• Welcome
• About
• News & Updates
• Guidelines
• Meetings
• Michael Morley Lecture
• Travel Bursaries
• Science
• Cinical Trials
• Links

Paraproteins: a review article and MCQ's

by

Dr. Joanna Sheldon PhD, SRCS, MRCPath
Professor Pamela Riches PhD, SRCS, FRCPath
Protein Reference and Immunology Department
St. George's Hospital
Blackshaw Road
London SW17 0RE

Tel: 0208 725 5752
e-mail:

This article is © Rila Publications Ltd. Visit the Rila educational website, which includes the dedicated CME Haematology journal and studies approved for CPD activity.

Abstract

Paraproteins are the earliest described tumour markers and remain an essential part of the investigation, diagnosis and monitoring of patients with B cell dyscrasias.

Electrophoresis is the only reliable way of detecting a paraprotein in biological fluids. Serum electrophoresis should always be accompanied by measurement of serum IgG, IgA and IgM concentrations. Samples with raised IgA and IgM concentrations that cannot be confirmed as polyclonal by the electrophoresis pattern should be analysed by immunofixation to exclude small paraprotein bands obscured by one of the normal zones.

Presence of monoclonal free light chains or Bence Jones protein in urine is suggestive of B cell malignancy and may be the only tumour marker. Whatever analytical system is being used, a trace of albumin should be visible in every urine sample to indicate adequate sensitivity.

The detection of a paraprotein in serum or urine must be followed-up with typing. Immunofixation remains the method of choice for paraprotein typing because it is fast, specific, flexible and easy to interpret. It is also more sensitive than electrophoresis and may detect paraprotein bands that are not visible on routine electrophoresis.

The method of choice for quantification of paraprotein is densitometric scanning of the electrophoretogram. Immunochemical quantitation is unreliable.

The investigation of serum and/or urine for paraproteins remains a subjective analysis. Sophisticated technology and automation cannot replace technical skill, scientific knowledge and experience.

Introduction

Paraproteins is the term given to abnormal proteins that occur in the blood, urine and tissues and many important discoveries were made before electricity, yet alone computerisation, was commonplace! The first report of an abnormal protein related to a tumour was made by Henry Bence Jones^[1]. A patient, Mr. Thomas Alexander McBean, was vaulting out of an underground cavern and felt something snap within his chest, this caused him 'intense agony' but a strengthening plaster to the chest gave some relief. The pain recurred, a pound of blood was removed and leeches were applied for 'maintenance therapy' - not surprisingly, McBean suffered considerable and increasing weakness!

Eventually, McBean was referred to a Harley Street Physician, Dr. William Macintyre who examined McBean's urine because oedema had been observed. The urine was described to 'abound in animal matter' but heating and the addition of nitric acid made it clear and cooling made the precipitate reappear^[2]. A sample of this urine was sent to Henry Bence Jones, a renowned Chemical Pathologist at St. George's Hospital who confirmed these strange urine properties. McBean's death on the first of January 1846 was recorded as due to 'atrophy from albuminuria'. Post mortem examination showed soft, brittle ribs filled with a 'gelatiniform' substance of a blood red colour and unctous feel'; the sternum and thoracic and lumbar vertebra showed the same changes. It was Bence Jones who recognised the potential importance of the abnormal urine protein in the diagnosis of what he described as mollities ossium and what we now know as multiple myeloma.

In 1937, Tiselius separated serum globulins into three parts, which he called the α, β and γ globulins. The term paraprotein was introduced by Apitz in 1940 to describe the abnormal proteins in blood, urine and tissues that are produced by myeloma cells. In 1959, Heremans used the term 'immunoglobulins' for the 5 groups of proteins (designated IgG, IgA, IgM, IgD and IgE) within the gamma globulins.

The immunoglobulins have a common basic monomeric unit of a Y-shaped molecule consisting of two identical heavy chains (α, γ, μ, δ, ε) and two identical light chains (κ and λ). The monomers combine to form dimers (the predominant form of IgA in secretions) or pentamers (IgM). The immunoglobulin chains are divided into two regions referred to as the constant region, which confers function e.g. macrophage binding or complement binding, to the immunoglobulin molecule and the variable region where antigen binding occurs. In this region, the amino acid sequence of the heavy and light chains makes a three dimensional shape that fits an antigen.

As the name suggests, the variable region shows huge diversity enabling the recognition of an enormous range of antigens. There are various estimates of the potential number of variations ranging from 10⁶ to 10¹⁰ possible different specificities. The different amino acid sequences within the variable region results in variation of charge and the very broad zone of gamma globulins that we are familiar with in the serum protein electrophoretic separation.

Immunoglobulins are the product of B lymphocytes. Proliferation of a single B cell produces a 'clone' producing immunoglobulin with identical heavy and light chains, exactly the same amino acid sequence and therefore electrophoretic mobility. This will show as an abnormal band on the electrophoretic separation - a so called paraprotein.

The B cells usually produce a slight excess of light chains; these are polyclonal, of low molecular weight and are cleared through the kidney. A malignant (or non-malignant) transformation of a B cells may lead to proliferation of that cell, secretion of a monoclonal immunoglobulin and secretion of excess monoclonal free light chains or Bence Jones protein.

There are a number of reasons why the investigation of serum and/or urine for a paraprotein is a useful and these include:

• the presence of a paraprotein in serum and/or urine can be used as a diagnostic criterion in some B cell malignancies
• in B cell malignancy, the paraprotein is a sensitive tumour marker
• immnoglobulin fragments (particularly Bence Jones protein) are rarely found in non-malignant conditions
• the serum and/or urine gives an 'overview' of the B cells products while B cell disease may be patchily distributed
• some symptoms may be directly related to the presence of a paraprotein e.g. hyperviscosity with high concentrations of IgM paraproteins

Electrophoresis is the only reliable method for the detection of paraproteins in serum and or urine. Automated quantification of kappa and lambda light chains and calculation of the ratio has been proposed as an alternative. In practice, this method has many limitations; it is poor at identifying serum paraproteins at low concentrations (below 5g/l) and identifying situations where there are two distinct paraprotein types. It is also poor at detecting Bence Jones protein in addition to an intact immunoglobulin paraprotein and identifying paraproteins where a polyclonal increase in immunoglobulins is dominated by one light chain type. This system does not lend itself to typing IgD or IgE paraproteins simply because these antisera are rarely available for nephelometric analyses.

Detection of paraproteins

Agarose electrophoresis is the most common method currently in use in the U.K. for the detection of paraproteins in serum and urine^[3]. There are still a few users of cellulose acetate as a support medium. There has been a trend towards automation of electrophoresis and automated agarose gel systems, suitable for both serum and urine electrophoresis, are now available. Automated capillary zone electrophoresis (CZE) systems are also available but to date, only suitable for serum electrophoresis.

Serum electrophoresis

The electrophoresis is the initial screening procedure and should therefore have sufficient resolution to do this adequately. Ideally agarose separations should be long enough (3 - 4 cm) to allow good separation of zones and show a good spread of the beta-gamma zone, which is achieved by properties of the agarose gel and buffer system that produce high endosmotic flow. The majority of serum paraproteins will be found in the region from the start of the beta to the end of the gamma zones. Occasionally paraprotein bands appear in the alpha-zones and in the post gamma region.

Paraprotein bands may be 'missed' if they are at low serum concentration (<5.0 g/l) and or where their mobility coincides with other bands such as β globulins. It is also possible to miss paraproteins where there is no suppression of normal immunoglobulin concentrations. In our laboratory, serum electrophoresis and immunoglobulin quantification are always done together as it is impossible to comment upon the clonality of immunoglobulins without serum electrophoresis and impossible to comment upon IgA and IgM concentrations from electrophoresis alone.

Immunofixation should be done on samples when no obvious paraprotein band is detected but where there is raised IgA or IgM without the increased staining of the beta-gamma region that is associated with a polyclonal increase in IgA or IgM. In our experience capillary zone electrophoresis more readily detects such polyclonally raised IgA and IgM than does agarose gel ^[4].

IgD paraproteins and free heavy chains are susceptible to post-synthetic degradation, which results in diffuse paraprotein bands on electrophoresis. These may be missed if present at low concentrations or if there is an expectation of seeing a clearly defined band.

There are a number of situations where a band is seen in a serum electrophoretic separation that is not monoclonal immunoglobulin; these include:

• additional bands in the alpha-1 region due to allotypic variation in α-1 antitrypsin
• split alpha-2 zone due to the different mobility of the haptoglobin-haemoglobin complex after intravascular haemolysis
• an additional band in the beta-gamma region due to high concentrations of C-reactive protein
• additional bands in the fast gamma region due to the presence of fibrinogen.

It is also worth noting that some paraproteins precipitate at temperatures below 37^oC - so called cryoproteins. Samples where cryoprotein is being considered must be collected, transported and separated at 37^oC. Failure to do this may result in the precipitation of the cryoprotein which will be subsequently discarded with the call pellet.

Urine electrophoresis

Urine analysis is an essential component in the investigation of patients with paraproteinaemia or with suspected B cell malignancies. An important feature used to help distinguish malignant from non-malignant conditions is the finding of immunoglobulin fragments produced by tumour cells. The fragments, which may not be detectable on serum electrophoresis separations, are usually of lower molecular weight than intact immunoglobulin molecules, pass readily through the kidney, and may be clearly visible in the urine due to the concentration effect. The finding of Bence-Jones protein (BJP) therefore provides a high index of suspicion for malignancy, although it does occur in apparently benign conditions. Even low concentrations (10mg/l) may be significant so that high sensitivity electrophoresis of urine is essential. This degree of sensitivity can be achieved by using agarose electrophoresis of un-concentrated urine and a sensitive stain.

Alternatively, concentrated urine (at least 100-fold concentration of early-morning urine is preferable) may be run in the serum systems. Whatever system is used, a trace of albumin must be visible in all urine samples; if this is not the case, the sample should be re-run or further concentrated. Where a BJP is present in significant concentrations (>100mg/l) with no accompanying glomerular or tubular proteinuria, detection is straightforward and the immunofixation identification step is unequivocal. However, the renal damage associated with BJ proteinuria, frequently results in complex, non-standard patterns requiring immunofixation to resolve the possible presence of BJP.

A low concentration of BJP may also accompany significant glomerular proteinuria in patients with light chain renal amyloidosis; the urine of any such patient should be investigated by immunofixation even in the absence of a band suggestive of BJP. Patients with serum paraproteins may show a "leak" of the serum paraprotein into the urine. This may occur with or without Bence Jones protein and immunofixation is essential to distinguish these.

A number of other proteins may appear as discrete bands on urine electrophoretic separations, particularly where there is an element of tubular proteinuria. These include the α- and β-microglobulins, lysozyme (migrating in the slow gamma region), degraded fragments of glomerular origin and rarely seminal fluid proteins. In some samples the β₂-microglobulin will be present in high concentrations and give a very prominent band.

Identification of paraproteins

The detection of a paraprotein band must always be followed up with the specific typing of the band. It is important that the heavy and light chain components are identified because this confirms monoclonality and the paraprotein type may give the clinician additional information about the underlying tumour and prognosis. Electrophoretic pattern of a patient's sample may change during the course of their disease or treatment so the initial investigations can serve as a point of reference. Complete disappearance of the paraprotein is rare but is occurring increasingly with treatment regimens using high dose chemotherapy and bone marrow or stem cell transplantation. An oligoclonal-banding pattern is sometimes seen in patients after bone marrow transplantation and it is important to distinguish this from the original paraproteinaemia.

Agarose gel immunofixation

Immunofixation on agarose gels, has been the method of choice for the detection of paraproteins for many years^[5]. It is quick, flexible and simple to interpret. It is recommended that immunofixation is done initially to test for the α, γ or μ heavy chains and the κ and λ light chains. Where a serum shows monoclonal light chains without a corresponding α, γ or μ heavy chain test for the delta and epsilon heavy chains should be done.

Occasionally some IgA or IgD paraproteins do not react with light chain antiserum due to their conformational arrangement. Also some large aggregated IgM paraproteins may precipitate into the gel on application and do not separate electrophoretically; these samples show a reaction in all antiserum lanes. In both these situations the true identity can be revealed by mild sulphydryl reduction using dithiothreitol prior to electrophoresis and immunofixation. The immunofixation is very flexible and by using appropriate antiserum, a variety of other proteins may be identified. The most likely proteins to exclude when investigating samples for monoclonal gammopathies are fibrinogen, C reactive protein, β₂-microglobulin and complement components.

In recent years, the suggested follow-up protocol for patients post haematopoietic stem cell transplantation for myeloma has included the detection of monoclonal immunoglobulin to establish whether the paraprotein has disappeared and to detect minimal residual disease. It is recommended that the samples be immunofixed even if there is no paraprotein band visible on the electrophoretic separation. Immunofixation is more sensitive and specific than electrophoresis for the identification of paraproteins in these circumstances and has been shown to have comparable sensitivity to the detection of monoclonal immunoglobulin gene rearrangement[6].

Immunofixation for Bence Jones protein is done with the same antibodies used in the serum system. The anti-lambda and anti-kappa antibodies are unable to distinguish between free light chains and those incorporated into an immunoglobulin molecule but the presence of free light chains is assumed if no heavy chain reaction is present. Antibodies reacting only with free light chains are commercially available but these are generally of low titre and react poorly in immunofixation systems.

The major difficulty in the interpretation of urine electrophoresis and immunofixation patterns is the distinction between monoclonal light chains and light chain fragments generated from normal immunoglobulin catabolism. The light chain fragments typically show as a "ladder-like" pattern, particularly in the immunofixation reaction with antiserum to kappa light chains^[7,8].

Paraprotein identification in CZE

The Immunofixation technique, where antigen-antibody complexes are precipitated into a gel matrix, is not applicable to the automated CZE instruments where the immunosubtraction method is used. In this, the sample is pre-reacted with sepharose beads coated with antiserum to the various heavy and light chains. Large macromolecular antigen-antibody complexes form on the coated beads and these sediment leaving a supernatant that should not contain the antigen corresponding to the antibody on the bead. The supernatants are aspirated into the system and electrophoretic separations are generated; one showing the pattern after IgG has been 'subtracted', after IgA has been 'subtracted' and so on. The interpretation of this technique relies on the disappearance of a whole peak, part of a zone or a change in shape of a zone.

Unlike immunofixation, immunosubtraction does not amplify the signal that is read and it is entirely dependent upon a paraprotein band being clearly visible in order for its disappearance to be noticeable. Immunosubtraction does work well for the typing of large paraproteins superimposed on a gamma zone of reduced intensity. However, in our experience, it is not satisfactory for typing small paraproteins particularly with gamma zones of normal intensity and for distinguishing polyclonal increases of the gamma where one of the light chain types predominates. To date, immunosubtraction methods show neither the flexibility nor the sensitivity of immunofixation.

Quantification of serum paraproteins

Immunochemical quantification of paraproteins is unreliable^[9,10]. The densitometric scan of gel electrophoretic separation is recommended for measurement of paraprotein concentration. It is important to be aware that there is a differential dye binding between albumin and the globulins, therefore the most precise estimation of paraprotein is derived from the percentage of relative dye-binding of the paraprotein band compared with the total globulin fraction rather than the total protein. It is also important to note that there is a non-linear relationship between dye-binding and protein concentration at high paraprotein concentrations. The CZE system directly reads the protein concentration at the 'exit' end of the capillaries by UV absorption and therefore in not influenced by the problems related to the staining stage of the electrophoretic technique.

Measurement of serum total protein and albumin are generally reliable and it can be useful to use these two concentrations as a 'rough check' of the paraprotein quantitation. The albumin concentration added to the paraprotein concentration cannot exceed the total protein concentration and (accepting that there may be differential albumin to globulin binding) bands of similar areas should ultimately be of similar concentrations.

Quantification of Bence-Jones protein

Quantitation of BJP is being recommended as a criterion for response, progression of relapse of multiple myeloma treated by high-dose therapy and stem cell transplantation^[6]. We recommend that this is done, like serum paraprotein quantitation, by densitometry of the urine electrophoresis and calculation of the paraprotein band with respect the urine total protein (either random or 24 hour).

Conclusion

There can be no doubt that testing for Bence Jones protein has 'stood the test of time'! Our techniques have improved considerably over the last 150 or-so years and now paraprotein analysis is widely available. It is important that those laboratories that offer paraprotein analysis are aware of the most appropriate way of investigating samples and of interpreting the results of paraproteins investigations. The subjective nature of the tests results in many potential sources of error or confusion. These cannot be minimised by replacing technical skill and scientific knowledge and experience with increasingly sophisticated technology and automation.

MCQ's

Answer TRUE or FALSE to the following statements (double-click inside the square brackets to reveal answer):

Concerning the immunoglobulins:

1. Immunoglobulins are the product of B lymphocytes [TRUE ]
2. In secretions, IgA is typically a pentamer [FALSE]
3. The constant region of the immunoglobulin molecule binds to complement components and macrophages [TRUE ]
4. In normal agarose gel electrophoresis, IgG, IgA and IgM are seen as separate distinct zones [FALSE]
5. The 5 immunoglobulin heavy chains are α, β, γ, δ and ε [FALSE]

Concerning the investigation of paraproteins in serum:

1. Paraprotein bands can appear anywhere on the electrophoretic separation from the alpha-zone to the post gamma region [TRUE ]
2. Small paraprotein bands run in the beta region of electrophoresis while larger bands run in the gamma region [FALSE]
3. Paraproteinaemia can be excluded if the total immunoglobulin concentration is normal [FALSE]
4. C-reactive protein, at high concentrations, may appear as a band in the alpha-2 region of the normal electrophoretic separation [FALSE]
5. Samples for the investigation of cryoproteins must be collected, transported and separated at 37^oC [TRUE ]

Concerning the investigation of paraproteins in urine:

1. Urine electrophoresis should be capable of detecting Bence Jones protein at a concentration of 10mg/l [TRUE ]
2. A trace of albumin should be visible on the electrophoretic separation of all urine samples [TRUE ]
3. Bence Jones protein can be distinguished from intact monoclonal immunoglobulin by its electrophoretic appearance [FALSE]
4. High concentrations of Beta-2 microglobulin in the urine will show as a sharp band on electrophoresis [TRUE ]
5. Marked glomerular proteinuria seen in light chain amyloid will obscure low concentrations of Bence Jones protein [TRUE ]

Concerning the identification of paraproteins:

1. Immunofixation is the method of choice for detecting paraproteins [FALSE]
2. In myeloma patients post stem cell transplantation, disappearance of the paraprotein should be confirmed by immunofixation [TRUE ]
3. In samples that do not react with IgG, IgA and IgM antiserum and react with only light chain antiserum it is essential to exclude an IgD or an IgE paraprotein [TRUE ]
4. Sulphydryl reduction prior to immunofixation is recommended for samples showing polyclonal reactions with antiserum to IgG, IgA, IgM, kappa and lambda. [FALSE]
5. Antiserum to free light chains is recommended for the detection of monoclonal free light chains (Bence Jones Protein) [FALSE]

References

While the advice and information on this website are believed to be true and accurate at the time of publication, neither the authors, nor the UKMF accept any legal responsibility for their content.