Study of Gammopathies with Immunofixation Electrophoresis TSIEH SUN, M.D., YORK Y. LIEN, PH.D., AND THOMAS DEGNAN, M.D.
Sun, Tsieh, Lien, York Y., and Degnan, Thomas: Study of gammopathies with immunofixation electrophoresis. Am J Clin Pathol 72: 5-11, 1979. Immunoflxation electrophoresis and immunoelectrophoresis were compared in 60 samples (51 sera and 9 urines) containing apparently homogenous bands, detected by electrophoresis. To determine the correlation with clinicalfindings,the patients were divided into three groups: (1) symptomatic with monoclonal immunoelectrophoretic patterns; (2) asymptomatic with monoclonal immunoelectrophoretic patterns; (3) asymptomatic with polyclonal immunoelectrophoretic patterns. The results from immunoelectrophoresis and immunofixation electrophoresis were consistent with each other in all cases of Groups I and III in terms of clonality (i.e., whether monoclonal or polyclonal) and immunoglobulin class; whereas in Group II, which was composed of asymptomatic patients, two sera and three urines were identified to have monoclonal changes by immunoelectrophoresis but polyclonal changes by immunofixation electrophoresis. It is recommended that immunoelectrophoresis still be used for routine clinical service, but supplemented by immunofixation electrophoresis in equivocal cases, namely: (1) light-chain changes masked by an umbrella effect of immunoglobulin G (IgG); (2) abnormal bands located in atypical areas or overlapped with a normal serum protein; (3) a normal-looking "free" light chain present in the urine, mimicking Bence Jones protein; (4) controversial cases of biclonal gammopathy; (5) mini-monoclonal or oligoclonal protein bands; (6) immune complexes. (Key words: Immunofixation electrophoresis; Gammopathies; Immunoelectrophoresis.)
Departments of Pathology and Medicine, Cornell University Medical College, New York, New York, and Departments of Laboratories and Medicine, North Shore University Hospital, Manhasset, New York
Materials and Methods Patients'
Samples
Patients with apparently homogenous bands in serum or urine (60 cases), detected by electrophoresis, were divided into three groups according to clinical history and further immunochemical evaluation: Group I. Patients were symptomatic with a monoclonal IEP pattern. Clinical diagnoses were plasma-cell myeloma or primary macroglobulinemia. These included 35 sera and five urine specimens. Immunochemically, the 35 sera contained monoclonal proteins consisting of 12 IgG-kappa, four IgG-lambda, four IgA-kappa, three IgA-lambda, four IgM-kappa, one IgM-lambda, two IgD-lambda, two biclonal gammopathies (IgGkappa and IgM-lambda; IgG-kappa and IgA-kappa), one kappa Bence Jones protein, and two lambda Bence Jones protein. The five urine specimens consisted of two kappa Bence Jones protein and three lambda Bence Jones protein. Group II. Asymptomatic with a monoclonal IEP pattern, this group contained ten sera and three urines. The clinical diagnoses were five benign monoclonal gammopathy or early myeloma (or macroglobulinemia), two benign prostatic hypertrophy, one each septicemia, acute glomerulonephritis, lymphoma, urinary tract infection, and syncope of unknown etiology, and one
IMMUNOFIXATION ELECTROPHORESIS is similar to immunoelectrophoresis (IEP) in principle, but instead of allowing antigen and antibody to diffuse in the supporting medium, antisera are applied, following electrophoresis, to the surface of the medium. The formation of an immunoprecipitin with a monospecific antiserum identifies the protein of interest. This basic concept was first described by Afonso in 1964,' but it was not a practical technic until Alper and Johnson modified it in 1969.3 Since then immunofixation electrophoresis has been used for the study of protein polymorphism, 2 genetic typing of alpha,-antitrypsin, 917 and identification of serum protein fractions. 516 The applicaReceived March 27, 1978; received revised manuscript and accepted for publication June 6, 1978. This work was done at North Shore University Hospital and was partly presented at the 29th National Meeting of the American Association for Clinical Chemistry, Chicago, Illinois, July 20, 1977. Address reprint requests to Dr. Sun: Department of Laboratories, North Shore University Hospital, Manhasset, New York 11030.
0002-9173/79/0700/0005 $00.85 © American Society of Clinical Pathologists
5
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
tion of immunofixation electrophoresis for the study of gammopathies was reported recently by Cawley and associates 4 and by Ritchie and Smith. 18 Included in these two reports were a few cases that illustrated the usefulness of immunofixation electrophoresis. However, the exact number of patients investigated was not specified. The present paper reports a systematic study of different types of gammopathy by immunofixation electrophoresis and compares it with IEP. A modified technic of immunofixation electrophoresis is also described. 21
6
SUN, LIEN, AND DEGNAN
A.J.C.P. • July 1979
Antisera tx»sv
The antisera used, including polyvalent, trivalent (IgG, IgA, IgM), monospecific for IgG (y) IgA (a), IgM (/x), IgD (8), IgE (e), kappa, and lambda, were from Kallestad (Chaska, Minnesota), Millipore (Bedford, Massachusetts), Behring (Somerville, New Jersey), Meloy (Springfield, Virginia) and Hyland (Costa Mesa, California). When equivocal results were obtained, antisera from different commercial sources were used for confirmation. Serum and Urinary
• ' (k
p K
c
A high-resolution electrophoresis system manufactured by Millipore Biomedica was used as described previously. 20 A battery of six electrophoretograms was obtained for each sample. Fifteen microliters of undiluted specimen were applied to the first strip of a buffered-agarose-coated plastic film as a control. Equal amounts of saline-diluted specimen (1:5) were applied to the other five strips. In the case of IgD myeloma, or other myelomas with very high concentrations of immunoglobulins, further dilutions were made to a point that antibody was in excess or in equivalence with the specific antigen. The antigen:antibody ratio was estimated by quantitation of individual immunoglobulins with the radial immunodiffusion technic (Kallestad) or by Ouchterlony double-diffusion analysis. Electrophoresis was run for 45 min at 200 v. lmmunofixation
FIG. 1. Biclonal gammopathy. Immunoelectrophoretic pattern shows abnormal mu, kappa, and lambda arcs. This can be interpreted as two abnormal IgM or as one abnormal IgM with one Bence Jones protein. N = normal serum; A = alpha antiserum; L = lambda antiserum; P = patients' serum; M = mu antiserum; T = trivalent or polyvalent antiserum; G = gamma antiserum; K = kappa antiserum.
case with an unknown diagnosis. Immunochemically, the ten sera were composed of monoclonal proteins consisting of four IgG-kappa, one IgG-lambda, one IgA-kappa, two IgA-lambda and two IgM-kappa. The three urines all contained kappa Bence Jones protein. Group III. Patients were asymptomatic with a polyclonal IEP pattern. The clinical diagnoses were one each chronic lymphocytic leukemia, autoimmune hemolytic anemia, syphilis, epilepsy, multiple tumors, and rheumatoid arthritis. The group also included one healthy laboratory technician. Immunochemically, all six sera and the one urine showed no abnormal precipitin arc by IEP.
After electrophoresis, the first strip of agarose-coated film was cut and stained with 0.2% amido black or, in case the protein concentration was low, with 0.2% Coomassie brilliant blue. The unstained strips were kept in a moist chamber until ready for use. The abnormal bands were then located from the stained strip and aligned with the unstained strips. Fifteen microliters of the respective monospecific antisera against IgG, IgA, IgM, kappa, or lambda were applied through the slots of a plastic sheet onto the unstained strips at the locations corresponding to the abnormal bands. In cases that showed an abnormality of light chains only, IgD and IgE antisera were also used. After 10 min, excess antisera were removed with blotter paper, and the agarose-coated film was pressed with one moistened and three or four dry sheets of blotter paper plus a weight of 1-2 kg for 10 min. The film was then deproteinized with saline solution for 2 - 6 hours, rinsed in tap water for 10 min, and finally air-dried and stained with amido black or Coomassie brilliant blue. Immunoprecipitin is usually visible within an hour
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
M
Electrophoresis
Vol. 72 • No. 1
7
GAMMOPATHY STUDY WITH IMMUNOFIXATION
after antiserum application. When necessary, the deproteinizing and staining procedures can be omitted. Immunoelectrophoresis A parallel Immunoelectrophoresis was performed on every specimen studied for comparison. The IEP method used was that of Grabar and Williams,8 the apparatus from Millipore Biomedica. For the interpretation of immunoelectrophoretic patterns, the readers are referred to Penn and Batya.15 Results
Discussion Immunoelectrophoresis is a very useful technic for quantitative and semiquantitative studies of immuno-
M
I1
FIG. 2. Biclonal gammopathy. Immunofixation pattern shows that one monoclonal band consists of IgM-kappa, another IgMlambda. L = lambda antiserum; K = kappa antiserum; M = mu antiserum.
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
In monoclonal gammopathy, immunofixation electrophoresis showed well-defined bands reacting to the particular monospecific antisera (e.g., IgG and kappa or lambda for IgG myeloma; IgM and kappa or lambda for macroglobulinemia) (Figs. 1 and 2). In polyclonal gammopathy, the bands were either irregular (Fig. 6) or well defined and reacted to all antisera (Fig. 4). In Group I, the results of IEP and immunofixation electrophoresis were consistent with each other in all 40 cases. In Group II, the results of these two technics were identical in eight cases but different for two sera and three urine specimens. The two sera both
showed IgM-kappa abnormalities by IEP but reacted to IgM, IgG, kappa, and lambda antisera by immunofixation electrophoresis. The clinical diagnoses for these two patients were benign prostatic hypertrophy and syncope of unknown etiology. The three urine specimens showed only the presence of kappa light chains without other immunoglobulins by IEP (Fig. 5), but all three specimens showed all immunoglobulins by immunofixation electrophoresis. The clinical diagnoses for these three patients were urinary tract infection, acute glomerulonephritis, and septicemia. The results for Group III were identical with the two methods in terms of clonality (i.e., all were polyclonal by both methods). The case of rheumatoid arthritis showed polyclonal IgM by IEP, but additional IgG fraction was identified by immunofixation electrophoresis (Figs. 3 and 6). Therefore, immunofixation electrophoresis proved its immune complex nature, and IEP failed to do so. A positive latex-fixation test further indicated that the immune complex was due to rheumatoid factor.
SUN, LIEN, AND DEGNAN
8 N
r
T P G N A P
globulins as well as other serum proteins. However, there are several faults existent with this method: Light-chain Study. Since all immunoglobulins contain kappa and lambda light chains, the kappa and lambda arcs demonstrated by IEP represent the entire mixture of immunoglobulins. In cases of early IgA myeloma or IgM macroglobulinemia, for instance, the changes in light chain may be masked by the predominant
light chain components of IgG, the so-called umbrella effect.716 Immunofixation electrophoresis, on the other hand, demonstrates specifically the light-chain component of the monoclonal protein, thus avoiding confusion. Localization of Monoclonal Band. IEP can indicate the approximate location of a monoclonal band, which is usually not a problem when it is located in the gamma zone. However, when the monoclonal band is located in the alpha or beta zone, especially when it is superimposed with a normal band, the identification may be difficult (Fig. 7). In two of our cases, electrophoretic patterns looked normal at first glance. The abnormal bands were identified only by immunofixation electrophoresis retrospectively. Bence Jones Proteinuria. In cases where renal permeability is increased, or renal tubular function has deteriorated, or even in normal persons, small numbers of light chains can be present in the urine.619 Since the lambda light-chain concentration in normal situations is only half as high as the kappa light-chain concentration,6 it may disappear earlier in urine after catabolism in renal tubules or enzymatic cleavage in the excreted urine and leave kappa light chains alone in the urine specimen. This may be misinterpreted as monoclonal free light chains by inexperienced individuals. This happened in three of our cases where only kappa was demonstrated by IEP (Fig. 5), but all immunoglobulins were detected by immunofixation electrophoresis. Clinical findings in these three cases indicate that the immunofixation electrophoresis results are more likely to be correct. Biclonal Gammopathy. When there are two monoclonal bands in an electrophoretic pattern with only a slight difference in electrophoretic mobilities, the immunofixation electrophoresis is still able to distinguish them (Figs. 1 and 2). However, it is usually difficult to interpret the IEP pattern, especially when heavy chains in both clones belong to the same class. 414 Mini-monoclonal or Oligoclonal Protein Bands. A small band detected in the gamma zone may represent early monoclonal gammopathy or other proteins such as C-reactive protein,11 lysozyme,10 or post-gamma globulin.12 IEP is able to identify monoclonal gammopathy, but not non-immunoglobulins, in routine studies. Immunofixation electrophoresis, on the other hand, can distinguish a monoclonal gammopathy from a nonimmunoglobulin band; in the former, the protein band reacts to a monospecific immunoglobulin antiserum only; in the latter, the protein band does not react to any immunoglobulin antiserum, or all immunoglobulins are present in the immunofixation electrophoresis pattern because the antisera react with background protein (Fig. 4). The non-immunoglobulin band can be
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
FIG. 3. Immune-complex disease. Immunoelectrophoretic pattern shows prominent arcs of mu, kappa, and lambda at the same position, indicating polyclonal IgMgammopathy. Note that the kappa and lambda light chain antisera react with the immune complexes (arrows), and these additional precipitins are distinguished from the normal kappa and lambda arcs. A possible interpretation is that the rheumatoid factor, as an antibody, reacts with immune complexes within or generated by the kappa and lambda antisera, forming the inner precipitin lines (arrows). These immune complexes may be created by absorption of the antisera or by the antiserum reacting with a serum constituent (IgG), generating a small immune complex. N = normal serum; A = alpha antiserum; L = lambda antiserum; P = patients' serum; M = mu antiserum; T = trivalent or polyvalent antiserum; G = gamma antiserum; K = kappa antiserum.
A.J.C.P. • July 1979
Vol. 72 • No. 1
GAMMOPATHY STUDY WITH IMMUNOFIXATION
no obvious change in the configuration of IgG after mercaptoethanol treatment. On the other hand, immunofixation electrophoresis clearly demonstrated the IgG component in addition to a polyclonal IgM, thus confirming the diagnosis of immune complexes (Figs. 3 and 6). Admittedly, immune complexes due to monoclonal rheumatoid factor may sometimes react with all antisera and be indistinguishable from polyclonal gammopathy. Benign versus Malignant Monoclonal Gammopathy. In our study, we divided patients into three groups in order to determine which technic correlates with the clinical findings better. In Group II, which included asymptomatic patients, two sera and three urines were positive by IEP but negative by immunofixation electrophoresis. The three urines were discussed under Bence Jones protein. The discrepancy in the two sera may be due to the fact that the immunofixation electrophoresis demonstrated not only the monoclonal IgM but also the
K
M
»
I I
FIG. 4. Immunofixation pattern of polyclonal gammopathy. A narrow band in the gamma region reacting to all antisera. Note that the pale centers in mu, kappa, and lambda chains indicate antigen excess. L = lambda antiserum; K = kappa antiserum; M = mu antiserum; A = alpha antiserum; G = gamma antiserum.
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
easily identified with specific antiserum, e.g., C-reactive protein antiserum. When multiple oligoclonal bands are present, IEP is usually useless, due to incomplete resolution of the protein bands because of diffusion; immunofixation electrophoresis, however, is capable of identifying the oligoclonal bands, since diffusion does not occur during the procedure. Nevertheless, for some well-established patterns, such as cerebrospinal fluid oligoclonal banding in multiple sclerosis, routine use of immunofixation electrophoresis is unnecessary. Immune Complexes. The distinction of monoclonal gammopathy from monoclonal immune complexes is very difficult, although possible, by IEP. Penn and Davis advocate the use of 2-mercaptoethanol to destroy the rheumatoid factor activity, if this is part of the immune complex.13 After the treatment, IgG will be released from the immune complex and resume a normal configuration and the clonality of IgM will be clearly identified. In our particular case, however, there was
9
10
SUN, LIEN, AND DEGNAN
A.J.C.P. • July 1979
I FIG. 5. Equivocal case of Bence Jones proteinuria. The immunoelectrophoretic pattern shows only a kappa light chain without corresponding heavy chains or lambda light chain. Note a fragment of heavy chain (h) is present with a fast electrophoretic mobility, indicating enzymatic cleavage has taken place. Electrophoretic pattern shows mini-monoclonal banding, which was identified by immunofixation (not shown) as polyclonal in origin. N = normal serum; A = alpha antiserum; L = lambda antiserum; P = patients' serum; M = mu antiserum; T = trivalent or polyvalent antiserum; G = gamma antiserum; K = kappa antiserum.
background IgG, which was relatively high in these two cases. It seems unlikely that either the immunofixation electrophoresis or the IEP can help to distinguish a benign from a malignant monoclonal gammopathy. Testing Time. IEP usually takes 18-48 hours to complete, depending on which manufacturer's products are used. Immunofixation electrophoresis, when stained, is nearly as time-consuming as IEP. However, one can inspect the unstained IEP pattern within a few hours, so that the clinicians can obtain some early information. The major drawback of immunofixation electrophoresis is that it cannot tell whether the band that is present in the pattern is normal or abnormal. The
1 1 G
f
A
FIG. 6. Immune-complex disease. Immunofixation shows gamma, mu, kappa, and lambda chains, indicating a polyclonal immune complex. G = gamma antiserum; A = alpha antiserum; M = mu antiserum; K = kappa antiserum; L = lambda antiserum.
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
range of optimum concentration of antigen or the optimum antigen:antibody ratio is narrower in immunofixation electrophoresis than in IEP. For instance, in two cases of IgD myeloma, where the undiluted patients' sera reacted to the IgD antiserum, showing abnormal arcs by IEP, there was no reaction by immunofixation electrophoresis until certain dilutions were made. That may be one of the reasons immunofixation electrophoresis is less reproducible than IEP and takes more skill to perform. In conclusion, IEP is still a method of choice for routine clinical service, as it clearly distinguishes a normal arc from an abnormal arc, and a pure protein from a contaminated protein, which immunofixation electrophoresis is unable to accomplish. However, for the situations mentioned above, immunofixation electrophoresis should be used to supplement IEP. The immunofixation electrophoresis technic described in this paper, applying antiserum through slots of a plastic sheet, is advantageous in that it saves antiserum and produces well-defined precipitin bands. Layering antiserum directly onto the entire electrophoretic pattern and laying an antiserum-moistened strip of cellulose acetate or filter paper onto the abnormal zone, according to our experience, are not as ideal as the present method in terms of definition of the precipitin band and quantity of serum consumed.
GAMMOPATHY STUDY WITH IMMUNOFIXATION
Vol. 72 . No. I
11
However, when the background immunoglobulins have a density equal to or higher than that of the overlapped monoclonal band, the antisera will react with both and produce an inevitable polyclonal pattern by our modified technic. Under most circumstances, though, the background immunoglobulins are of lower density than the monoclonal band and do not react with the antisera, because after the specimen is routinely diluted, their concentration is too low for detection. When there is a high immunoglobulin background, antisera should be applied over a broader area to demonstrate the relative densities of the monoclonal band and the adjacent proteins. When the former has a higher density than the latter when reacting to a monospecific antiserum, a ' monoclonal gammopathy is present. On the other hand, when the densities of the two are equal, this band contains non-immunoglobulin protein.
References 1. Afonso E: Quantitative immunoelectrophoresis of serum proteins. Clin Chim Acta 10:114-122, 1964 2. Alper CA: Genetic polymorphism of complement components as a probe of structure and function, Progress in Immunology, First International Congress of Immunology. Edited by Amos B. New York, Academic Press, 1971, pp 609-624 3. Alper CA, Johnson AM: Immunofixation electrophoresis: A technique for the study of protein polymorphism. Vox Sang 17:445-452, 1969 4. Cawley LP, Minard BJ, Tourtellotte WW, et al: Immunofixation electrophoretic techniques applied to identification of proteins in serum and cerebrospinal fluid. Clin Chem 22:1262-1268, 1976 5. Chang CH, InglisNR: Convenient immunofixation electrophoresis on cellulose acetate membrane. Clin Chim Acta 65:91-97, 1975 6. Deegan MJ: Bence Jones protein: Nature, metabolism, detection and significance. Ann Clin Lab Sci 6:38-46, 1976 7. Frangione B, Franklin EC: Heavy chain diseases: Clinical features and molecular significance of the disordered immunoglobulin structure. Semin Hematol 10:53-64, 1973 8. Grabar P, William CA: Methode permettant L'etude conjugee des proprtetes electrophoretiques et immunocrimiques d'un melange de proteines. Application du serum sanguin. Biochim Biophys Acta 10:193-194, 1953 9. Johnson AM: Genetic typing of alpha,-antitrypsin in immunofixation electrophoresis. Identification of subtypes of Pi M. J Lab Clin Med 87:152-163, 1976 10. Kohn J: The laboratory investigation of paraproteinemia, Recent Advances in Clinical Pathology. Series 6. Edited by Dyke SC. Edinburgh and London, Churchill Livingstone, 1973, pp 363-401 11. Laurell CB: Composition and variation of the gel electrophoretic fractions of plasma, cerebrospinal fluid, and urine. Scan J Clin Lab Invest 29 (suppl 124):71-82, 1972 12. Leterre EC, Callawaert A, Heremans JF, et al: Electrophoretic morphology of gamma globulins in cerebrospinal fluid of multiple sclerosis and other diseases of the nervous system. Neurology 20:982-990, 1970 13. Penn GM, Davis T: Identification of Myeloma Protein. Chicago, American Society of Clinical Pathologists, 1976, pp 26-27 14. Penn GM, Batya J, Moore S: Diclonal gammopathy. Clinical Immunology Check Sample Critique, No. CI-2. Chicago, American Society of Clinical Pathologists, 1977
FIG. 7. IgG-lambda myeloma. A monoclonal protein band is present at atypical region overlapping with beta 2 , which is identified by immunofixation as IgG-lambda (only gamma antiserum is illustrated). N = normal serum: A = alpha antiserum; L = lambda antiserum: P = patients' serum; M = mu antiserum; T = trivalent or polyvalent antiserum; G = gamma antiserum; K = kappa antiserum.
15. Penn GM, Batya J: Interpretation of Immunoelectrophoretic Patterns. Chicago, American Society of Clinical Pathologists, 1978 16. Ritchie RF, Smith R: Immunofixation. I. General principles and application to agarose gel electrophoresis. Clin Chem 22: 497-499, 1976 17. Ritchie RF, Smith R: Immunofixation. II. Application to typing of alpha,-antitrypsin at acid pH. Clin Chem 22:1735-1737, 1976 18. Ritchie RF, Smith R: Immunofixation. III. Application to the study of monoclonal proteins. Clin Chem 22:1982-1985, 1976 19. Solomon A: Bence Jones proteins and light chains of immunoglobulins [Second of two parts]. N Engl J Med 294:91-98, 1976 20. Sun T, Chan SK, Gross S: Evaluation of a high-resolution electrophoresis system. Am J Clin Pathol 67:247-250, 1977 21. Sun T, Lien Y, Kunins M: Comparison of immunoelectrophoresis and immunofixation technique in the study of monoclonal gammopathy. Clin Chem 23:1154-1155, 1977
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
Acknowledgment. The specimen used for Figures 1 and 2 is the Clinical Immunology Check Sample (ASCP) CI-2, 1977.
Sun, Tsieh, Lien, York Y., and Degnan, Thomas: Study of gammopathies with immunofixation electrophoresis. Am J Clin Pathol 72: 5-11, 1979. Immunoflxation electrophoresis and immunoelectrophoresis were compared in 60 samples (51 sera and 9 urines) containing apparently homogenous bands, detected by electrophoresis. To determine the correlation with clinicalfindings,the patients were divided into three groups: (1) symptomatic with monoclonal immunoelectrophoretic patterns; (2) asymptomatic with monoclonal immunoelectrophoretic patterns; (3) asymptomatic with polyclonal immunoelectrophoretic patterns. The results from immunoelectrophoresis and immunofixation electrophoresis were consistent with each other in all cases of Groups I and III in terms of clonality (i.e., whether monoclonal or polyclonal) and immunoglobulin class; whereas in Group II, which was composed of asymptomatic patients, two sera and three urines were identified to have monoclonal changes by immunoelectrophoresis but polyclonal changes by immunofixation electrophoresis. It is recommended that immunoelectrophoresis still be used for routine clinical service, but supplemented by immunofixation electrophoresis in equivocal cases, namely: (1) light-chain changes masked by an umbrella effect of immunoglobulin G (IgG); (2) abnormal bands located in atypical areas or overlapped with a normal serum protein; (3) a normal-looking "free" light chain present in the urine, mimicking Bence Jones protein; (4) controversial cases of biclonal gammopathy; (5) mini-monoclonal or oligoclonal protein bands; (6) immune complexes. (Key words: Immunofixation electrophoresis; Gammopathies; Immunoelectrophoresis.)
Departments of Pathology and Medicine, Cornell University Medical College, New York, New York, and Departments of Laboratories and Medicine, North Shore University Hospital, Manhasset, New York
Materials and Methods Patients'
Samples
Patients with apparently homogenous bands in serum or urine (60 cases), detected by electrophoresis, were divided into three groups according to clinical history and further immunochemical evaluation: Group I. Patients were symptomatic with a monoclonal IEP pattern. Clinical diagnoses were plasma-cell myeloma or primary macroglobulinemia. These included 35 sera and five urine specimens. Immunochemically, the 35 sera contained monoclonal proteins consisting of 12 IgG-kappa, four IgG-lambda, four IgA-kappa, three IgA-lambda, four IgM-kappa, one IgM-lambda, two IgD-lambda, two biclonal gammopathies (IgGkappa and IgM-lambda; IgG-kappa and IgA-kappa), one kappa Bence Jones protein, and two lambda Bence Jones protein. The five urine specimens consisted of two kappa Bence Jones protein and three lambda Bence Jones protein. Group II. Asymptomatic with a monoclonal IEP pattern, this group contained ten sera and three urines. The clinical diagnoses were five benign monoclonal gammopathy or early myeloma (or macroglobulinemia), two benign prostatic hypertrophy, one each septicemia, acute glomerulonephritis, lymphoma, urinary tract infection, and syncope of unknown etiology, and one
IMMUNOFIXATION ELECTROPHORESIS is similar to immunoelectrophoresis (IEP) in principle, but instead of allowing antigen and antibody to diffuse in the supporting medium, antisera are applied, following electrophoresis, to the surface of the medium. The formation of an immunoprecipitin with a monospecific antiserum identifies the protein of interest. This basic concept was first described by Afonso in 1964,' but it was not a practical technic until Alper and Johnson modified it in 1969.3 Since then immunofixation electrophoresis has been used for the study of protein polymorphism, 2 genetic typing of alpha,-antitrypsin, 917 and identification of serum protein fractions. 516 The applicaReceived March 27, 1978; received revised manuscript and accepted for publication June 6, 1978. This work was done at North Shore University Hospital and was partly presented at the 29th National Meeting of the American Association for Clinical Chemistry, Chicago, Illinois, July 20, 1977. Address reprint requests to Dr. Sun: Department of Laboratories, North Shore University Hospital, Manhasset, New York 11030.
0002-9173/79/0700/0005 $00.85 © American Society of Clinical Pathologists
5
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
tion of immunofixation electrophoresis for the study of gammopathies was reported recently by Cawley and associates 4 and by Ritchie and Smith. 18 Included in these two reports were a few cases that illustrated the usefulness of immunofixation electrophoresis. However, the exact number of patients investigated was not specified. The present paper reports a systematic study of different types of gammopathy by immunofixation electrophoresis and compares it with IEP. A modified technic of immunofixation electrophoresis is also described. 21
6
SUN, LIEN, AND DEGNAN
A.J.C.P. • July 1979
Antisera tx»sv
The antisera used, including polyvalent, trivalent (IgG, IgA, IgM), monospecific for IgG (y) IgA (a), IgM (/x), IgD (8), IgE (e), kappa, and lambda, were from Kallestad (Chaska, Minnesota), Millipore (Bedford, Massachusetts), Behring (Somerville, New Jersey), Meloy (Springfield, Virginia) and Hyland (Costa Mesa, California). When equivocal results were obtained, antisera from different commercial sources were used for confirmation. Serum and Urinary
• ' (k
p K
c
A high-resolution electrophoresis system manufactured by Millipore Biomedica was used as described previously. 20 A battery of six electrophoretograms was obtained for each sample. Fifteen microliters of undiluted specimen were applied to the first strip of a buffered-agarose-coated plastic film as a control. Equal amounts of saline-diluted specimen (1:5) were applied to the other five strips. In the case of IgD myeloma, or other myelomas with very high concentrations of immunoglobulins, further dilutions were made to a point that antibody was in excess or in equivalence with the specific antigen. The antigen:antibody ratio was estimated by quantitation of individual immunoglobulins with the radial immunodiffusion technic (Kallestad) or by Ouchterlony double-diffusion analysis. Electrophoresis was run for 45 min at 200 v. lmmunofixation
FIG. 1. Biclonal gammopathy. Immunoelectrophoretic pattern shows abnormal mu, kappa, and lambda arcs. This can be interpreted as two abnormal IgM or as one abnormal IgM with one Bence Jones protein. N = normal serum; A = alpha antiserum; L = lambda antiserum; P = patients' serum; M = mu antiserum; T = trivalent or polyvalent antiserum; G = gamma antiserum; K = kappa antiserum.
case with an unknown diagnosis. Immunochemically, the ten sera were composed of monoclonal proteins consisting of four IgG-kappa, one IgG-lambda, one IgA-kappa, two IgA-lambda and two IgM-kappa. The three urines all contained kappa Bence Jones protein. Group III. Patients were asymptomatic with a polyclonal IEP pattern. The clinical diagnoses were one each chronic lymphocytic leukemia, autoimmune hemolytic anemia, syphilis, epilepsy, multiple tumors, and rheumatoid arthritis. The group also included one healthy laboratory technician. Immunochemically, all six sera and the one urine showed no abnormal precipitin arc by IEP.
After electrophoresis, the first strip of agarose-coated film was cut and stained with 0.2% amido black or, in case the protein concentration was low, with 0.2% Coomassie brilliant blue. The unstained strips were kept in a moist chamber until ready for use. The abnormal bands were then located from the stained strip and aligned with the unstained strips. Fifteen microliters of the respective monospecific antisera against IgG, IgA, IgM, kappa, or lambda were applied through the slots of a plastic sheet onto the unstained strips at the locations corresponding to the abnormal bands. In cases that showed an abnormality of light chains only, IgD and IgE antisera were also used. After 10 min, excess antisera were removed with blotter paper, and the agarose-coated film was pressed with one moistened and three or four dry sheets of blotter paper plus a weight of 1-2 kg for 10 min. The film was then deproteinized with saline solution for 2 - 6 hours, rinsed in tap water for 10 min, and finally air-dried and stained with amido black or Coomassie brilliant blue. Immunoprecipitin is usually visible within an hour
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
M
Electrophoresis
Vol. 72 • No. 1
7
GAMMOPATHY STUDY WITH IMMUNOFIXATION
after antiserum application. When necessary, the deproteinizing and staining procedures can be omitted. Immunoelectrophoresis A parallel Immunoelectrophoresis was performed on every specimen studied for comparison. The IEP method used was that of Grabar and Williams,8 the apparatus from Millipore Biomedica. For the interpretation of immunoelectrophoretic patterns, the readers are referred to Penn and Batya.15 Results
Discussion Immunoelectrophoresis is a very useful technic for quantitative and semiquantitative studies of immuno-
M
I1
FIG. 2. Biclonal gammopathy. Immunofixation pattern shows that one monoclonal band consists of IgM-kappa, another IgMlambda. L = lambda antiserum; K = kappa antiserum; M = mu antiserum.
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
In monoclonal gammopathy, immunofixation electrophoresis showed well-defined bands reacting to the particular monospecific antisera (e.g., IgG and kappa or lambda for IgG myeloma; IgM and kappa or lambda for macroglobulinemia) (Figs. 1 and 2). In polyclonal gammopathy, the bands were either irregular (Fig. 6) or well defined and reacted to all antisera (Fig. 4). In Group I, the results of IEP and immunofixation electrophoresis were consistent with each other in all 40 cases. In Group II, the results of these two technics were identical in eight cases but different for two sera and three urine specimens. The two sera both
showed IgM-kappa abnormalities by IEP but reacted to IgM, IgG, kappa, and lambda antisera by immunofixation electrophoresis. The clinical diagnoses for these two patients were benign prostatic hypertrophy and syncope of unknown etiology. The three urine specimens showed only the presence of kappa light chains without other immunoglobulins by IEP (Fig. 5), but all three specimens showed all immunoglobulins by immunofixation electrophoresis. The clinical diagnoses for these three patients were urinary tract infection, acute glomerulonephritis, and septicemia. The results for Group III were identical with the two methods in terms of clonality (i.e., all were polyclonal by both methods). The case of rheumatoid arthritis showed polyclonal IgM by IEP, but additional IgG fraction was identified by immunofixation electrophoresis (Figs. 3 and 6). Therefore, immunofixation electrophoresis proved its immune complex nature, and IEP failed to do so. A positive latex-fixation test further indicated that the immune complex was due to rheumatoid factor.
SUN, LIEN, AND DEGNAN
8 N
r
T P G N A P
globulins as well as other serum proteins. However, there are several faults existent with this method: Light-chain Study. Since all immunoglobulins contain kappa and lambda light chains, the kappa and lambda arcs demonstrated by IEP represent the entire mixture of immunoglobulins. In cases of early IgA myeloma or IgM macroglobulinemia, for instance, the changes in light chain may be masked by the predominant
light chain components of IgG, the so-called umbrella effect.716 Immunofixation electrophoresis, on the other hand, demonstrates specifically the light-chain component of the monoclonal protein, thus avoiding confusion. Localization of Monoclonal Band. IEP can indicate the approximate location of a monoclonal band, which is usually not a problem when it is located in the gamma zone. However, when the monoclonal band is located in the alpha or beta zone, especially when it is superimposed with a normal band, the identification may be difficult (Fig. 7). In two of our cases, electrophoretic patterns looked normal at first glance. The abnormal bands were identified only by immunofixation electrophoresis retrospectively. Bence Jones Proteinuria. In cases where renal permeability is increased, or renal tubular function has deteriorated, or even in normal persons, small numbers of light chains can be present in the urine.619 Since the lambda light-chain concentration in normal situations is only half as high as the kappa light-chain concentration,6 it may disappear earlier in urine after catabolism in renal tubules or enzymatic cleavage in the excreted urine and leave kappa light chains alone in the urine specimen. This may be misinterpreted as monoclonal free light chains by inexperienced individuals. This happened in three of our cases where only kappa was demonstrated by IEP (Fig. 5), but all immunoglobulins were detected by immunofixation electrophoresis. Clinical findings in these three cases indicate that the immunofixation electrophoresis results are more likely to be correct. Biclonal Gammopathy. When there are two monoclonal bands in an electrophoretic pattern with only a slight difference in electrophoretic mobilities, the immunofixation electrophoresis is still able to distinguish them (Figs. 1 and 2). However, it is usually difficult to interpret the IEP pattern, especially when heavy chains in both clones belong to the same class. 414 Mini-monoclonal or Oligoclonal Protein Bands. A small band detected in the gamma zone may represent early monoclonal gammopathy or other proteins such as C-reactive protein,11 lysozyme,10 or post-gamma globulin.12 IEP is able to identify monoclonal gammopathy, but not non-immunoglobulins, in routine studies. Immunofixation electrophoresis, on the other hand, can distinguish a monoclonal gammopathy from a nonimmunoglobulin band; in the former, the protein band reacts to a monospecific immunoglobulin antiserum only; in the latter, the protein band does not react to any immunoglobulin antiserum, or all immunoglobulins are present in the immunofixation electrophoresis pattern because the antisera react with background protein (Fig. 4). The non-immunoglobulin band can be
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
FIG. 3. Immune-complex disease. Immunoelectrophoretic pattern shows prominent arcs of mu, kappa, and lambda at the same position, indicating polyclonal IgMgammopathy. Note that the kappa and lambda light chain antisera react with the immune complexes (arrows), and these additional precipitins are distinguished from the normal kappa and lambda arcs. A possible interpretation is that the rheumatoid factor, as an antibody, reacts with immune complexes within or generated by the kappa and lambda antisera, forming the inner precipitin lines (arrows). These immune complexes may be created by absorption of the antisera or by the antiserum reacting with a serum constituent (IgG), generating a small immune complex. N = normal serum; A = alpha antiserum; L = lambda antiserum; P = patients' serum; M = mu antiserum; T = trivalent or polyvalent antiserum; G = gamma antiserum; K = kappa antiserum.
A.J.C.P. • July 1979
Vol. 72 • No. 1
GAMMOPATHY STUDY WITH IMMUNOFIXATION
no obvious change in the configuration of IgG after mercaptoethanol treatment. On the other hand, immunofixation electrophoresis clearly demonstrated the IgG component in addition to a polyclonal IgM, thus confirming the diagnosis of immune complexes (Figs. 3 and 6). Admittedly, immune complexes due to monoclonal rheumatoid factor may sometimes react with all antisera and be indistinguishable from polyclonal gammopathy. Benign versus Malignant Monoclonal Gammopathy. In our study, we divided patients into three groups in order to determine which technic correlates with the clinical findings better. In Group II, which included asymptomatic patients, two sera and three urines were positive by IEP but negative by immunofixation electrophoresis. The three urines were discussed under Bence Jones protein. The discrepancy in the two sera may be due to the fact that the immunofixation electrophoresis demonstrated not only the monoclonal IgM but also the
K
M
»
I I
FIG. 4. Immunofixation pattern of polyclonal gammopathy. A narrow band in the gamma region reacting to all antisera. Note that the pale centers in mu, kappa, and lambda chains indicate antigen excess. L = lambda antiserum; K = kappa antiserum; M = mu antiserum; A = alpha antiserum; G = gamma antiserum.
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
easily identified with specific antiserum, e.g., C-reactive protein antiserum. When multiple oligoclonal bands are present, IEP is usually useless, due to incomplete resolution of the protein bands because of diffusion; immunofixation electrophoresis, however, is capable of identifying the oligoclonal bands, since diffusion does not occur during the procedure. Nevertheless, for some well-established patterns, such as cerebrospinal fluid oligoclonal banding in multiple sclerosis, routine use of immunofixation electrophoresis is unnecessary. Immune Complexes. The distinction of monoclonal gammopathy from monoclonal immune complexes is very difficult, although possible, by IEP. Penn and Davis advocate the use of 2-mercaptoethanol to destroy the rheumatoid factor activity, if this is part of the immune complex.13 After the treatment, IgG will be released from the immune complex and resume a normal configuration and the clonality of IgM will be clearly identified. In our particular case, however, there was
9
10
SUN, LIEN, AND DEGNAN
A.J.C.P. • July 1979
I FIG. 5. Equivocal case of Bence Jones proteinuria. The immunoelectrophoretic pattern shows only a kappa light chain without corresponding heavy chains or lambda light chain. Note a fragment of heavy chain (h) is present with a fast electrophoretic mobility, indicating enzymatic cleavage has taken place. Electrophoretic pattern shows mini-monoclonal banding, which was identified by immunofixation (not shown) as polyclonal in origin. N = normal serum; A = alpha antiserum; L = lambda antiserum; P = patients' serum; M = mu antiserum; T = trivalent or polyvalent antiserum; G = gamma antiserum; K = kappa antiserum.
background IgG, which was relatively high in these two cases. It seems unlikely that either the immunofixation electrophoresis or the IEP can help to distinguish a benign from a malignant monoclonal gammopathy. Testing Time. IEP usually takes 18-48 hours to complete, depending on which manufacturer's products are used. Immunofixation electrophoresis, when stained, is nearly as time-consuming as IEP. However, one can inspect the unstained IEP pattern within a few hours, so that the clinicians can obtain some early information. The major drawback of immunofixation electrophoresis is that it cannot tell whether the band that is present in the pattern is normal or abnormal. The
1 1 G
f
A
FIG. 6. Immune-complex disease. Immunofixation shows gamma, mu, kappa, and lambda chains, indicating a polyclonal immune complex. G = gamma antiserum; A = alpha antiserum; M = mu antiserum; K = kappa antiserum; L = lambda antiserum.
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
range of optimum concentration of antigen or the optimum antigen:antibody ratio is narrower in immunofixation electrophoresis than in IEP. For instance, in two cases of IgD myeloma, where the undiluted patients' sera reacted to the IgD antiserum, showing abnormal arcs by IEP, there was no reaction by immunofixation electrophoresis until certain dilutions were made. That may be one of the reasons immunofixation electrophoresis is less reproducible than IEP and takes more skill to perform. In conclusion, IEP is still a method of choice for routine clinical service, as it clearly distinguishes a normal arc from an abnormal arc, and a pure protein from a contaminated protein, which immunofixation electrophoresis is unable to accomplish. However, for the situations mentioned above, immunofixation electrophoresis should be used to supplement IEP. The immunofixation electrophoresis technic described in this paper, applying antiserum through slots of a plastic sheet, is advantageous in that it saves antiserum and produces well-defined precipitin bands. Layering antiserum directly onto the entire electrophoretic pattern and laying an antiserum-moistened strip of cellulose acetate or filter paper onto the abnormal zone, according to our experience, are not as ideal as the present method in terms of definition of the precipitin band and quantity of serum consumed.
GAMMOPATHY STUDY WITH IMMUNOFIXATION
Vol. 72 . No. I
11
However, when the background immunoglobulins have a density equal to or higher than that of the overlapped monoclonal band, the antisera will react with both and produce an inevitable polyclonal pattern by our modified technic. Under most circumstances, though, the background immunoglobulins are of lower density than the monoclonal band and do not react with the antisera, because after the specimen is routinely diluted, their concentration is too low for detection. When there is a high immunoglobulin background, antisera should be applied over a broader area to demonstrate the relative densities of the monoclonal band and the adjacent proteins. When the former has a higher density than the latter when reacting to a monospecific antiserum, a ' monoclonal gammopathy is present. On the other hand, when the densities of the two are equal, this band contains non-immunoglobulin protein.
References 1. Afonso E: Quantitative immunoelectrophoresis of serum proteins. Clin Chim Acta 10:114-122, 1964 2. Alper CA: Genetic polymorphism of complement components as a probe of structure and function, Progress in Immunology, First International Congress of Immunology. Edited by Amos B. New York, Academic Press, 1971, pp 609-624 3. Alper CA, Johnson AM: Immunofixation electrophoresis: A technique for the study of protein polymorphism. Vox Sang 17:445-452, 1969 4. Cawley LP, Minard BJ, Tourtellotte WW, et al: Immunofixation electrophoretic techniques applied to identification of proteins in serum and cerebrospinal fluid. Clin Chem 22:1262-1268, 1976 5. Chang CH, InglisNR: Convenient immunofixation electrophoresis on cellulose acetate membrane. Clin Chim Acta 65:91-97, 1975 6. Deegan MJ: Bence Jones protein: Nature, metabolism, detection and significance. Ann Clin Lab Sci 6:38-46, 1976 7. Frangione B, Franklin EC: Heavy chain diseases: Clinical features and molecular significance of the disordered immunoglobulin structure. Semin Hematol 10:53-64, 1973 8. Grabar P, William CA: Methode permettant L'etude conjugee des proprtetes electrophoretiques et immunocrimiques d'un melange de proteines. Application du serum sanguin. Biochim Biophys Acta 10:193-194, 1953 9. Johnson AM: Genetic typing of alpha,-antitrypsin in immunofixation electrophoresis. Identification of subtypes of Pi M. J Lab Clin Med 87:152-163, 1976 10. Kohn J: The laboratory investigation of paraproteinemia, Recent Advances in Clinical Pathology. Series 6. Edited by Dyke SC. Edinburgh and London, Churchill Livingstone, 1973, pp 363-401 11. Laurell CB: Composition and variation of the gel electrophoretic fractions of plasma, cerebrospinal fluid, and urine. Scan J Clin Lab Invest 29 (suppl 124):71-82, 1972 12. Leterre EC, Callawaert A, Heremans JF, et al: Electrophoretic morphology of gamma globulins in cerebrospinal fluid of multiple sclerosis and other diseases of the nervous system. Neurology 20:982-990, 1970 13. Penn GM, Davis T: Identification of Myeloma Protein. Chicago, American Society of Clinical Pathologists, 1976, pp 26-27 14. Penn GM, Batya J, Moore S: Diclonal gammopathy. Clinical Immunology Check Sample Critique, No. CI-2. Chicago, American Society of Clinical Pathologists, 1977
FIG. 7. IgG-lambda myeloma. A monoclonal protein band is present at atypical region overlapping with beta 2 , which is identified by immunofixation as IgG-lambda (only gamma antiserum is illustrated). N = normal serum: A = alpha antiserum; L = lambda antiserum: P = patients' serum; M = mu antiserum; T = trivalent or polyvalent antiserum; G = gamma antiserum; K = kappa antiserum.
15. Penn GM, Batya J: Interpretation of Immunoelectrophoretic Patterns. Chicago, American Society of Clinical Pathologists, 1978 16. Ritchie RF, Smith R: Immunofixation. I. General principles and application to agarose gel electrophoresis. Clin Chem 22: 497-499, 1976 17. Ritchie RF, Smith R: Immunofixation. II. Application to typing of alpha,-antitrypsin at acid pH. Clin Chem 22:1735-1737, 1976 18. Ritchie RF, Smith R: Immunofixation. III. Application to the study of monoclonal proteins. Clin Chem 22:1982-1985, 1976 19. Solomon A: Bence Jones proteins and light chains of immunoglobulins [Second of two parts]. N Engl J Med 294:91-98, 1976 20. Sun T, Chan SK, Gross S: Evaluation of a high-resolution electrophoresis system. Am J Clin Pathol 67:247-250, 1977 21. Sun T, Lien Y, Kunins M: Comparison of immunoelectrophoresis and immunofixation technique in the study of monoclonal gammopathy. Clin Chem 23:1154-1155, 1977
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
Acknowledgment. The specimen used for Figures 1 and 2 is the Clinical Immunology Check Sample (ASCP) CI-2, 1977.
