Phenotyping Polyclonal Kappa and Lambda Light Chain Molecular

In 2 patients with immune disorders and hypergammaglobulinemia, we observed a skewed polyclonal molecular mass distribution which translated into bias...
0 downloads 0 Views 4MB Size
Article pubs.acs.org/jpr

Phenotyping Polyclonal Kappa and Lambda Light Chain Molecular Mass Distributions in Patient Serum Using Mass Spectrometry David R. Barnidge,*,† Surendra Dasari,‡ Marina Ramirez-Alvarado,§ Adrian Fontan,† Maria A. V. Willrich,† Renee C. Tschumper,∥ Diane F. Jelinek,∥ Melissa R. Snyder,† Angela Dispenzieri,†,⊥ Jerry A. Katzmann,† and David L. Murray† †

Department of Laboratory Medicine and Pathology, ‡Biomedical Statistics and Informatics, §Department of Biochemistry and Molecular Biology, ∥Department of Immunology, and ⊥Department of Medicine, Mayo Clinic, Rochester, Minnesota 55905, United States S Supporting Information *

ABSTRACT: We previously described a microLC-ESI-Q-TOF MS method for identifying monoclonal immunoglobulins in serum and then tracking them over time using their accurate molecular mass. Here we demonstrate how the same methodology can be used to identify and characterize polyclonal immunoglobulins in serum. We establish that two molecular mass distributions observed by microLC-ESI-Q-TOF MS are from polyclonal kappa and lambda light chains using a combination of theoretical molecular masses from gene sequence data and the analysis of commercially available purified polyclonal IgG kappa and IgG lambda from normal human serum. A linear regression comparison of kappa/lambda ratios for 74 serum samples (25 hypergammaglobulinemia, 24 hypogammaglobulinemia, 25 normal) determined by microflowLC-ESI-Q-TOF MS and immunonephelometry had a slope of 1.37 and a correlation coefficient of 0.639. In addition to providing kappa/lambda ratios, the same microLC-ESI-Q-TOF MS analysis can determine the molecular mass for oligoclonal light chains observed above the polyclonal background in patient samples. In 2 patients with immune disorders and hypergammaglobulinemia, we observed a skewed polyclonal molecular mass distribution which translated into biased kappa/lambda ratios. Mass spectrometry provides a rapid and simple way to combine the polyclonal kappa/lambda light chain abundance ratios with the identification of dominant monoclonal as well as oligoclonal light chain immunoglobulins. We anticipate that this approach to evaluating immunoglobulin light chains will lead to improved understanding of immune deficiencies, autoimmune diseases, and antibody responses. KEYWORDS: Immunoglobulins, monoclonal, polyclonal, kappa/lambda ratios, autoimmune diseases



INTRODUCTION

normally distributed peaks which were confirmed to be kappa and lambda using top-down mass spectrometry and purified kappa and lambda IgG preparations. In addition, a theoretical molecular mass distribution for polyclonal kappa and lambda light chains was generated using full length gene sequence data. The population means found using the molecular mass distributions from the gene sequence data was in good agreement with the molecular mass distributions observed using the mass spectrometer. In addition to determining light chain molecular mass distributions, the methodology can be used to determine the kappa to lambda ratio using the peak area of each isotype distribution. We demonstrate this by comparing the kappa/lambda ratios found by microflowLCESI-Q-TOF MS and immunonephelometry for 25 patients with

Clinical laboratories evaluate serum immunoglobulins using a combination of protein gel electrophoresis (PEL), immunofixation (IFE), and immunonephelometry. In a healthy individual, a broad evenly dispersed electrophoretic migration pattern is observed for intact immunoglobulin molecules. The broad migration pattern is a result of the large number of immunoglobulins generated as a function of gene rearrangements and somatic hypermutation (approximately 2 × 1012 unique immunoglobulin combinations are possible).1 Recently we reported using microflow liquid chromatography positive ion electrospray ionization coupled to a quadrupole time-offlight mass spectrometer (microLC-ESI-Q-TOF MS), for identifying, quantifying, and isotyping monoclonal immunoglobulins in serum and urine.2 Here we report the use of this method to identify polyclonal molecular mass distributions specific to kappa and lambda light chain isotypes. Mass spectra from pooled normal serum enriched for immunoglobulins and reduced with dithiothreitol (DTT) showed two distinct, © 2014 American Chemical Society

Special Issue: Proteomics of Human Diseases: Pathogenesis, Diagnosis, Prognosis, and Treatment Received: June 13, 2014 Published: August 18, 2014 5198

dx.doi.org/10.1021/pr5005967 | J. Proteome Res. 2014, 13, 5198−5205

Journal of Proteome Research

Article

Figure 1. Theoretical V, J, and C region molecular mass distributions. Graphical representation of the theoretical molecular mass distributions from the variable (V), joining (J), and constant (C) regions of kappa and lambda light chains. The molecular masses from the V and J regions should have a normal distribution due to random gene rearrangement as predicted by the central limit theorem.

Serum

hypergammaglobulinemia, 24 patients with hypogammaglobulinemia, and 25 normal controls. This unique mass spectrometry based approach to describing an individual’s immunoglobulin light chain repertoire is rapid, easy to perform, and uses low-cost reagents. As an analytical technique it has the potential to redefine how the immune system is assessed for immune deficiencies, autoimmune diseases, and antibody/vaccine responses.



A volume of 20 μL of serum was enriched for immunoglobulins using 180 μL of Melon Gel following the manufacturer’s instructions. After immunoglobulin enrichment, 20 μL of sample was reduced by adding 20 μL of 100 mM DTT and 20 μL of 50 mM ammonium bicarbonate then incubated at 55 °C for 30 min before injection. Samples were placed into 96 deepwell PCR plates (300 μL volume) at 9 °C while waiting for injection. LC Conditions

MATERIALS AND METHODS

An Eksigent Ekspert 200 microLC (Dublin, CA) was used for separation; mobile phase A was water + 0.1% FA, and mobile phase B was 90% acetonitrile + 10% 2-propanol + 0.1% FA. A 2 μL injection was made onto a 1.0 × 75 mm Poroshell 300SBC3 column with 5 μm particle size flowing at 25 μL/min, while the column was heated at 60 °C. A 25 min gradient was started at 80%A/20%B, held for 1 min, ramped to 75%A/25%B over 1 min, ramped to 65%A/35%B over 10 min, ramped to 50%A/ 50%B over 4 min, ramped to 95%A/5%B over 2 min held for 5 min, ramped to 80%A/20% over 1 min, and equilibrated at 80% A/20% for 1 min.

Serum and Immunoglobulin Reagents

Waste samples were collected from the Clinical Immunology Laboratory protein electrophoresis assay. Purified IgG kappa and IgG lambda from normal donors was purchased from Bethyl Laboratories (Montgomery, TX). Total kappa and total lambda light chains were quantitated with Siemens immunoassay reagent sets on the BNII immunonephelometer (Siemens Diagnostics, Marburg, Germany). Reagents

ESI-Q-TOF MS

Ammonium bicarbonate, dithiothreitol (DTT), and formic acid were purchased from Sigma-Aldrich (St. Louis, MO). Melon Gel was purchased from Thermo-Fisher Scientific (Waltham MA). Water, acetonitrile, and 2-propanol were purchased from Honeywell Burdick and Jackson (Muskegon, MI).

Spectra were collected on an ABSciex TripleTOF 5600 quadrupole time-of-flight mass spectrometer (ABSciex, Vaughan ON, CA) in ESI positive mode with a Turbo V dual ion source with an automated calibrant delivery system. Source conditions were IS, 5500; Temp, 500; CUR, 45; GS1, 35; GS2, 5199

dx.doi.org/10.1021/pr5005967 | J. Proteome Res. 2014, 13, 5198−5205

Journal of Proteome Research

Article

Figure 2. Deconvoluted mass spectrum of normal pooled serum. Deconvoluted form of the mass spectrum shown in Supporting Information Figure S3. The Bayesian reconstruction algorithm converts the multiply charged ions from m/z to molecular mass. The spectrum shows the same peak pair with normally distributed molecular masses in the ranges observed for kappa and lambda gene sequences converted to molecular mass shown in the inset.

30; CE, 50 ± 5. TOF MS scans were acquired from m/z 600− 2500 with an acquisition time of 100 ms. Fragment ion scans were acquired from m/z 350−2000 with an acquisition time of 100 ms. The instrument was calibrated every 5 injections through the CDS using calibration solution supplied by the manufacturer.

converted to average molecular mass using the ExPASy Compute pI/Mw tool (http://web.expasy.org/compute_pi) and then added to the molecular mass of the corresponding isotype constant region. Each molecular mass was placed into 100 Da width bins, and the software package JMP 10.0.0 was used to produce histograms, calculate the mean molecular mass, and model the normal distribution of calculated molecular masses.

MS Data Analysis

Analyst TF v1.6 was used for instrument control. Data were viewed using Analyst TF v1.6 and PeakView v1.2.0.3. Multiply charged ion peak centroids were used to calculate average molecular mass and the peak area value used for quantification through BioAnalyst software provided with Analyst TF. Multiple ion deconvolution was performed using the Bayesian Protein Reconstruct software package in BioAnalyst. The following settings were used: start mass (Da) = 22,000; Stop mass (Da) = 26,000 Da; step mass (Da) = 1; S/N threshold = 20; minimum intensity% = 0; iterations = 20; adduct, hydrogen. A limited mass range was used with a start m/z = 1100 and a stop m/z = 2500. Deconvoluted and multiply charged ions were manually integrated using the Manual Integration 33 script in Analyst TF.



RESULTS Since the light chain V and J region gene sequences are randomly rearranged, the central limit theorem4 predicts that the amino acid sequences of the expressed light chains should have normally distributed molecular masses. Figure 1 presents an example of the expected theoretical molecular mass distributions that would be observed for each of the three regions; variable (V), joining (J), and constant (C) for both the kappa and lambda light chains. The images under the V and J regions show the expected normally distributed molecular masses of the translated regions, while the images under the C regions show single bars. Since the kappa constant region has only one conserved amino acid sequence it is represented by a single molecular mass bar, while the image under the C region for lambda shows four different bars, each representing the four different lambda constant region molecular masses L1, L2, L3, and L7.5 A table with the C region amino acid sequences for kappa and lambda light chains and the calculated mass difference between the kappa and lambda C regions can be found in Supporting Information Figure S1. We constructed a histogram of the molecular masses obtained from the gene sequences for 785 kappa and 1087 lambda sequences each containing the entire V, J, and C regions. The nucleotide sequence information for each light chain was converted to the amino acid sequence and then converted to molecular mass. Histograms were constructed using the calculated molecular masses displayed in 100 Da bin widths for kappa and lambda

Bioinformatics Data Analysis

The normal distribution used to model the kappa and lambda light chain molecular mass distribution was generated using kappa and lambda gene sequences from the Boston University ALBase, supported by HL68705 (http://albase.bumc.bu.edu/ aldb). Gene sequences were uploaded into the IMGT alignment tool V-QUEST (http://www.imgt.org/IMGT_ vquest/vquest)3, and each sequence was aligned from the variable (V) region Frame 1 (N-terminus) through the joining (J) region to the beginning of the constant (C) region. Only gene sequences that included the entire V region through the J region were used (785 kappa and 1087 lambda). The gene sequence was then translated into the corresponding amino acid sequence using the ExPASy Translate tool (http://web. expasy.org/translate). This amino acid sequence was then 5200

dx.doi.org/10.1021/pr5005967 | J. Proteome Res. 2014, 13, 5198−5205

Journal of Proteome Research

Article

Figure 3. Deconvoluted mass spectra of IgG kappa and IgG lambda purified from serum. Deconvoluted mass spectra comparing normal pooled serum (top), IgG kappa purified serum (middle), and IgG lambda purified serum (bottom). The spectra from the purified sera show the absence of polyclonal light chains from the isotype not purified verifying that the molecular mass distributions observed by microLC-ESI-Q-TOF MS represent kappa and lambda polyclonal light chains.

visible in the mass spectrum in Supporting Information Figure S3 as seen by the series of multiply charged ions and the two sets of broad normally distributed peaks that are labeled with arrows. The deconvoluted form of the mass spectrum is displayed in Figure 2 and clearly shows two sets of peaks each with a normal distribution. The mean molecular mass for the smaller peak is 22,809.92 Da, and the mean molecular mass for the larger peak is 23,433.45 Da. This translates into a difference of 623.53 Da, which is 95.36 Da greater in mass than the calculated difference between kappa and lambda light chains using the gene sequence data. The inset to Figure 2 shows the calculated normal distributions fitted to the kappa and lambda histograms found using the gene sequence data illustrating the similarities between the data collected using the mass spectrometer and the models made using the gene sequence data. To experimentally confirm that the two molecular mass distributions were representative of the kappa and lambda light chain isotypes, commercially available purified IgG kappa and purified IgG lambda obtained from pooled normal serum were analyzed using the microLC-ESI-Q-TOF MS method. Figure 3 shows the results from deconvoluted mass spectra for normal pooled serum (top), IgG kappa purified normal pooled serum (middle), and IgG lambda purified normal pooled serum (bottom). The figure shows the absence of the lambda molecular mass distribution in the IgG kappa preparation and the absence of the kappa polyclonal molecular mass distribution

light chain isotypes (see Supporting Information Figure S1). A normal distribution was fitted to each of the histograms, and the mean molecular mass for kappa (μK) was found to be 23,373.41 Da, while the mean molecular mass for lambda (μL) was found to be 22,845.24 Da (indicated by red vertical lines). This translates into a difference of 528.17 Da between μK and μL. This value is 164.62 Da greater than the difference in molecular mass of the kappa and lambda constant regions alone. Assuming that the V and J region amino acid sequences are truly random, then the expected difference between μK and μL should be the difference in the masses of the constant regions. We theorize that the additional mass difference of approximately 164.62 Da is due primarily to the difference in the molecular mass of the amino acids in the N-terminal framework region (FR1) portion of the kappa and lambda variable regions. These regions contain amino acids that are largely conserved between the two isotypes and are not hypermutated to the extent of the CDR regions. The calculated molecular masses found using kappa and lambda gene sequence data were compared to the molecular masses observed in normal pooled serum analyzed using microLC-ESI-Q-TOF MS. The total ion chromatogram observed for the analysis of normal pooled serum and the mass spectra observed after summing the spectra over the elution time for the light chains (4.0 to 7.5 min) is shown in Supporting Information Figures S2 and S3. The multiply charged ions created by electrospray ionization6 are clearly 5201

dx.doi.org/10.1021/pr5005967 | J. Proteome Res. 2014, 13, 5198−5205

Journal of Proteome Research

Article

Figure 4. Kappa and lambda molecular mass distributions. Mass spectra comparing the kappa and lambda polyclonal molecular mass distributions for the +11 charge state obtained using microLC-ESI-Q-TOF MS. The trace in black was observed in serum from a patient with hypergammaglobinemia, while the trace in red is from pooled normal control serum.

in the IgG lambda preparation. We confirmed the purity of the IgG kappa and IgG lambda preparations using top-down MS as described in our previous work.2 In addition to providing the polyclonal kappa and lambda light chain molecular mass distributions, microLC-ESI-Q-TOF MS can also provide the relative abundance of each isotype. The total kappa/lambda light chain ratio in normal serum (TLC κ/λ) is often quoted as being between 1.5 and 2. The ratio remains consistent in normal individuals whether it is found using ELISA, immunohistochemical, or immunonephelometric methods.7−10 Since the electrospray ionization process creates multiply charged ions, there are three different approaches to determine the total kappa/lambda ratio from the mass spectrometry data. The first is to deconvolute the mass spectrum using software, as seen in Figure 3, and then calculate the peak areas of the kappa and the lambda light chains using the deconvolved data. Using the deconvolution approach we found that the total kappa/lambda ratio was consistently 1.5 times higher than the accepted ratio in normal serum samples that had a normal ratio as determined by immunonephelometry. We speculate this is due to the fact the Bayesian protein reconstruction algorithm was originally designed to provide molecular mass and abundance data for proteins with a single molecular mass as opposed to the heterogeneous unresolved mass spectra observed for the

polyclonal kappa and lambda light chains. The second approach to determine the total kappa/lambda ratio from the mass spectrometry data is to calculate the kappa and lambda peak areas for each charge state separately and then combine them. Using this technique we calculated kappa and lambda peak areas and ratios for charge state pairs +20 through +10 and combined them. The peak area ratios ranged from 1.22 (charge state +19) to 2.61 (charge state +12) with a mean value of 1.63. The third approach is to calculate the kappa/lambda ratio using the peak areas from a single charge state. We chose the +11 charge state pair since it consistently showed a total kappa/ lambda ratio using pooled normal serum that was in close agreement with those found using immunonephelometry. The precision of the +11 charge state approach was checked using 10 normal pooled serum replicates, and a %CV of 1.9 was observed with a range of 1.6 to 1.8 and a mean of 1.7. In addition, using the +11 charge state rather than the deconvoluted spectrum or the multiple charge states has the added benefit of simplifying the determination of the final kappa/lambda ratio by eliminating several computational steps. A linear regression analysis was performed to compare the total kappa/lambda ratios found using the +11 charge state vs those found using immunonephelometry for serum samples that had PEL gamma fractions that indicated hypergammaglobulinemia (n = 25), hypogammaglobulinemia (n = 24), or were 5202

dx.doi.org/10.1021/pr5005967 | J. Proteome Res. 2014, 13, 5198−5205

Journal of Proteome Research

Article

Figure 5. Immune system-dependent changes in molecular mass distributions. Mass spectra showing the changes in the total kappa/lambda light chain ratios in serum for the +11 charge state ions as a function of immune system status. The spectra are (top) normal control, (middle) patient with chronic inflammation, and (bottom) patient with Sjögren’s syndrome. Each spectrum is labeled with the calculated total kappa/lambda light chain ratio (TLC κ/λ).

a patient with a monoclonal gammopathy. The spectrum on the top is from a serum sample prior to the patient receiving a peripheral blood stem cell transplant, and the monoclonal kappa light chain is clearly visible. The spectrum on the bottom of the figure is from a serum sample 183 days after the transplant, and the monoclonal kappa light chain peak is no longer visible above the polyclonal background. In addition, the spectrum shows a more normal polyclonal kappa and lambda light molecular mass distribution with a kappa/lambda ratio of 1.78. Figures 5 and 6 illustrate the unique ability of microLCESI-Q-TOF MS to give a visual expression of oligoclonal kappa and lambda light chains while at the same time providing the total kappa/lambda light chain ratios. Only mass spectrometry can provide the level of resolution and mass measurement accuracy to enable such a distinctive view of an individual’s immunoglobulin light chain phenotype.

normal (n = 25). A plot of these data can be found in Supporting Information Figure S4. The results from this regression analysis showed that the kappa/lambda ratio values had a slope of 1.37 and a correlation of 0.640. An example of two mass spectra used for the regression analysis are shown overlaid in Figure 4 where the +11 charge state kappa and lambda light chain ions from a patient with hypergammaglobulinemia is shown in the black trace and a normal control is shown in the red trace. The spectrum from the patient with hypergammaglobulinemia clearly shows elevated light chain concentrations as compared to the normal control along with monoclonal light chains that protrude above the polyclonal background resulting in an oligoclonal appearance to the spectrum. Several of the patients with hypergammaglobulinemia showed a pronounced polyclonal skewing of the +11 kappa/lambda light chain charge states. Figure 5 shows a close up view of the +11 charge state ions observed from a normal control (TLC κ/λ = 1.6), a patient with a chronic inflammatory response of unknown origin (TLC κ/λ = 0.62), and an autoimmune patient with Sjögren’s syndrome (TLC κ/λ = 5.4). The spectra from the patients with chronic inflammation and Sjö gren’s syndrome show skewed polyclonal light chain molecular mass distributions. Figure 6 shows two up-close spectra of the +11 charge state observed in serum samples from



DISCUSSION Our previous work focused on using microLC-ESI-Q-TOF MS to identify and monitor monoclonal immunoglobulins in serum from patients with a monoclonal gammopathy. We have focused on monitoring kappa and lambda light chains since they do not undergo post-translational modifications compared to the heavy chains and have fewer amino acids making it easier 5203

dx.doi.org/10.1021/pr5005967 | J. Proteome Res. 2014, 13, 5198−5205

Journal of Proteome Research

Article

Figure 6. Kappa and lambda molecular mass distributions before and after PBSCT. Mass spectra showing the changes in the total kappa/lambda light chain ratios in serum for the +11 charge state ions in a patient with multiple myeloma before and after a peripheral blood stem cell transplant. The samples were taken 183 days apart.

calculation of kappa/lambda peak area ratios. The correlation between ratios found using mass spectrometry vs standard immunonephelometry was greater than the standard tolerance (±20%). However, many of these samples were from patients with a history of monoclonal gammopathy or who currently had a polyclonal protein electrophoresis pattern. Mass spectra from patients with the greatest discrepancy between ratios had monoclonal peaks, which biased the overall peak area calculation reducing the correlation between mass spectroscopy and nephelometry kappa/lambda ratios. Nevertheless, the observations for the patients with polyclonal hypergammaglobulinemia highlight the ability of a single mass spectrometry based assay to be used for determining the kappa/lambda light chain ratio while at the same time allowing the visualization of abnormalities such as an oligoclonal or restricted polyclonal immunoglobulin synthesis. We have expanded the application of our previous work by applying it to the assessment of polyclonal kappa and lambda light chain repertoires in serum. Alternative methods based on next generation sequencing (NGS) of DNA isolated from bone marrow B cells typically take a day to prepare the sample for sequencing, a day to perform the sequencing, and a day for bioinformatics interpretation. Here we demonstrate a technique that can produce results in under an hour from a routine blood draw. This initial work has focused on modeling the kappa and lambda light chain molecular masses in pooled normal control serum in order to understand the use of this technology. Future efforts will concentrate on the distribution of molecular masses from diverse normal control individuals in an effort to better understand the range of kappa to lambda ratios and the distributions of normal polyclonal light chains compared to the

to determine their accurate molecular mass. Initial experiments with normal pooled serum showed paired sets of unresolved peaks that were normally distributed and observed at lower levels than monoclonal immunoglobulins observed in patients with dysproteinemia. Using a combination of molecular mass data determined by gene sequence and MS, along with purified kappa and lambda controls, we determined that the normally distributed masses represent the entire kappa and lambda light chain immunoglobulin repertoire. The difference in mass between kappa and lambda light chains is due to the difference in the amino acids making up the constant regions along with contributions from the FR region within the V domain. The shape of each distribution is a function of the central limit theorem and the random nature of somatic hyper mutation. Kappa and lambda light chains have been well characterized using electrophoretic methods such as capillary electrophoresis1112 and SDS-PAGE,13,14 but these approaches do not have the resolving power, sensitivity, and speed compared to microLC-ESI-Q-TOF MS. Initial characterization of polyclonal immunoglobulin light chains was done over a decade ago; however, the kappa and lambda polyclonal molecular mass distributions were unresolved due to the low resolution of the instrumentation used for the analysis.15 Currently, much of the characterization of immunoglobulins using high-resolution mass spectrometers has focused on the quality control of therapeutic monoclonal immunoglobulins due to the superior sensitivity and specificity of the technique as compared to other methods16.17 The benefit of using a high resolution, accurate molecular mass instrument is seen in the spectra presented here where the kappa and lambda polyclonal light chain repertoire is clearly displayed. In addition, microLCESI-Q-TOF MS can provide a simple approach to the 5204

dx.doi.org/10.1021/pr5005967 | J. Proteome Res. 2014, 13, 5198−5205

Journal of Proteome Research

Article

characterization of plasma monoclonal free light chains in amyloidosis and multiple myeloma: a pilot study of intact and truncated forms of light chains and their charge properties. Clin. Chem. Lab. Med. 2008, 46, 335−341. (15) Adamczyk, M.; Gebler, J. C.; Wu, J. Profiling of polyclonal antibody light chains by liquid chromatography/electrospray ionization mass spectrometry. Rapid Commun. Mass Spectrom. 2000, 14, 49− 51. (16) Murray, D.; Barnidge, D. Characterization of immunoglobulin by mass spectrometry with applications for the clinical laboratory. Crit Rev. Clin. Lab. Sci. 2013, 50, 91−102. (17) Zhang, H.; Cui, W.; Gross, M. L. Mass spectrometry for the biophysical characterization of therapeutic monoclonal antibodies. FEBS Lett. 2014, 588, 308−317.

light chain distributions in autoimmune diseases, immunodeficiencies, and postvaccination sera.



ASSOCIATED CONTENT

S Supporting Information *

Molecular mass distributions calculated from light chain gene sequences; microLC-ESI-Q-TOF MS total ion chromatogram; mass spectrum of normal pooled serum; and linear regression comparison of kappa/lambda ratios. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*(D.R.B.) E-mail: [email protected]. Notes

The authors declare no competing financial interest.



REFERENCES

(1) Lefranc, M. P. IMGT, the international ImMunoGeneTics information system. Cold Spring Harbor Protoc. 2011, 2011, 595−603. (2) Barnidge, D. R.; Dasari, S.; Botz, C. M.; Murray, D. H.; Snyder, M. R.; Katzmann, J. A.; Dispenzieri, A.; Murray, D. L. Using mass spectrometry to monitor monoclonal immunoglobulins in patients with a monoclonal gammopathy. J. Proteome Res. 2014, 13, 1419− 1427. (3) Brochet, X.; Lefranc, M. P.; Giudicelli, V. IMGT/V-QUEST: the highly customized and integrated system for IG and TR standardized V-J and V-D-J sequence analysis. Nucleic Acids Res. 2008, 36, W503− 508. (4) Mukhopadhyay, N.; Chattopadhyay, B. A tribute to Frank Anscombe and random central limit theorem from 1952. Sequential Anal. 2012, 31, 265−277. (5) McBride, O. W.; Hieter, P. A.; Hollis, G. F.; Swan, D.; Otey, M. C.; Leder, P. Chromosomal location of human kappa and lambda immunoglobulin light chain constant region genes. J. Exp. Med. 1982, 155, 1480−1490. (6) Mann, M.; Meng, C. K.; Fenn, J. B. Interpreting mass-spectra of multiply charged ions. Anal. Chem. 1989, 61, 1702−1708. (7) Chui, S. H.; Lam, C. W.; Lai, K. N. Light-chain ratios of immunoglobulins G, A, and M determined by enzyme immunoassay. Clin. Chem. 1990, 36, 501−502. (8) Feiner, H. D. Pathology of dysproteinemia: Light chain amyloidosis, non-amyloid immunoglobulin deposition disease, cryoglobulinemia syndromes, and macroglobulinemia of Waldenstrom. Hum. Pathol. 1988, 19, 1255−1272. (9) Jefferis, R.; Deverill, I.; Ling, N. R.; Reeves, W. G. Quantitation of human Total IgG, kappa IgG and lambda IgG in serum using monoclonal antibodies. J. Immunol. Methods 1980, 39, 355−362. (10) Haraldsson, A.; Kock-Jansen, M. J.; Jaminon, M.; van Eck-Arts, P. B.; de Boo, T.; Weemaes, C. M.; Bakkeren, J. A. Determination of kappa and lambda light chains in serum immunoglobulins G, A and M. Ann. Clin. Biochem. 1991, 28 (Pt 5), 461−466. (11) Petersen, J. R.; Okorodudu, A. O.; Mohammad, A.; Payne, D. A. Capillary electrophoresis and its application in the clinical laboratory. Clin. Chim. Acta 2003, 330, 1−30. (12) Chartier, C.; Boularan, A. M.; Dupuy, A. M.; Badiou, S.; Bargnoux, A. S.; Cognot, C.; Eliaou, J. F.; Cristol, J. P. Evaluation of two automated capillary electrophoresis systems for human serum protein analysis. Clin. Biochem. 2011, 44, 1473−1479. (13) Lavatelli, F.; Brambilla, F.; Valentini, V.; Rognoni, P.; Casarini, S.; Di Silvestre, D.; Perfetti, V.; Palladini, G.; Sarais, G.; Mauri, P.; Merlini, G. A novel approach for the purification and proteomic analysis of pathogenic immunoglobulin free light chains from serum. Biochim. Biophys. Acta 2011, 1814, 409−419. (14) Kaplan, B.; Ramirez-Alvarado, M.; Dispenzieri, A.; Zeldenrust, S. R.; Leung, N.; Livneh, A.; Gallo, G. Isolation and biochemical 5205

dx.doi.org/10.1021/pr5005967 | J. Proteome Res. 2014, 13, 5198−5205