because the dimer is formed by covalent bonds, the denaturing methods do not yield the component monomers. However, the trimolecular complexes of dimers are held together by noncovalent interactions, which are disrupted by the acidic pH and acetonitrile.
|
|
Discussion |
|---|
|
TopAbstractIntroductionCase ReportMaterials and MethodsResultsDiscussionReferences |
|---|
20% of MM cases (10). The patient reported here had a large M-spike by serum protein electrophoresis; however, there was no evidence of a heavy chain, which led to the discovery that there was significant BJ proteinemia (Fig. 1A
). Approximately two-thirds of patients with a serum M-protein also have BJPs in the urine (14). In this patient the BJ proteinuria was negligible (Fig. 1B
). Although only one-fourth of MM patients have overt renal disease, more than one-half have more subtle renal pathology (2)(15)(16). Serum creatinine concentrations do not reflect these changes until the glomerular filtration rate is reduced by almost 50% (17). Therefore, use of creatinine values alone may produce an underestimation of the incidence of renal involvement in MM (17). Approximately 65% of patients with light chain myeloma have impaired renal function at the time of diagnosis if estimated creatinine clearance is used as the variable (15). In the patient studied here, renal function appeared to be normal based on serum creatinine and blood urea nitrogen concentrations (0.008 and 0.135 g/L, respectively).
The marked increase in the serum FLCS (Table 1
) compared with the serum M-peak was an unusual finding. Because the nephelometric FLC assay was calibrated for a preparation of monomeric light chains, we speculate that the FLC quantification was artifactually high. Therefore, the 344 g/L may be a pseudo-result, as the unusual hexameric clustering of light chains may have produced stronger interactions with the polyclonal FLC antisera when compared with the normal monomeric or dimeric calibrators, thereby producing an inaccurate estimate in the nephelometric quantification.
Approximately 24 cases have been reported with tetrameric monoclonal light chains (18). All of these patients, with rare exceptions, had MM or related disorders. Under normal circumstances, light chains in serum are in reversible association–disassociation equilibrium among monomers, dimers, and tetramers (1).
Previously, noncovalent multisubunit aggregate formation in proteins was deduced largely by gel electrophoresis methods (Fig. 2B
) or analytical ultracentrifugation (Table 2
and Fig. 3
). Although these methods do provide limited information on complex formation, they are entirely quantitative in terms of determining accurate molecular mass values. Among the advantages of microelectrospray ionization mass spectrometry is its ability to separate different molecular species, based on mass, from a complex mixture (19). It is also able to accurately determine molecular masses of large proteins (20). The absence of other molecular mass species, such as monomers, dimers, or tetramers, in the SEC fraction (Fig. 2A
) of the patient’s serum suggests that it was composed exclusively of the trimer-dimer complex (Fig. 4
).
It has been suggested that light chain myeloma patients tend to have a predilection for renal damage (19)(21), although there have been other reports to the contrary (3). It has been presumed that the higher association constant for light chains may be of clinical relevance in the development of renal disease (22). Some tetrameric light chains have been shown to be composed of four noncovalent monomers (23) that are otherwise typically found as stable monomers or noncovalent dimers. It has been clearly shown that BJPs preferentially exist as covalent dimers, in contrast to light chains, which are found as monomers or noncovalent dimers. The formation of covalent light chain dimers is highly dependent on the accessibility of free cysteinyl residues (24). On the other hand, noncovalent dimerization is related to the KD (association constant) of the individual monomeric subunits (25). Light chains with an unusually high KD may have a greater propensity to form tetramers or larger multimeric structures.
Although the presence of monoclonal light chain proteins has been associated with pathologic conditions such as MM or primary (AL) amyloidosis, the exact mechanism of formation or the pathophysiologic relevance of the multimer conformers has not been established. It is quite likely that they are metabolized more slowly than the average dimer or monomer, because of their high molecular mass (26). It is also reasonable to assume that the multimeric forms, by virtue of their larger size, are not as readily filtered through the glomeruli. This patient did not have overt signs of renal failure, although the disparity between the concentrations of BJPs in the urine and serum strongly suggests a reduction in renal excretion of the FLCs. The concordance in the molecular masses obtained by four independent techniques—SEC, gel electrophoresis, analytical ultracentrifugation, and microelectrospray ionization mass spectrometry—clearly indicates that the FLCs in this patient were unusual complex aggregates of independent dimer subunits. Because this appears to be the first time such molecular complexes has been reported in MM, the correlation of circulating, molecular aggregates of clonal protein to the pathophysiologic state has to be established.
In conclusion, this case allowed the identification of an unusual aggregation of light chain dimers into trimolecular complexes in a patient with MM who had a relapse 1 year after a stem cell transplant. The presence of these multimeric species prevented renal excretion of the light chains, producing a significantly large serum M-spike with a correspondingly small urinary M-spike.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionCase ReportMaterials and MethodsResultsDiscussionReferences |
|---|
and chains of human immunoglobulin. J Biol Chem 1969;244:4299-4307. and free immunoglobulin light chains: relative sensitivity for detection of monoclonal light chains. Clin Chem 2002;48:1437-1444.
-DNA transcription complexes: analysis by microelectrospray mass spectrometry. Rapid Commun Mass Spectrom 2001;15:1011-1016.[Web of Science][Medline]
[Order article via Infotrieve]
-globulin B chains. Nature 1965;205:1171-1173.
I light chains: role of the third hypervariable region. Proc Natl Acad Sci U S A 1980;77:1144-1148.The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  I. Herzum, H. Renz, and H. G. Wahl Immunochemical Quantification of Free Light Chains in Urine Clin. Chem., June 1, 2005; 51(6): 1033 - 1035. [Full Text] [PDF]
|
|
|
|
|
|
J. R. Tate, D. Gill, R. Cobcroft, and P. E. Hickman Practical Considerations for the Measurement of Free Light Chains in Serum Clin. Chem., August 1, 2003; 49(8): 1252 - 1257. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. S. Abraham, S. M. Geyer, T. L. Price-Troska, C. Allmer, R. A. Kyle, M. A. Gertz, and R. Fonseca Immunoglobulin light chain variable (V) region genes influence clinical presentation and outcome in light chain-associated amyloidosis (AL) Blood, May 15, 2003; 101(10): 3801 - 3807. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
