The immunoglobulins: AreviewThe immunoglobulins: A review which individual chains are composed has...

J. clin. Path. (1969), 22, 117-131

The immunoglobulins: A reviewN. H. MARTIN

From St George's Hospital, London

In 1959 Heremans proposed the embracing term'immunoglobulins' for those globulins primarilyassociated with the lymphoreticular system. Theterm 'lymphoreticular system' includes specific andnonspecific proliferations of the plasma cell series.This review deals with disease characterized by aprimary proliferation of one of the cellular elementsof the lymphoreticular system, in which there isassociated an excessive production of a specific im-munoglobulin or fragment of a specific immuno-globulin and its release into the vascular space. Theclassical examples are myelomatosis and essentialmacroglobulinaemia. The normal immunologicalmechanisms of defence are frequently associated withproduction and release of immunoglobulins, but itis not the intention of this review to cover this muchwider field. Nevertheless, unusual antigenic stimuli,or the pattern of response of an 'unsalted' popu-lation can produce a bizarre immunological responsewhich may be confused by the unwary with theprimary lymphoreticular dysplasias.

Porter's elucidation (1962) of the structure of yG,with the dissection of the molecule into a pair oflight chains and a pair of heavy chains, enabledthose interested to look with much greater precisionat the globulins produced in excess in the primarylymphoreticular dyscrasias, particularly mye-lomatosis. Thus, it was soon evident that 'BenceJones protein' was in fact free light chain or a pro-tein very like it. It was also clear that whereas yGglobulin extracted from normal serum consisted ofa variety of species, the composition of whose lightand heavy chains varied one from another in theirsequential pattern, in myelomatosis the species pro-duced was restricted. It was on this basis that theconcept developed that myelomata represented theproliferation of a single clone of cells producing asingle species of protein. If, and it was clear thatthey did, these malignant clones displaced the normalcell species, the disease myelomatosis would beassociated with varying degrees of immunologicalparalysis. This indeed turned out to be the case.The macroglobulinaemias are still in some state of

confusion. The structures of the large molecule yMis being clarified and it is clear that the essentialdifference from its smaller cousin, yG, lies in the

detailed structure of its heavy chain. The light chainsare to all intents and purposes identical.Once an understanding of the structure of the

molecules had been acquired, workers were able toforecast further protein dyscrasias such as Fc and,u monomer disease on a rational basis, akin toFisher's forecasting of genetic defects using hisstatistical skills.More recently the use of immunochemical tech-

niques, particularly the Mancini application ofimmunodiffusion to quantitative assay, has con-tributed to further advances in the identification ofpatients suffering from myelomatosis and alliedconditions. Recent series have regularly includedsubstantial groups of patients whose serum proteinpattern suggested a monoclonal dyscrasia, butwhose clinical picture did not tally with the classicalpicture of myelomatosis (Osserman and Takatsuki,1964; Bachmann, 1965; Hobbs, 1967a). Notable isa detailed analysis by Hallen (1966) of 150 suchpatients.

THE NOMENCLATURE OF THE IMMUNOGLOBULINS

A memorandum issued by WHO (1964) on thenomenclature of immunoglobulins made certainbasic suggestions which have been universallyadopted.

Individual chains should be referred to by Greekletters. Thus, the two principal groups of light chainsare referred to as K and A. The five groups of heavychains so far identified are referred to as y, a, ,t, 8,and E. The whole molecule is defined by a capitalletter G, A, M, D or E, supplemented by a capitalK or L where the nature of the light chain has beenidentified.The whole molecule may be referred to in one of

two ways thus:yG or IgG, subgroups yGK yGL

or IgGK IgGLyM or IgM, subgroups yMK yML

or IgMK IgMLThe molecular formula of the molecule is pre-

sented in Greek lettering thus: yAK or IgAK iswritten a%K2, implying that the molecule IgA is madeup of two cx heavy chains and two K light chains.

117

on May 22, 2020 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.22.2.117 on 1 March 1969. D

ownloaded from

http://jcp.bmj.com/

N. H. Martin

Similarly Igm will be written Q"K2)5 or (P25,depending on which light chain is incorporated inthe molecule and the assumption that the macro-globulin circulating in the serum is a pentamer.As genetic loci are identified on individual chains

these may be appended as a suffix to the individualnotation in the molecular formulae.yG or IgG replaces the old terminology y, 7Sy, Y2,y52;yA or IgA replaces theold terminologyfl2A,y1A;yM or IgM replaces the old terminology yjM, f2M.

CHEMICAL CHARACTERISTICS OF THEIMMUNOGLOBULINS

IgG Normal serum concentrations are 800 to 1,600mg/100 ml, and the rate of synthesis is 20 to 40mg/kg/day.

In 1962 Porter made the tentative suggestion thatIgG consisted of four peptide chains, two large andtwo small, linked together by disulphide bridges.The smaller chains are of two main groups, K and A,and have a molecular weight of the order of 20,000,while the larger chains have a molecular weight ofthe order of 50,000. In general, one disulphide bondlinks each heavy chain to its respective light chain,the linkage being at or near the carboxy terminalgrouping of the light chain. The heavy chains arelinked to each other by a disulphide bond which lieson the carboxy terminal side of the point of cleavageof the whole molecule by the enzyme papain.

If we consider the breakdown of the molecule bypapain, it appears to consist of three fragments: thetwo Fab fragnents, which are made up of the aminoterminal half of the heavy chain still attached to thelight chain, and one fragment, the Fc fragnent,which consists of the residual carboxy terminal por-tions of the two heavy chains linked together by adisulphide bond. The amino terminal portion of theheavy chain is referred to in isolation as the Fdfragment.

THE FAB FRAGMENT Enzymic degradation of specificantibodies shows that the antibody activity of aparticular globulin is concentrated in the Fab frag-ment. It was for this reason that the symbol wasadopted. The actual site of activity is no more than3 to 4% of the total Fab fragment, that is, approxi-mately 10 to 12 amino acid residues. Full activityrequires the association of a portion of the Fd frag-ment and the light chain operating together. Eachantibody molecule of the IgG series contains twocombining sites which correspond neatly to the twoFab fragments known to exist in the individual IgGmolecules.

The Fc fragment This fragment represents the

Aminoterminal

Carboxy sterminal Fd

sinTh~% % Heavy chain-Specificca&S or p

;., ~Light chain -Common toFc Fab k or X all molecules

FIG. la. A simplified basic structural formula of theimmunoglobulins.

IgM pentamer IgM monomer

tS-S fS-Sr

t , fragment

FIG. l b. The pentameric formula of the macroglobulinIgM, together with the fragments occasionally identified inthe blood stream of patients suffering from macroglo-bulinaemia.

'exposed' portion of the heavy chain on the diagram.It is readily crystallized (hence the adoption of thesymbol Fc). It does not carry the antibody activityof the molecule. It is the fragment to which the car-bohydrate is attached.

It is the Fc fragment that contains the sequentialarrangement responsible for the cytophilic propertiesof the intact IgG molecule. It is as if the carboxyterminal portion of the whole molecule acted as ananchor where anchorage is necessary. Sequentialanalysis suggests that the pattern of this anchordiffers very little from one mammal to the next.Though the Fc fragment contains no antibodyactivity as such, it has other important biologicalactivities such as the capacity to fix complement andto combine with the rheumatoid factor.

The light chains Reference has already beenmade to the 'light chains' as an integral portion ofthe Fab fragment responsible for antibody activity.These are now classified into two main serologicalgroups, the K and the A chains, and the completesequential analysis of the 143 amino acid residues of

118



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/

The immunoglobulins: A review

which individual chains are composed has beencompleted (Milstein, 1966a and b). Seventy per centof normal human immunoglobulins carry the Kchain, while 30% carry the A chain. It has beenclaimed that the IgD globulins are predominantly Aand the cold agglutinins predominantly K chaincarriers.When the light chains separated from immuno-

globulin IgG are examined by starch gel electro-phoresis 10 components are identifiable (Cohen,1966), and even the individual chains separated bythis technique may be further differentiated by theirantigenic behaviour. This antigenic differentiationappears to be under genetic control and is identi-fiable on the chain as the Inv. locus. Within the twomain groups of K and A light chains there is a con-siderable degree of detailed heterogeneity mainlyconcentrated in the amino terminal portion of themolecule. The whole chain appears to fall naturallyinto two sections analogous to the Fd and Fc fractionof the heavy chain.

These light chains are common to all the principalgroups of immunoglobulins. No worker to date hasfound K and A chains side by side on the same mole-cule. It is the characteristics of the heavy chaincontaining the Fd and Fc fragments which differ-entiate IgG, IgA, IgD, and IgE one from the other.

The heavy chains The heavy chain of IgG hasnow been analysed in some detail. There are 34prolines in a total of 422 amino acid residues. Theseproline residues are not uniformly distributed.Thus, in the first 84 amino acid residues from theamino terminal end there are only four prolineresidues, whereas in the first 29 amino acid residuesin the carboxy terminal fragment adjoining the pointof papain cleavage there are nine proline residues.Structurally this implies that the Fd fragment, whichcontributes to antibody specificity, is a relativelyflexible structure and that the Fc fragment is rela-tively rigid in the area next to the point of papaincleavage.

IgA The normal serum levels lie between 140 and420 mg/100 ml, and the normal rate of synthesisbetween 3 and 50 mg/kg/day.The monomer of normal IgA has a molecular

weight of the order of 160,000, but it may sediment ina heterogeneous fashion producing components of7S, llS, 13S, and on occasion, 15S.

In the myelomas associated with excess pro-duction of IgA, ultracentrifugal analysis may showsimilar heterogeneity (Kekwick, 1940), or a singleunit with a sedimentation coefficient of about 9(patient C in a series studied by McConnell andMartin, 1958).The light chains of the two molecules IgG and

IgA are similar, but the heavy chains differ both intheir antigenic behaviour and in their carbohydratecontent, the a chains of IgA containing four timesas much hexose as the y chain of IgG, being muchricher in neuraminic acid.Clamp, Dawson, and Hough (1966) have analysed

the a chain in some detail and found evidence tosuggest that the carbohydrate is not concentrated inone area of the chain but that there are severalpoints of attachment.IgA is of special biological interest because it is

the principal protein in human colostrum; it isexcreted in the saliva, in the lachrymal glands, andin the mucous glands of the intestine.

In its exocrine form it is attached to an immuno-chemically identifiable polypeptide, the 'secretorypiece' of Hanson and Johansson (1967), referred toby Tomasi, Tan, Solomon, and Prendergast (1965)as the 'transport piece'.A comparison of specific purified IgA (exocrine)

and normal IgA (serum) from the data of Tomasiand Hanson show that the S coefficient is of theorder of 11 compared with 7. Treatment with fmercapthanol does not alter the S value of eitherIyAex or IgAS as compared with IyM.The total carbohydrate is somewhat higher in

exocrine IgA than serum IgA, but the neuraminicacid content is 0-1 to 10% in the excretion com-pared with 1-7 to 1-8% in the serum IgA. This is inkeeping with the slightly slower electrophoreticmobility.The IgG/IgA ratio in saliva is less than 1 com-

pared with an IgG/IgA ratio O/over 6 in serum.Immunofluorescent studies using specific anti a

chain sera indicate that the IgA immunoglobulinsare concentrated in the plasma cells of the laminapropria of the stomach, duodenum, jejunum, colon,appendix, and rectum (Crabbe and Heremans, 1966).When antiserum to the 'secretory piece' is used it

is noted that the 'exocrine' form of IgA appears tooccur principally in the goblet cells excreting mucus.Such antisera did not produce fluorescence in thesubepithelial plasma cells in contrast to a prepara-tion using ax chain antiserum.

In newborn infants, unlike adults, there are few, ifany, subepithelial plasma cells, though the infantshave normal quantities of 'secretory piece' in theirexternal secretions. The evidence suggests, therefore,that the secretory piece is manufactured separatelyand subsequently coupled to IgA.

igm The normal serum concentration ranges be-tween 50 and 190 mg/100 ml, and the rate of syn-thesis lies between three and 17 mg/kg/day.The IgM immunoglobulin was first identified by

ultracentrifugal analysis. It stands out as a well

119



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/

N. H. Martin

defined peak with a sedimentation velocity ofapprox-imately 19 compared with 7 for normal IgG. Adetailed analysis of the ultracentrifugal behaviourof the serum protein of normal persons shows thatthe 19S component is composed of two proteins, ancx globulin and a y globulin. These are almost cer-tainly different in their detailed architecture andbiological function. The macroglobulinaemia ofWaldenstrom is regularly associated with increasesin the y macroglobulin. Estimates of the molecularweight of this molecule range from 890,000 to 970,000.It is a pentamer consisting of five monomers linkedby covalent bonds. These are probably :S:S: bonds.Estimates of the monomer size range from 160,000to 170,000. The light chains are similar to, if notidentical with, those in IgG. The heavy chains havea molecular weight of between 65,000 and 67,000and are some five times richer in carbohydrate andin neuraminic acid content than the heavy chains ofIgG (Martin, 1968).

Detailed antigenic studies (Seligman and Mihaesco,1967) indicated that the IgM molecules are as diversein their individual characters as IgG molecules.Recalculation of physicochemical data (Martin,1968), as a result of a critical essay by Creeth andKnight (1965) on the assessment of macromolecularsymmetry, indicates that IgM is not markedly moreasymmetrical than IgG. These calculations, takentogether with the electron microscope studies ofSvehag, Chesebero, and Bloth (1967), suggest thatof the various configurations discussed by Millerand Metzger (1965 and 1966) the 'rosette' is themost likely. Three theoretical possibilities arise fromthe acceptance of such a structure: first, that up tohalf of the 10 antibody sites may normally be buriedand unavailable within the rosette; second, that acondition exists, 'monomer' disease, in which theerror is in the final condensation of the individualmonomers to form the pentamer; third, that the 'tails'of the rosette may be synthesized independently asfragments analogous to the Fc fragments of yGdescribed by Franklin, Lowenstein, Bigelow, andMeltzer (1964). If these do exist in vivo in serumthey should be demonstrable as having immuno-logical characteristics related to the ,u chain, but noreaction to K or A sera.

igD The normal serum concentrations lie between0'5 and 40 mg/100 ml and the normal rate of syn-thesis is between 0 03 and 1-4 mg/kg/day.Immunoglobulin D has a sedimentation constant

similar to IgG, 6-5 to 7-0. It contains two light and twoheavy chains. Its site in the electrophoretic analysisof the serum proteins suggests that it has a slightlyhigher mobility than IgG. Moreover, Rowe andFahey (1965), carrying out degradation studies with

papain, noted that the Fc fragment had a highermobility than that of IgG, indicating that it may bericher in sialic acid though no figures are given.The light chains resemble those in IgG and are

predominantly of the A species. The heavy chains areantigenically distinct. Hobbs, Slot, Campbell, Clein,Scott, Crowther, and Swan (1966) have describedsix patients suffering from myelomatosis in whosesera the myeloma cell protein reacted with IgDantisera. The clinical picture was not significantlydifferent from patients suffering from IgG. But inonly one patient was the total serum protein levelraised significantly. Since the normal levels of IgDare only 0 5 to 40 mg/100 ml, it is evident that pro-duction could be enhanced five to tenfold withoutgross disturbance of the protein pattern. BenceJones protein was being excreted in large amountsin three of the patients described by Hobbs et a!(1966) and was detectable in the urine of all followingconcentrations. Its very low normal serum concen-tration and its relatively high metabolic turnoversuggest that unequivocally abnormal serum levelsmay be attained without demonstrable gross dis-tortion of the serum protein pattern by the usualtechniques employed in routine laboratories.

IgE The normal serum concentrations lie between10 and 140 nanogram/100 ml. The synthesis rateis not known.

In 1968 Johansson and Bennich described a newprotein in the serum of a patient suffering frommyelomatosis which they designated IgND. Itappears that this protein is of the family normallydesignated IgE. By reason of its dimensions and itscarbohydrate content it may be grouped with IgA,though it does not appear to have the same tendencyto form polymers.

Its sedimentation value was 7-29. The total car-bohydrate content was 10-71 and the neuraminicacid content about 1 %. According to Bennich andJohansson (1967) the carbohydrate is attached toat least six prosthetic groups in the heavy chain.One carbohydrate pairing lies to the N terminal sideof the point of papain cleavage that is in the Fabfragment. The Fd fragment is rich in methionine.Johansson, Bennich, and Wide (1968) showed thatthe Fc fragment of IgE is twice the size of theFc fragment of IgD.

THE PHYSIOLOGY OF THE IMMUNOGLOBULINS

The immunoglobulins are produced in the cells ofthe lymphoreticular system. In the normal ebb andflow of response to antigenic stimulus, their pro-duction and release into the vascular space is asso-ciated with an increase in the plasma cell population

120



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/


of the bone marrow and/or the lymphatic glands. Itis strikingly illustrated in the hypergammaglobu-linaemias associated with chronic progressivehepatitis and may be observed in a variety of chronicinfections, in diffuse lupus, rheumatoid arthritis,sarcoidosis, chronic beryllium poisoning, certain'hypersensitivity diseases' such as serum sickness,and acquired haemolytic anaemia. Analysis of thistype of immunoglobulin production indicates thatthough there is an increase in specific antibodyproduced, there is also associated a generalizedproduction of IgG giving a diffuse band on electro-phoresis and a demonstrable increase in the 7Scomponent on ultracentrifugal analysis.

Proliferation of other cells of the lymphoreticularseries, such as the reticulum cell and the precursorsof the lymphocyte are sometimes, but not regularly,associated with extracellular release of excessivequantities of immunoglobulins. This is the situationin the primary macroglobulinaemia of Waldenstromwhere an excessive amount of IgM released isassociated with a proliferation of cells resemblinglymphocytes, and in the 'heavy chain' syndrome ofFranklin in which excessive production and releaseinto the vascular space of Fc fragments is oftenassociated with lymphomata, and, in certain in-stances, enlarged glands having a histological picturesuggestive of Hodgkin's disease (Ellman and Bloch,1968).In the normal adult, 50% of the free IgG is con-

centrated in the vascular space, about 5% of thisbeing catabolized daily.IgA is distributed in approximately the same

proportions, but the loss from the vascular space isabout three times greater than that of IgG. IgA ispeculiar in that, coupled to a polypeptide the 'secre-tory piece', it is excreted against the concentrationgradient into the saliva, bronchial secretions, andintestinal juice. In these it exists in much higherconcentrations than it does in the blood.IgM is localized almost entirely in the vascular

space, its rate of loss from that space being aboutthe same as that of IgA. Humphreys (see Holborrow,1967) has shown that IgM is 100 times more efficientat lysing red cells than IgG. It is a particularlyefficient agglutinator, suggesting that it is speciallyadapted to handling bacteria and other particulate

antigens. IgG may be more effective in neutralizingantigen in solutions at a molecular level. Table Isummarizes these findings.The work of Fink, Miller, Dorward, and Lo-

Spalluto (1962) suggests that the primary responseto extrinsic stimulation is associated with an increasein IgM globulins.

Swedlund, Gleich, and Chodirker (1968), investi-gating a patient with dysgammaglobulinaemia inwhose serum there was a deficiency of both IgG andIgA, showed that the IgM response to bacterio-phage a x 174 was both marked and prolonged,lasting 13 days. In the normal adult the IgM responsewould have been short lived and overshadowed bothin extent and duration by the IgG response. Thus, itis clear that the stimulus responsible for activeproduction of the individual immunoglobulins maybe independent and complementary.

SPECIFIC GLOBULIN DYSCRASIAS

Gutman (1948) coined the phrase 'M protein' forthe discrete excess of protein strikingly demonstratedby classical electrophoresis in the sera of patientssuffering from myelomatosis. Riva (1957) used theterm for a discrete, tightly packed band seen whenserum proteins are analysed by electrophoresis ona discontinuous medium. Vast numbers of patientsshowing this serum protein anomaly have beenexamined in different parts of the world. While it isevident that the majority of such patients are suffer-ing from myelomatosis and macroglobulinaemia oran allied lymphoreticular dyscrasia, a significantpercentage are not. Indeed, by the relatively crudetechniques used, 'M' bonds could be observed in apercentage of patients with malignant growths suchas prostatic carcinoma. Quantitative analysis indi-cates that the concentrations are rarely as high asthose commonly seen in myelomatosis. The proteinobserved may resemble IgG, IgA, or IgM immuno-logically.More refined techniques have indicated that these

sera show multiple banding with one zone pre-dominating, as if there were polyclonal stimulation,but that one clone was more sensitive to the specificstimulant than others.The term 'paraprotein' has been used by some

TABLE IDISTRIBUTION, SERUM CONCENTRATION, RATES OF SYNTHESIS AND CATABOLISM, AND APPROXIMATE TOTAL OF INDIVIDUAL

GLOBULINS

IgA IgD IgG IgMDistribution (5% in IV pool)Serum concentrationRate of synthesis (mg/kg/day)Catabolic rate (Y. of IV pool)Approximate total globulin (g/kg in 70 kiloman)

40140-420207-5514-340 1-0 25

63-860 3-4-00-3-1*4918-60

0-01-0-05

48-62800-186020-404-7

0 75-19

65-10050-1903-2-16-914-25

0-05-0-19

121



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/

N. H. Martin

workers to refer to proteins which do not occurnormally in the serum. There seem to be few syn-dromes in which such a claim can be made withconfidence. It would seem reasonable to abandonthe use of the term and, as exemplified by Franklin'sdescription of heavy chain disease, to refer specifi-cally to any rogue protein as and when it is identifiedprecisely, together with the clinical circumstances inwhich it occurs.

CLINICAL FEATURES OF MONOCLONAL PROLIFERATION

In the following sections the clinical and clinico-pathological features of monoclonal proliferationare discussed against the background of experienceof 225 patients investigated in the past 15 years,because clinical colleagues suspected for one reasonor another that the primary disease process fromwhich they suffered might be associated with anunregulated production of one or other of theimmunoglobulins. Thus, the information obtainedis from a preselected group of the total population.Moreover, because the majority, though not all, ofour collaborators have been consultants themselves,there has been in a sense a double process of selec-tion. This experience is tempered by examination ofdata from equivalent series published in the litera-ture, though it seems clear that the majority of suchseries are also the subject of selection if only thatthey are patients in hospital clinics. A notable ex-ception is the Varmland survey by Axelsson, Bach-mann, and Hallen (1966). In this survey electro-phoretic analysis of the serum proteins was made on6,995 persons representing 70% of the total popu-lation above the age of 25. 'M' component wasidentified in 64 of these, but in only 10 was the levelabove 100 mg/100 ml serum. Immunochemicalanalysis indicated that in 61 % the discrete proteinexcess was IgG, in 270% IgA, and in 80% IgM, buttwo of the persons showed excess of both IgG andIgM and of IgG and IgA. Fifty-nine of the 64persons were examined clinically and 36 were foundto have an abnormality which did not impairactivity. Seven of the 64 persons were over 80.

Detailed investigation of the immunoglobulinsisolated from patients suffering from myelomatosis

TABI

and allied conditions suggest a homogeneity of thecirculating immunoglobulin that is not presentwhen the immunoglobulins from normal healthyindividuals are examined by similar techniques.Instead of the 10 or so bands demonstrable on starchgel analysis of the separated light chains, only oneor perhaps two are seen. A similar homogeneity ispresent in the heavy chains though these are techni-cally more difficult to handle. This observation,coupled with the assumption that an individualimmunocyte produces only one type of immuno-globulin, has led to the concept of monoclonalproliferation associated with the massive productionof immunoglobulins of limited structural range. Theprincipal conditions in which this peculiar type ofproliferation occurs are the myelomatoses, the essen-tial macroglobulinaemias of Waldenstrom, theso-called idiopathic hyperglobulinaemias, and theH chain disease of Franklin. Occasionally, anindividual immunoglobulin is produced in excesscomplicating lymphomata and even totally unrelatedneoplasms such as bronchial carcinoma or carcinomaof the breast, prostate, and stomach (Table II).

In the future, to the breakdown shown in TableIII will be added patients in whom the monoclonaldisturbance is confined to IgD, to IgE, and to otheridentified immunoglobulins of specific character.IgD and IgE are present in normal serum in suchsmall quantities that gross disturbances in their

TABLE IIPERCENTAGE DISTRIBUTION OF MONOCLONAL PROTEINSFROM THREE SOURCES TOTALLING OVER 1,000 PATIENTS

No. of Cases

Multiple myelomaMacroglobulinaemiaHeavy chain diseaseLymphoma and lymphatic leukaemiaUnrelated neoplasiaIdiopathic

400165 510 250 755.757.7510 0

45626712

646

22536515

874

5Osserman and Takatsuki (1964)2Hobbs, J. R. (1967). Brit. nied. J., 2, 699.3Martin (unpublished series)Hobbs also refers to 'other diseases, liver gut etc-' 4% and 'tran-sient'. We have 10 cases unplaced, including fouLr 'transient' cases.

LE IIIBREAKDOWN ACCORDING TO SPECIFIC IMMUNOGLOBULIN ANALYSIS OF 500 CASES

Prinicipal Deviatiotn in Serum ( %) Proportiotn of Coluonn' Bence Jones K ( %)2Protein in Urine ( %)

lgG 68 (47) 45 (15)IgA 23 (20) 50 6)IgM 4 (14) 10 ( 1)No gross change in serum 5 (20) 100 (20)'Data from series of Osserman and Takatsuki (1964).2Data on K and A chains from series of Imhof, BallieLlx, Mul, and Poen (1966).

,A (e /)2

39394043

61616057

122



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/


level may occur without disturbing the total serumprotein electrophoretic pattern. Such cases un-doubtedly account for some of the patientsdescribed in whom Bence Jones protein wasdemonstrable in the urine without gross evidenceof disturbance of the serum protein distribution.This is exemplified in three out of the six patientswith IgD myeloma described by Hobbs et al (1966).

THE COMMON CLINICAL STIGMATA ASSOCIATED WITHMONOCLONAL PROLIFERATION

These are Bence Jones proteinuria; recurrent in-fection; renal damage associated with proteinuria;anaemia, (a)normocyticnormochromic with a raisedESR, and (b) leucoerythroblastic anaemia, usuallya terminal event resulting from extensive marrowinvasion; space-occupying lesions, which may beskeletal localized vertebral collapse, spontaneousfractures, diffuse generalized osteoporosis, withserum alkaline phosphatase levels normal in 90%of patients, or visceral with enlarged lymph glandsor spleen; cryoglobulinaemia, Raynaud's pheno-mena; hypercalcaemia; amyloidosis; and serumhyperviscosity.

BENCE JONES PROTEINURIA The increasingly commonpractice of concentrating fresh urine specimens bypressure dialysis and other means before subjectingthem to electrophoresis and immunological in-spection has given a fresh view on the incidence offree light chains passed in the urine.

Patients with Bence Jones proteinuria-if by thatwe imply those in whom the peculiar thermol-ability of the protein first described by Bence Jonesis demonstrable in the fresh, unconcentrated speci-men of urine-number between 40 and 60% of atotal series (Kubota, Schwartz, and Putman, 1956;Martin, 1961) when four consecutive specimens areexamined following the initial diagnosis. Sur-prisingly it is only demonstrable in 10% of patientssuffering from primary Waldenstrom macro-

globulinaemia.Berggard and Edelman (1963), examining con-

centrates of pooled normal human serum, demon-strated the presence of proteins having thermolabileproperties similar to Bence Jones protein. Physico-chemical investigation suggested that these were infact light chains either the result of immunoglobulindegradation, or, as in myeloma, the result of over-

production and extracellular release of uncoupledlight chains. Martin (1961) emphasized that repeatedexaminations of urine were required before theabsence of Bence Jones protein might be assuredeven by old-fashioned techniques. Using modemnmethods of concentration, dialysis, and immuno-

chemical techniques of identification, it is probablethat light chains will be found in every urine andthat the differentiation between normal and ab-normal persons depends on the amount passed andthe characteristics of the electrophoretic analysis ofthe urinary proteins (Dominico and Waldenstrom,1968). The light chains excreted from normal urineor from polyclonal hypergammaglobulinaemiaswill present as a relatively diffuse band or as a seriesof bands, those from the monoclonal conditions asa discrete tightly packed band.Hobbs (1967b), using a relatively simple technique

for the concentration of urine, found light chainconcentrations above 1 mg/100 ml in 172 out of 223patients examined. He emphasizes that had theurines not been concentrated, Bence Jones proteinwould have been missed in 90 patients. His resultsfor the group of patients with excess of IgM in theserum show that by refined techniques 32 out of 34patients were excreting light chains, though in 26of these the urinary concentration rate was less than20 mg for 100 ml urine.

Six patients with IgD myeloma (Hobbs et al,1966) are interesting because in three the grossserum concentration of IgD and IgG immuno-globulins totalled 920, 660, and 920 mg/100 ml,that is, well within normal limits of immunoglobulinconcentration. Yet IgD levels were six times higherthan normal, and, since IgD has a much shorterhalf life than IgG, it implies a considerable increasein the specific plasma cell population. Two of thepatients in the series were producing profuseamounts of Bence Jones protein. It may well be thatsome of the cases reported in the earlier literatureas having a normal serum protein pattern with grossBence Jones proteinuria were, in fact, cases of IgDmyeloma. If they were, the striking 'flaming appear-ance' in the cytoplasm of the plasma cells seen onmarrow sampling may give a hint of the true diag-nosis.

RECURRENT INFECTION Boe (1945) described twopatients having classical myelomatosis who hadmultiple pneumonic episodes. Marks (1953) madea detailed study of antibody formation in 17 patientssuffering from myelomatosis when challenged withstaphylococcal antitoxin and streptolysin 0 (Fig. 2).These were compared with the response of 112normal persons and show a clear-cut diminution incapacity to produce antibody. Other patients withhypergammaglobulinaemia, the consequence ofhepatitis and rheumatoid arthritis, showed no suchdiminution in their response. Lawson, Stuart, Paull,Phillips, and Phillips (1955) reported on nine cases,stressed their liability to pyogenic infections, andshowed that all had a constant and characteristic

123



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/

N. H. Martin

40

30a)(n

0~

a) 20ox

M Myelomna,:I Controls II

II I

I I

I I

I I

__I_I

==m

< A0 17 l20 40 801'0trAnti-streptolysin '6' titre

FIG. 2. Plot ofantistreptolysin titre ofmyelomapatients com-pared with normal controls showing the defective reactionofthe patients with myelomatosis. Datafrom Marks (1953).

deficiency in circulating antibody. One of theirpatients had had three attacks of lobar pneumoniain the four months before admission to hospital andanother had seven attacks between 1950 and 1954.These patients in their clinical history were somewhatreminiscent of Bruton's boy (1952) who had a pri-mary globulin deficiency, and it is evident thatmyeloma must be regarded as a form of immuno-logical paralysis.

Porges (1956) reiterated the liability of thesepatients to pyogenic infection, but stressed that thedeformities of the chest wall associated with bonylesions and the anaemia might be significant con-

tributory factors in the recurrence of pulmonarycomplications. This was clearly not the situation inthe patient reported by Baker and Martin in 1959,who had had a pleural effusion in 1951 and was

under surveillance at a chest hospital in 1952, forit was not until 1957 that she developed bone lesions.The proliferation of a single or very limited num-

ber of clones of cells (Ballieux, Imhof, Mul, Zegers,and Stoop, 1968) is always associated with some

degree of immunoparesis though the cell-mediatedimmune response may remain undamaged. Indeed,in 100% of patients with malignant immunopro-liferative diseases recurrent infection is the primarymanifestation. In all these conditions there is some

demonstrable reduction in normal levels of immuno-globulin whether the technique employed be elec-trophoretic (Laurell, 1961) or specifically immuno-logical (Fahey, Scoggins, Utz, and Szwed, 1963;Cwynarski, 1968). Patients with myelomatosis showa diminished response to pneumococcal polysac-charide and to typhoid and to influenza vaccines.Fahey et al (1963) stressed the difference in primaryand secondary response in three patients, while Cone

and Uhr (1964) showed that only 50% of theirpatients could be sensitized by dintrofluorobenzene.

In their detailed study of the pattern of recurrenceof infection in 46 patients with myelomatosis ormacroglobulinaemia, Fahey et al (1963) indicatedthat pneumonia and pyelonephritis were by far thecommonest areas of infection noted and that theincidence of infection per patient per month wassignificantly higher in the patients with myeloma-tosis than in those with macroglobulinaemia. Sixof the patients suffering from macroglobulinaemiawere in relatively good health despite impairedantibody response.

Miller (1962) has studied the pattern of immuno-logical deficiency in the lymphomas and leukaemias.Of 61 patients suffering from chronic lymphaticleukaemia, 21 had associated hypogammaglobulin-aemia. It is not surprising, therefore, that thereshould be a high incidence of infection in thesepatients. Patients with 'Hodgkin's disease', takingthat group to embrace the granulomata, the para-granulomata, and the sarcomata, were not so proneto bacterial infection, 1250% against 37.0o% of thecases with chronic lymphatic leukaemia. Delayedhypersensitivity as indicated by skin tests showed amarked difference between the two groups: 80600of patients with chronic lymphatic leukaemia re-sponded as compared with 25 % of the patients withHodgkin's disease. Cone and Uhr (1964) have com-pared the immunological response of small groupsof patients suffering from chronic lymphatic leu-kaemia and myeloma in some detail. Both showeddelayed type sensitivity to at least one of a range ofcommonly encountered antigens. But while five outof 11 patients with myeloma gave a positive responseto sensitization with 2-4-denitrofluorobenzene onlyone out of nine of the patients suffering from chroniclymphatic leukaemia did likewise.

RENAL DAMAGE More than 50% of patients haveimpaired renal function; 5% present with acuterenal failure and uraemia is second only to pneu-monia as the ultimate cause of death in this groupof diseases.Compared with other causes of chronic renal

damage, oedema, haematuria, and hypertension arenot common.The primary renal lesions as observed at necropsy

or by renal biopsy lie in the distal segments of thenephron: the tubular epithelium in the area isswollen and degenerate and the lumen is blockedby obstructing casts of protein which extend upwardto involve the entire length of the tubule (Oliver,1944-45).Renal biopsy studies indicate that initially the

glomeruli are not involved and function adequately,

124



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/


Bence Jones protein being filtered through themeffectively (Greenwald, Bronfin, and Auerbach,1953).

Batts (1939), from a review of 40 cases, indicatesthat the severity of renal damage cannot be corre-lated directly with the amount of Bence Jones proteinpassed at any time in the urine.The functional capacity of the kidney must be

affected by a variety of other factors, such as dehy-dration, hypercalcaemia, anaemia, and increasedblood viscosity in individual cases.

Occasionally the renal damage may be the con-sequence of paraamyloid infiltration or diffuseinfiltration with plasma cells.

Figure 3 shows the levels of blood urea in 100consecutive patients on their admission to hospital.

30h

20-cnc0

ar-10

0 20 40 60 80 100over

100

Urea (mg./l0rnil.)

FIG. 3. Levels of blood urea in 100 consecutive patientsadmitted to hospital.

Of the four with blood urea levels higher than 100mg/100 ml, three were admitted in advanced renalfailure, the diagnosis of myelomatosis not havingbeen suspected, and in two of these the monoclonalactivity was producing IgA.

ANAEMIA Some degree of anaemia is present innearly all cases during the course of the disease.Figure 4 shows the Hb levels on admission of 140consecutive patients suffering from IgG and IgAmyelomatosis compared with 45 consecutive patientssuffering from macroglobulinaemia (IgM). Speci-mens of blood taken from the latter patients fre-quently show clumping and rouleaux formation. Theanaemia is usually normocytic and normochromicand frequently refractory to haematinics. The originsof the anaemia are complex, being a combination ofthe invasion of normal marrow, the direct inter-ference in the development and survival of theerythron, and, particularly in the macroglobulin-aemias, persistent intermittent haemorrhage, withinterference in the blood clotting mechanism arising

from interaction of the immunoglobulin with theplatelets and the coagulation factors VI and VII andfibrinogen (Nilehn, 1962; Frick, 1955).Glueck and Hong (1965) have described the

interference with the action of the antihaemophilicglobulin by an IgA globulin from a patient withmyelomatosis.

307M

20-

10

4 6 8 lb 12 1'4 1'6, 0 Hb(g. 100m1.)

10 _

20 -

-)A & YG

30r_

FIG. 4. The haemoglobin levels ofpatients suffering frommacroglobulinaemia at first examination compared withthose suffering from myelomatosis at first examination.

SPACE-OCCUPYING LESIONS These are as a ruleosterlytic but in a few cases may be visceral.

Skeletal The characteristic lesions in myelo-matosis are osteolytic and in 90% of patients areassociated with normal serum alkaline phosphataselevels (Fig. 5). The first indication of the lesion maybe vertebral collapse or a spontaneous fracture. Fiveto 10% of patients have a diffuse osteoporosis with-out the characteristic localized osteolytic lesions.Discrete skeletal lesions are rare in the primarymacroglobulinaemias.

Visceral lesions In 20% of patients the depositsare primarily visceral in the spleen or a lymph node,though in these marrow puncture almost invariablyshows an excess of plasma cells. The plasma cellleukaemias merge imperceptibly with this group.Many of these patients develop demonstrableskeletal lesions as the disease progresses, though theprimary macroglobulinaemias are an obstinateexception.

125



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/

N. H. Martin

40

30

"' 20(C

it0

O0

___1

50 100 150Alkaline phosphatase (I.U.)

20O

15~

1n

CZ

5

200

FIG. 5. The serlulm alkalitie phosplhatase levels in a series ofpatienits suffrring firom1i nieloniatosis at first examiinlationi.

CRYOGLOBULINAEMIA AND RAYNAUD S PHENOMENA

Cryoglobulinaemia is not pathogenomic of themonoclonal proliferations, for it may be observedin any condition in which there is gross overpro-

duction of immunoglobulins, particularly wN herethere is an associated marked disturbance of thealbumin/globulin ratio.

Because of the physicochemical properties of 1gManid the relatively low temperatures at which theymay precipitate, peripheral circulatory disturbances,sLuch as Raynaud's phenomena, accompanied bysuperficial skin ulceration, may be the presentingsynmptonm in patients with mnacroglobulinaemia. Offive patients we have seen in whorn this was thesituation, one had a markedly positive Wassermannreactioni, and had been receiving treatment forsyphilis (Martin, 196)0).

FI\PERCALCAEMIA Figure 6 shows the calciumlevels in 80 consecutive patients at first examination.A patient admitted after the collection of this

series deserves mention. Admitted from the Out-patient Department complaining of polyuria, herapidly became disorientated and lapsed into coma.

He had marked nitrogen retention (blood urea 110mg/100 ml) and a serum calcium level of 17 6 mg/100ml) Under vigorous treatment with prednisone anddisodium phosphate his serum calcium level droppedto 9-0 mg/100 ml over the next 20 days and his bloodurea to 37 mg/100 ml. There was no history or evi-

dence of any discrete skeletal lesions on radiographs.The diagnosis of myelomatosis of the IgG group was

established from the findings in the serum, the urine,and on examination of the bone marrow. He made a

8 10 12 14 16 186Calcium (nmg./100ml.)

FIG. 6. Calciulmt levels in cl ser-ies of patients sqfleiiricfromn mn1velomahitosis onfirst admiis sion.

good immediate clinical recovery on immuno-suppressive drugs, but died the following year wheithe diagnosis was confirmed.

Lyall Watson (1964). in a study of the effects ofhvdrocortisone levels on hypercalcaemnia in neo-plastic disease, showed that of 12 patients eightshowed a well mar-ked drop in plasnia calcium. Inone the osteol\tic lesions appeared to diminislh inextent.

AMNYLoIDOSIS Amloidosis is niot a commllon cOIlm-plicatioin of the monoclonal diseases. Five to 1()"show paraamyloid deposits. Contrarim ise, anypatient with so-called primary amyloid lesions in theskin or tongue should be examined to exclude thepresence of myelomatosis.

\VISCOlS SERtUM AND THL HYP[R\ISCOSITY SNDROME

Steel (1959) and Falhey, Barth, and Solomoni (1965)have pointed out the striking difference in viscositybetween IgG and IgM, when their relative viscosityis plotted against protein concentration. Frangleii(1968) has pointed out an equally striking differencein the temperature viscosity ratio. Fahey, Barth, andSolomon (1965) have outlined a clinical picture ofrecurrent spontaneous haemorrhage associated witlhsymptoms of impaired circulation, together withocular and neurological manifestations which occurwhen the relative serum viscosity exceeds six. Thesefeatures are most common in essential macro-globulinaemia with excess of circulating IgM, butthey are observed in patients with myelomatosisassociated with exce,s of IgA and occasionally wilhIgG.

126



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/


MACROGLOBULINAEMIAS The advances in our know-ledge of the distribution and detailed structure ofthe macroglobulins necessitates a recasting of theclassification of diseases associated with overpro-duction of these molecules.The classical Waldenstr6m macroglobulinaemia

is associated with massive amounts of an immuno-globulin having a sedimentation coefficient of about19 Svedberg units, which, on immunochemicalanalysis, can be differentiated from IgG and IgA,the antigenic behaviour of the ,u chain being specific.Cwynarski (1968), examining six cases of macro-globulinaemia, demonstrated that in them the levelsof IgG were diminished or at best normal. This is animportant criterion in establishing 'primary' macro-globulinaemia of the IgM type.Though there is considerable overlap in the clinical

picture, this form of clonal proliferation is usuallydistinguishable from myelomatosis in that malespredominate, that the incidence is at a somewhatolder age, that osteolytic lesions are uncommon,while striking haemorrhagic episodes are frequentlythe presenting symptom. On examination the patientoften has some degree of lymphadenopathy and apalpable spleen. The bone marrow is infiltrated withcells of the lymphocyte series. As mentioned earlier,Bence Jones proteinuria occurs in no more than 15%of patients. The sera display an exotic array ofagglutinating phenomena, the consequence of theexcess of IgM.Of 70 patients in whose sera we have been able to

demonstrate excess of macroglobulins, 45 mightreasonably be regarded as cases of primary Walden-strom type macroglobulinaemia. Almost withoutexception these patients showed some degree ofanaemia when first examined. In 17, severe localhaemorrhage or the dramatic consequence of localhaemorrhage was the presenting symptom. In 10 ofthese there were visual disturbances ranging fromunexplained diminution of vision to a quite suddenblurring of vision. Fortunately, these alarmingvisual disturbances usually subside.The typical retinal picture with the peripheral

microaneurysms has been well described by Ashton,Kok, and Foulds (1963). The picture is not strictlypathogenomic, occurring occasionally in patientssuffering from myelomatosis and polycythaemia.

Microdissection and histochemical investigationshow the aneurysmal walls to be degenerate and tocontain PAS-positive material. This material is alsoseen in the lumen surrounding clumps of red cells.These degenerative changes in the small vessels areseen in other parts of the body and contribute to theviolent nasopharyngeal and gingival haemorrhageexperienced by some patients. The interference withthe blood-clotting mechanism already referred to4

make these haemorrhages unusually difficult tocontrol.

Unexplained cyanosis, bruising, and Raynaud'sphenomena were the presenting symptoms in fivepatients and two presented with flat serpiginousulcers on the legs and back. In these the exotic sero-logy had led to an unwarranted tentative diagnosis.In six patients there was an antecedent history ofrespiratory infection.

Serial studies of the cause of infection in the new-born show that the initial response is accompaniedby an outpouring of the IgM immunoglobulinfollowed by a rising outflow of IgG. Serial studies ofthe serological changes in infective hepatitis andinfective mononucleosis show a marked initial risein IgM which may persist in some patients, while asignificant number of patients suffering from viralpneumonia with haemolytic complications showextremely high cold agglutinin titres and demon-strable increases in IgM. Though Gordon (1953) hasshown that cold agglutinins are macroglobulins, itcannot be assumed that all macroglobulins are coldagglutinins. Ritzmann, Thurm, Truax, and Levin(1960) report a patient in whose sera a gross excessof macroglobulin could be demonstrated but therewas no evidence of the presence of cold agglutinins.Finally, there are a number of protozoal infestations,such as those with Trypopanasoma leishmanii andmalaria, which produce a marked IgM response. Allthese situations must be regarded as examples of'secondary' macroglobulinaemia, and, although inthe majority the diagnosis will be evident, theoccasional atypical patient may present a problem.

It is vital, therefore, that any diagnosis should bebacked by serial differential analysis to demonstratethat the IgM titre is high and remains high and thatIgG and IgA titres are not markedly increased. Thus,there is the well documented and puzzling case of thepatient first seen in 1954 (reported by Ferriman andAnderson in 1956) who four years later (Andersonand Ferriman, 1960) was perfectly well and whosesera at that time showed no excess of macroglobulins.

In the vast majority of these secondary cases, aswith those in which an excess of IgM is associatedwith a pleomorphic tumour, the high level of IgMwill be associated with an increased level of IgG.

THE IYM MONOMERS The pentameric structure ofthe IgM molecule put forward by Miller andMetzger (1965, 1966) suggests immediately thepossibility of IgM monomers circulating either inconditions associated with excess production of thepentamer, Waldenstr6m's macroglobulinaemia, orindependently. When the serum globulins from suchpatients are analysed on a Sephadex 200 column,small peaks or irregularities on the forward edge of

127



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/

N. H. Martin

the 7S peak are sometimes seen (Fig. 7). Moreover,when the final effluent is concentrated and inspectedusing antisera specific for the ,u chain, in addition topositive reactions in the forward peak associatedwith the 19S globulin, reactions are noted withmaterial obtained from the leading edge of the areaassociated with the 7S globulin. This suggests, thoughit does not prove, that in certain patients the heavychain monomer or a fragment of that chain is manu-factured in excess and released into the serumalongside the pentamer in a manner analogous to theliberation of the Bence Jones protein in IgG andIgA myelomata. Though our present methods ofidentification may be crude and insensitive this doesnot appear to occur in more than 200% of patientswith macroglobulinaemia.Solomon and Kunkel (1967) published a detailed

clinical picture of a woman of 67 who appears tohave been suffering from IyM monomer disease.The general course of her disease was not unlikethat of patients suffering from heavy chain disease(Fc fragment). The total protein level, 6 3 g/100 ml,was not raised, but an anomalous peak was notedbetween the P and y globulins. It seems clear fromthe careful physicochemical and immunochemicalanalysis that this aberrant protein contained heavychains with immunological properties resemblingthe ,t chain and light chains of the A type. Ultra-centrifugal analysis of the serum indicated that therewas no demonstrable excess of 19S component.Ultracentrifugal analysis of the isolated anomalousprotein indicated that its sedimentation velocity was6 55. It could thus be regarded as the IgM monomeror molecule akin to the monomer. We have seen twoother patients in whose sera ,uG monomer appearedto be the predominant atypical protein.

Rothfield, Frangione, and Franklin (1965) hadnoted an antinuclear antibody in the sera of patientssuffering from systemic lupus erythematosus in theactive phase which has antigenic behaviour similarto IgM but migrates both in the ultracentrifuge andin the Sephadex column with the fractions rich inIgG and not with 19S IgM. Stobo andTomasi (1967)confirmed this finding in 52 cases examined sufferingfrom disseminated lupus erythematosus, the serafrom 17% of these containing demonstrable amountsof low molecular weight IgM. These were pre-dominantly males and particularly males with lowor absent serum IgA. It appeared that the presenceof IgM was inversely proportional to that of IgA.These observations suggest that the comparativeanatomy of the heavy chains of IgA and IgM re-quires investigation.Though IgM monomer is present in the serum of

20% of patients suffering from Waldenstrom'smacroglobulinaemia, there is as yet insufficient

III Eli p +ve

oC+ ve

+ve

E280

I 1

°o Bed volume

FIG. 7. Serial separation on a Sephadex column of theserum proteins from a patient suffering from macro-globulinaemia. The black blocks represent the immunolog-ical response of successive samples coming off a fraction-ation column to specific antisera, to ,u chain, a chain,and y chain.

evidence to decide whether the presence of the mono-mer is indicative of the primary or the secondarycondition (Martin, 1968).

HEAVY CHAIN' DISEASE Franklin et al (1964)described a syndrome associated with the circulationand excretion of a peculiar protein with a sedi-mentation value of 3 6 to 3 8. This they called 'heavychain' disease. Detailed studies of the protein indicatethat it is, or is closely related to, the Fc fragment ofIgG. Seven cases have been described in satisfactorydetail, others have been observed but not establishedbeyond doubt.The clinical picture is of a male over 40 with a

generalized lymphadenopathy and splenomegaly,often of rapid onset, and painful. This is accom-panied by fever and in three cases notable uvulaland palatal oedema. After the initial onset a propor-tion of cases regress, but all suffered from recurrentinfection, two dying of pneumonia and one ofgeneralized sepsis.The patients are anaemic and generally lympho-

penic. There is gross proteinuria, but no demon-strable Bence Jones protein. Though the ESR israised the levels do not approach those commonlyseen in patients suffering from myelomatosis. Serumelectrophoretic analysis displays a tightly packedband running ahead of the normal position of theIgG band. The protein isolated from both the urineand the serum does not react with either K or Aantiserum. It does react with antisera to the y heavychain and to antisera specifically prepared against

128



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/


the Fc fragment. Physicochemical measurements ofthe protein suggest a molecular weight of between52,000 and 58,000 and that the individual moleculesare symmetrical dimers stabilized by an S:S bondand show Gm specificity. Prahl (1967), examiningthe N and C terminal sequences of the protein chain,noted the persistence of the 'initiator' of the N area,but apparent deletion of the cistron correspondingto the main Fd area, biosynthesis recommencing inthe Fc area.

Alternatively, Askonas and Williamson (1966)have interpreted the quantitative data as suggestinga transposition of the initiating site so that an inter-nal codon is 'read' as initiator. These speculationsabout a rare disease suggest interesting ways inwhich 'freak' proteins may be produced.

LYMPHOMA, LYMPHATIC LEUKAEMIA, AND NEOPLASIASUNRELATED TO THE LYMPHORETICULAR SYSTEM Inan analysis of 110 patients suffering from chroniclymphatic leukaemia, Fairley and Scott (1961)described one patient in whose serum there wasunequivocal evidence of an excess of IgM. Hallen(1966), in a series of 59 patients with chroniclymphatic leukaemia, identified four in whom therewas an increase in IgM. Lymphatic leukaemia mightbe considered as a manifestation of a monoclonaldisease. However, in the patient described by Fairleyand Bodley Scott there were demonstrable excessesof 7S globulin (IgG). Excess of IgM has also beenobserved on occasion in such pleiomorphic condi-tions as Hodgkin's disease (Martin, 1968).

Finally, in any large series of cases there are aresiduum, 2 to 5% of patients, usually in the olderage group, with primary neoplasia unrelated to thelymphoreticular system, such as the epidermis,breast, stomach, and prostate, whose sera on elec-trophoretic analysis show a compact band in theregion of immunoglobulins. These globulins mayhave immunological relationships to IgG, IgA, orIgM (Waldenstrom, 1961).

'IDIOPATHIC' MONOCLONAL PROLIFERATION Theoccurrence of monoclonal proliferation producingan electrophoretic pattern in all ways similar to thatseen in myelomatosis in persons devoid of any symp-toms that could reasonably be associated with thebiochemical findings has remained something of apuzzle ever since Waldenstrom (1944, 1965) des-cribed 'essential hypergammaglobulinaemia'.Norgaard (1964) describes five patients in whom

there was a lapse of six, 10, 13, 13, and 17 years be-tween the first identification of a serum proteinpattern suggesting monoclonal proliferation andrecognition of overt myelomatosis. Wallerstein

(1951) described one patient in whom the time lapsewas 17 years. Well documented cases with three tofour years' history before overt symptoms are notuncommon (Baker and Martin, 1959; Hobbs,1966). In the light of this experience it is difficultto regard any monoclonal proliferation as 'benign'and it is prudent to regard each case as potentiallymalignant. This does not imply that with our presentknowledge these patients should be treated with ourpresent array of cytotoxic drugs but they should befollowed with scrupulous care.

CONCLUSION

The myelomatoses, the macroglobulinaemias, andassociated diseases characterized by the release ofexcessive amounts of specific immunoglobulins ortheir component chains into the vascular space,must be regarded as a group within the frameworkof the lymphoreticular dysplasias. Any neoplasiaof the lymphoreticular system may on occasion beaccompanied by an outpouring of excessive amountsof immunoglobulins, a partial expression of thedislocation of the immunological machinery withwhich the tissues are normally associated.The peculiarity of myelomatosis, primary macro-

globulinaemia, and of the directly related diseases,such as the ,G monomer disease, is that they areregularly associated with a gross overproduction ofa single immunoglobulin and/or its constituentparts. Rarely two immunoglobulins may be secreted,suggesting that two clones of cells are developingindependently. In these cases the analyses of theconstituent molecules of the isolated immuno-globulins never give the complex pattern displayedby their normal counterpart. In considering themachinery of recognition and defence, it is cus-tomary to consider individually the cellular andhumoral component. This is an arbitrary divisiondictated by the interests of the investigator. Themajority of lymphomata are associated with nogross disturbance ofthe circulatory levels ofimmuno-globulins except under challenge. When there is adisturbance it is usually a deficiency. Nevertheless, asmall number of lymphomata and lymphatic leu-kaemias are accompanied by an excessive release ofimmunoglobulins into the vascular space. The sameis true of the multicellular processes grouped underthe heading 'Hodgkin's disease'.A detailed analysis of the globulins produced in

myelomatosis suggests a relatively simple situationin which there is a rapid multiplication of one plasmacell series devoted to the production of a singleimmunoglobulin. In so far as the matter has beentaken there seems to be a concomitant reduction inthe production of other immunoglobulins, and

129



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/

N. H. Martin

devices normally calculated to stimulate humoralantibody produce only a poor response. Plasmacells are constantly being developed as a part of thenormal response to antigenic stimuli. The multi-plication of the 'rogue clone' inhibits either themultiplication of the normal plasma cells or sup-presses their capacity to produce the normal rangeof immunoglobulins.

In the expanding field of tissue recognition andtransplantation, it is recognized that the smalllymphocyte is a critical mediator in homograftrejection. In this field, in recent years, attention hasbeen focused on antilymphocytic globulins producedby the collection of peripheral lymphocytes or lym-phocytes formed from the thoracic duct of onespecies and their injection into another. Such glo-bulins have been prepared in horses from humansand used as an immuno suppressive adjuvant. Theaction appears to be restricted and primarily effectiveon the thymic-dependant lymphocytes. At thepresent time the preparation is crude and of variablepotency, but there is little doubt that with moderntechniques an effective element or elements will beisolated and identified. This will open the way to theproduction of synthetic homologues. Such a develop-ment might prove valuable to surgeons interested intransplantation.

Equally interesting, if the clonal concept of mye-lomatosis is correct, would be the production ofantiplasmocytic globulin against a specific plasmacell clone, presumably in the first instance yA or yG.The evidence seems to suggest that the 'rogueplasma cell' in myelomatosis is a pure bred line andshould be the ideal area in which to attempt theproduction of such serum. In spite of relative successwith synthetic totipotent immunosuppressives, mye-loma stills remains a lethal disease.The production and purification of 'naturally'

occurring antiplasmocytic substances could prove areal therapeutic advance in the treatment of thisgroup of lymphoreticular dysplasias.

My thanks are due to the British Empire Cancer Cam-paign for the support they have given while carrying outthe work described in this review.

REFERENCES

Anderson, A. B., and Ferriman, D. (1960). Brit med. J., 1, 277.Ashton, N., Kok, D. A., and Foulds, W. S. (1963). J. Path. Bact.,

86, 453.Askonas, B. A., and Williamson, A. R. (1966). Proc. roy. Soc. B.,

166, 232.Axeisson, U., Bachmann, R., and Hall6n, J. (1966). Acta med. scand.,

179, 235.Bachmann, R. (1965). Ibid., 178, 801.Baker, G. P., and Martin, N. H. (1959). Brit. med. J., 1, 953.Ballieux, R. E., Imhof, J. W., Mul, N. A. J., Zegers, B. J. M., and

Stoop, J. W. (1968). Clin. chim. Acta, 22, 7-13.Batts, M., Jr (1939). Arch. Surg., 39, 807.

Bennich, K., and Johansson, S.G. 0. (1967). In Gamma-Globulins(Nobel Symposium No. 3), p. 199. Edited by J. Killander.Interscience, New York.

Berggard, L., and Edelman,G. M. (1963).Proc. nat. Acad. Sci. (Wash.),49, 330.

B0e, J. (1945). Acta med. scand., 123, 110.Bruton, 0. C. (1952). Pediatrics, 9, 722.Bull. Wld Hlth Org. (1964). 30, 447.Clamp, J. R., Dawson, G., and Hough, L. (1966). Biochem.J., 100,

35c.Cone, L., and Uhr, J.W. (1964). J.clin. Invest., 43, 2241.Cohen,S. (1966). Proc. roy. Soc. B., 116, 114.Crabb6, P. A., and Heremans,J. F. (1966). Gastroenterology, 51, 305.Creeth, J. M., and Knight, C.G. (1965). Biochim. biophys. Acta (Amst.),

102, 549.Cwynarski, M. T. (1968). Clin. chim. Acta, 19, 1.Dominico, F., and Waldenstrom,J. (1968). Clin. Exp. Immunol.,

3, 911.ElIman, L. L., and Bloch, K. J. (1968). New Engl. J. Med., 278, 1195.Fahey, J. L., Barth, W. F., and Solomon, A. (1965). J. Anmer. med. Ass.,

192, 496.Fahey,J. L., et al (1964). Bull. Wld Hlth Org., 30, 447.

Scoggins, R., Utz, J. F., and Szwed, C. F. (1963). Amer. J.Med., 35, 698.

Fairley,G. H. and Scott, R. B. (1961). Brit. med. J., 2, 920.Ferriman, D. G., and Anderson, A. B. (1956). Ibid., 2, 402.Fink, C. W., Miller, W. E., Jr, Dorward, P., and LoSpalluto, J.

(1965). J. clin. Invest., 41, 1422.Franglen,G. (1968). Personal communication.Franklin, E. C., Lowenstein, J., Bigelow, B., and Meltzer, M. (1964).

Amer. J.Med., 37, 332.Frick, P.G. (1955). Amer. J. clin. Path., 25, 1263.Gleich,G.J., Uhr, J.W.,Vaughan, J. H., and Swedlund, H. A. (1966).

J. clin. Invest., 45, 1334.Glueck, H. I., and Hong, R. (1965). Ibid., 44, 1866.Gordon, R. S., Jr (1953). J. Immunol., 71, 220.Greenwald, H P., Bronfin, G. J., and Auerbach, 0. (1953). Amzer. J.

Med., 15, 198.Gutman, A. B. (1948). Recent Advances in Protein Chenmistry, Vol. 4,

pp. 156-240. Academic Press, New York.Hallen, J. (1966). Acta. med. scand., suppi.,462, 71.Hanson, L. A., and Johansson, B. G. (1967). In Gainnia-Globulins

(Nobel Symposium No. 3), p. 141. Edited by J. Killander.Interscience, New York.

Heremans, J. F. (1959). Clin. chim. Acta, 4, 639.and Crabbe, P. A. (1967). In Gaunima-Globulins (Nobel Sym-posium No. 3), p. 129. Edited by J. Killander. Interscience,New York.

Hobbs, J. R. (1967). Brit. itoed. J., 2, 699.(1967b). Proc. roy. Soc. Med., 60, 1250Slot, G. M. J., Campbell, C. H., Clyne, G. P., Scott, J. T.,Crowther, D., and Swan, H. T. (1966). Lancet, 2, 614.

Holborrow, E. J. (1967). Ibid., 1, 942.lmhof, J. W., Ballieux, R. E., Mul, N. A. H., and Poen, H. (1966).

Ac ta. med. scand., suppl. 445, 102.Johansson, S. G. O., and Bennich, H. (1968). Imnmunologys 13, 381.

, and Wide, L. (1968). Ibid., 14, 265.Kekwick, R. A. (1940). Biochem. J., 34, 1248.Kubota, C., Schwartz, S. O., and Putman, F. W. (1956). Acta haeniat.

(Basel). 16, 105.Laurell, A. H. F. (1961a). Acta. nied. scand., suppl. 367, 69.Laurell, C. B. (1961b). Ibid., suprl. 367, 9, 17, and 23.Lawson, H. A., Stuart, C. A.., Paull, A. M., Phillips, A. M., and

Phillips, R. W. (1955). New Engl. J. Med., 252, 13.McConnell, R. J., and Martin, N. H. (1958). Brit. J. Cantcer, 12, 264.Marks, J. (1953). J. clin. Path., 6, 62.Martin, N. H. (1960). Quart. J. Med. (n.s.), 29, 179.

(1961). Lancet 1, 237.(1968). Clin. chim. Acta, 22, 15.

Miller, D. G. (1962). Ann. intern. Med., 57, 703.Miller, F., and Metzger, H. (1965). J. biol. Chena., 240, 3325.,- (1966). Ibid., 241, 1732.

Milstein, C. (1966a). Proc. roy. Soc. B., 166, 138.(1966b). J. molec. Biol., 21, 203.

Nil6hn, J. E. (1962). Acta med. scand., 171, 491.Norgaard, 0. (1964). Ibid., 176, 137.Oliver, J. (1944-45). Harvey Lect., 40, 102.Ossermann. F. F., and Takatsuki, K. (1964). Amer. J. Mled., 37, 351.Porges, R. F. (1956). J. Lab. cliii. Med., 47, 960.

130



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/


Porter, R. R. (1962). In Basic Problems in Neoplastic Disease, p. 172.Edited by A. Gellhorn and E. Hirschberg, Columbia Univer-sity Press, New York.

Prahl, J. W. (1967). Nature (Lond.), 215, 1386.Ritzmann, S. E., Thurm, R. H., Truax, W. E., and Levin, W. C.

(1960). Arch. intern. Med., 105, 939.Riva, G. (1957). Das Serumeiweissbild. Huber, Berne.Rothfield, N. F., Frangione, B., and Franklin, E. C. (1965). J. clin.

Invest., 44, 62.Rowe, D. S., and Fahey, J. L. (1965). J. exp. Med., 121, 171.Seligman, M., and Mihaesco, C. (1967). In Gamma-Globulins (Nobel

Symposium No. 3), p. 182. Edited by J. Killander. Interscience,New York.

Svehag, S. E., Chesebero, B., and Bloth, B. (1967). Ibid., p. 267.Solomon, A., and Kunkel, H. (1967). Amer. J. Med., 42, 958.Steel, A. E. (1959). Clin. chim. Acta, 4, 503.Stobo, J. D., and Tomasi, T. B., Jr (1967). J. clin. Invest., 46, 1329.Swedlund, H. A., Gleich, G. J., and Chodirker, W. B. (1968). J

Immunol., 100, 1296.Tomasi, T. B., Jr, Tan, E. M., Solomon, A., and Prendergast, R. A.

(1965). J. exp. Med., 121, 101.Walderstrom, J. (1944). Acta med. scand., 117, 216.

(1961). Ibid., suppl. 367, 110.(1965). Advanc. metab. Dis., 2, 117.

Wallenstein, R. S. N. (1951). Amer. J. Med., 10, 325.Watson, L. (1964). Quart. J. Med., 33, 525.

131



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/

The immunoglobulins: AreviewThe immunoglobulins: A review which individual chains are composed has...

Documents

Transcript of The immunoglobulins: AreviewThe immunoglobulins: A review which individual chains are composed has...