The HLA system: structure andfunction · Centromere * ** HLA *Recombination 'hot' spot Fig1 SchematicmapoftheHLAregion. 00 97%ofthetimeallele Ss a1 a2 a3 tm c c c HLA-DPand HLA- samechromosomeas

J Clin Pathol 1987;40:948-958

The HLA system: structure and functionW F BODMER

From the Imperial Cancer Research Fund, London

SUMMARY The HLA system is the major histocompatibility system of man and was found througha search for blood group-like determinants on white blood cells that would be effective in matchingfor transplantation. The HLA system has its counterparts in other species of mammals, birds, andreptiles including the much studied H2 system of the mouse. The HLA system started from a seriesof antigens defined by a combination of relatively crude serology and genetics, supported by exten-sive statistical analysis. It has turned out to be a complex genetic region determining two major setsof cell surface products which mediate essential functional interactions between cells of the immunesystem, and so have a major role in the control of the immune response. Polymorphism in the HLAregion is thus associated with a wide variety of diseases with an immune aetiology.

In this brief review I shall survey some aspects of thestructure and function of the HLA system and theirrelation to disease susceptibility, and the normal andpathological tissue distribution of the two majorclasses of HLA determinants.

The HLA system

Early studies in the search for white cell blood groups,especially by Dausset, showed that there were whitecell agglutinins in the serum of polytransfused sub-jects which were used for the definition of an antigen,but these sera turned out to be too complex to begenerally useful. The further search for antigensdepended on the independent discovery by Van Roodand Payne in 1958 that leucocyte agglutinins wereproduced by fetal-matemnal stimulation. Using suchsera and a statistical approach based on the analysisof 2 x 2 associations of serum reactions on a panel ofcell donors, Van Rood defined a two allele systemwhich he called group 4 in 1962. These statisticalmethods were further developed by Payne et al,' whodefined an apparently independent system of antigenswhich they called LA (L for leukocytes and A for thefirst locus). Subsequent development of the systemmade use of a microcytoxicity assay dependent oncomplement that had been developed by Terasaki.This assay used partially purified peripheral bloodlymphocytes as the target cells, and together withfetal-maternal antisera, remains the mainstay ofHLAtyping, although monoclonal antibodies, as will bedescribed later, are gradually taking over. As furtherantigens associated with the LA and 4 groups weredefined it became clear that these antigens fell into

two series, each of which behaved as if they were con-trolled by a set of alleles at two closely linked loci,now known as the HLA-A and HLA-B loci. Sub-sequently, a third similar locus, HLA-C, was defined.

Following the discovery by Bach and Amos thatthe mixed lymphocyte culture reaction (in which theability of lymphocytes to stimulate each other todivide is detected by the incorporation of tritiatedthymidine) was controlled in association with theHLA antigens, the mixed lymphocyte culture test wasused to define a new set of determinants called HLA-DW. Subsequently, it was established that these wereparalleled by a set of serological determinants thatcould be identified predominantly on peripheralblood B lymphocytes, the HLA-DR, for D-relatedtypes. Further serological and cellular studiesidentified additional similar types called HLA-DQand DP.The development of the HLA system has been

greatly enhanced by a series of international collabo-rative workshops started by Amos in 1964, the ninthof which took place in 1984. The proceedings of theseworkshops effectively define the development of theHLA system, and the latest provides a comprehensivesurvey of the HLA system.2 Further reviews of thesystem are available.3"The loci controlling the two major sets of sero-

logically defined antigens, the HLA-A, B, C or class Ideterminants, and the HLA-D region or class IIdeterminants, are closely linked to each other at eitherend of the HLA region, which lies on the short arm ofchromosome 6. Fig 1 gives a schematic genetic map ofthe HLA system. The region spans a recombinationfraction of between 2 and 3%/o. In other words about

948

copyright. on F

ebruary 13, 2020 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.40.9.948 on 1 S

eptember 1987. D

ownloaded from

Centromere

* **

HLA

* Recombination 'hot' spot

Fig 1 Schematic map ofthe HLA region.

0 0 97% ofthetime alleleSs a1 a2 a3 tm c c c HLA-DP and HLA-

same chromosome asto offspring.

@a2 @a1 J The HLA-A, B, aregion, distal to the

{3 S S A control are presentsSS P2m body. Molecular dat

I least two sets of relafitm analogy with the moiCb lar structures but w

Fig 2 Schematic structure ofHLA-A, B, C molecules and and whose products Itheir genomic organisation. The typical HLA-A, B, C gene sue distribution. Allproduces a molecule with three domains outside the cell, with #2-microglobuliilabelled al, a2, and a3, a transmembrane segment (tm) and a single gene on chrorshort cytoplasmic tail (c). These molecules are associated contains some 25 or rwith h%-microglobulin outside the cell membrane, which however, are not expassociates particularly with the a3 domain. SS indicates the The HLA-D regiopresence ofdisulphide bonds. The domains are codedfor by formed from a functiseparate exons (open boxes) labelled SSfor the signal h (fi 3' bsequence, a], a2, a3 and tmfor the corresponding domains, chain (fig 3), but Inwhile the cytoplasmic tail has its coding sequence split into coded for in the HLAthree exons. mentioned, three maj

L oct ac2 CP,TJ,CY 3'UT

L p1 P2 CP,TM,CY 3'UT

p--- ~ ~ ~ ~~~~~~ OO0---

es at either end of the system, say-A, will be held together on thes they are passed on from parent

nd C loci lie at the end of thecentromere. The antigens theyon most nucleated cells in theLa suggest that there are also atLted loci, called QA and TL byuse H2 system, which have simi-rhich are evolutionarily distincthave a much more restricted tis-these products are associated

in (fig 2), which is coded for by amosome 15. The class I regionmore genes, about half of which,ressed.n or class II products are eachonal heterodimer of an cx and a Pthis case both sets of chains areL-D region. There are, as alreadyjor subsets of products, DP, DQ,

NH2 NH2

S

ASS

V( J

CP ICPTM TM

CYT TCYTFig 3 Schematic structure ofHLA-D region molecules and their genetic organisation. The molecules areformedfrom an aand a p chain, each ofwhich has two domains outside the cell, a] and a2, and PI and P2, a connecting peptide (CP) followed bythe transmembrane (TM), and cytoplasmic regions (CYT). Each ofthe two major external domains has its own correspondingexon. L standsfor leader sequence; 3'UTfor the 3'untranslated region; CP, TM, and CYfor the connecting peptidetransmembrane and cytoplasmic regions. The boxesfor the genomic organisation indicate the exons, and the linesjoining themthe intervening regions.

The HLA system: structure andfunction 949

II

copyright. on F

ebruary 13, 2020 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.40.9.948 on 1 S

eptember 1987. D

ownloaded from

950and DR and this set of genes is at the centromeric endof the region.Both the HLA-D region and HLA-A, B, C sets of

products function in controlling interactions betweencells in the immune response and, through this, play akey part in the regulation of the immune responsethrough interaction with the T lymphocyte antigenreceptor.The strongest evidence that the HLA system is

indeed the major histocompatibility system relevantfor matching for transplantation comes from the factthat kidney or bone marrow grafts exchangedbetween HLA identical siblings survive, with appro-priate immunosuppressive treatment, almost as wellas those between identical twins and far better thangrafts exchanged between mismatched siblings orother relatives. Because the loci controlling all theHLA determinants are closely linked within the HLAregion, on average there is a 25% chance that a pair ofsiblings will be HLA identical, the exceptions beingdue to genetic recombination occurring within theHLA region. Matching for HLA within the family,therefore, ensures that most of the time all thedifferences within the region are matched. Matchingunrelated donors and recipients for HLA-A, B, C andD region determinants has an important effect ongraft survival, but the results are much less strikingthan those with HLA identical siblings. This mayeither mean that there are other determinants con-trolled by the HLA region, which are not yet clearlyidentified, with respect to which matching for trans-plantation is important; or that combinations matterso that, for example, matching for either HLA-A, B,C alone or HLA-D region alone is not as good asmatching for both together. Morris et alt have writtena recent review of the effects of HLA matching ontransplantation.

In between the HLA-A, B, C and HLA-D sub-regions lies a series of genes for some complementcomponents and also for the 21-hydroxylase enzyme,deficient in those with congenital adrenal hyperplasia.The complement genes in HLA are C2, the secondcomponent of the classical complement pathway, fac-tor B (BF), the closely related product of the alterna-tive pathway, and C4 the fourth component of theclassical pathway. The C4 and 21-hydroxylase(210H) genes are duplicated adjacent to those for C2and BF. Recently, the genes for the tumour necrosisfactor a and ,B chains have also been shown to be inthe HLA region.6

Serological and cellular analysis has shown thatthere are at least 18 alleles at the A locus, 41 at B,eight at C, about 20 at DR, three at DQ, and six atDP. Each allele at these loci corresponds to a typedefined by the serological and cellular techniques.Each type is identified by a combination of a letter

Bodmerand a number, corresponding to the locus and theallele within the locus-for example, Al, B8, CW6 orDR4. (TheW in CW6 stands for "workshop" to indi-cate, originally, a less clearly defined specificity. Now,however, the W is mainly used to distinguish the Clocus products from the complement components.)The types thus fall into six different sets correspond-ing to the six loci, A, B, C, DR, DQ and DP, and eachindividual can carry up to two different types corre-sponding to two different alleles at any locus. Thuseach individual may have up to 12 different types, twofrom each of the six sets corresponding to the six loci.These types can create about 80 000 million differentcombinations and generate, overall, an extraor-dinarily high level of polymorphism.

Pairs of alleles at closely linked loci within the sys-tem are often highly associated in the population dueto the phenomenon of linkage disequilibrium orgametic association. Thus Al at the HLA-A locusoccurs in northern Europeans with a frequencyof about 31% and B8, controlled by an allele at theclosely linked HLA-B locus, with a frequency ofabout 21%. The combined phenotype Al and B8occurs with a frequency of about 17%, while if thetwo types were independent in the population theexpected frequency would only be 0-31 x 0-21 or6-5%. The excess of Al B8 types over that expected ifthey were independent is due to gametic associationor linkage disequilibrium-namely, the occurrence inthe population of a relatively high frequency of chro-mosomes that carry the alleles Al and B8 together onthe same chromosome. In the absence of disturbingeffects, such as natural selection, gametic associationsdecline eventually to zero at a rate (1 - r)', where r isthe recombination fraction between the loci and n thenumber of generations. The recombination fractionis, therefore, the major determinant of the extent ofgametic association, and strong associations probablyoccur only for recombination fractions which are lessthan 0-5-0-1%. In practice strong gametic associationis found between alleles at the DQ, DR, complementand HLA-B and C loci, while the average associationbetween alleles at HLA-A and HLA-B is much less,and there is mostly no detectable assocation betweenHLA-DP alleles and other alleles of the HLA regionloci. This fits in with the fact that the recombinationfraction between the HLA-A and HLA-B loci isabout 0-8%, while that between HLA-DP and DQ orDR is about 1-5%. The recombination fractionsbetween the other loci must be much less, as few ifanyrecombinants have been detected between any of theHLA-DQ, DR, complement, B and C loci.The patterns of gametic association and

recombination are reflected in the distribution ofrecombination "hot-spots" (fig 1). These are positionswhere recombination is presumed to occur much

copyright. on F

ebruary 13, 2020 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.40.9.948 on 1 S

eptember 1987. D

ownloaded from

The HLA system: structure andfunctionmore often than elsewhere in the region, and theyhave been clearly identified in the mouse I region,which is the equivalent of the HLA-D region.7 In theHLA region recombination hot-spots are presumedto occur between DP and DQ and between A and C,and these are identified directly by relatively highrecombination fractions-namely, between 0-5% and2%, given the established molecular distance betweenthe genes. In the case of the DQ subregion the hot-spot is identified by a relative lack of gametic associ-ation between alleles at different loci within the sub-region. In general, strong gametic associations maybe expected between alleles at loci within a regionbounded by recombination hot-spots, but notbetween alleles at loci separated by a recombinationhot-spot, even when both are in the HLA region.

Molecular genetics of the HLA region

All the known products of the HLA region have beencloned using standard techniques of molecular gen-etics, generally starting from cDNA clones through toselecting genomic clones. This has established thegenomic organisation of individual genes and of theregion as a whole.8 -10 Sequence data show that theHLA-A, B, C and D region products are members ofthe immunoglobulin supergene family, sharingimportant though limited homology and character-istic structural features with the immunoglobulins.

There is an obvious problem in naming the HLAregion products, which arises because of their havingoriginated from studies of genetic variation, as in thecase of blood groups, rather than as a protein prod-uct, like immunoglobulins and haemoglobins. Whenthe product is studied first, the gene loci are naturallynamed after the product. In the case of the HLA sys-tem, however, there are names for the genes but nogood names for the products. By analogy with theimmunoglobulins, the HLA and other species majorhistocompatibility products can conveniently becalled "histoglobulins." It then seems natural to callthe class I products histoglobulins I or HgI, and theclass II products histoglobulins II or HgII1 1; this pro-visional nomenclature will be used in the rest of thispaper.

951Analysis of the class I or HgI products and their

corresponding DNA sequence has shown them tohave three protein domains outside the cell, followedby a transmembrane region, and a short intracellularcytoplasmic tail. The three main external domainshave corresponding exons, as do the signal sequenceand transmembrane region, while the cytoplasmicdomain is distributed among three small exons (fig 2).All the HgI genes have essentially the same structure,and the major polymorphism of the HLA-A, B, Cgenes is found in the two domains most distal to thecell surface (al and a2). The membrane proximaldomain (a3) is the one with the greatest similarity toimmunoglobulins and is most closely associated with#2-microglobulin, itself a member of the immuno-globulin supergene family.The class II or HgII products have a similar struc-

ture and genomic organisation to HgI (fig 3). Each ofthe two chains has two external domains (cx1, a2 and81, P4 coded for by separate exons. It is also themembrane proximal domains which show the greatestsimilarity to the immunoglobulins.

Fig 4 gives a more detailed map of the HLA-Dregion. From this it can be seen that in addition to theDP, DQ, and DR subregions, other genes have beenfound between the DP and DQ subregions, pro-visionally called DZa and DO#. It is, however, not yetknown to what extent they are expressed and whattheir functional importance might be. The DP andDQ subregions contain duplicated pairs of a and ,Bgenes, only one member of which is expressed, whilethe DR subregion contains two and sometimes threeexpressed fi chain genes, but only one a chain gene.This organisation suggests an origin from a series ofduplication events and some selection for diversifica-tion of the HLA-D region products. By far the mostpolymorphic loci are those for DQ,B, DQa, and one ofthe DR# chains. The other DR3 chain and the DP3chain are also fairly polymorphic, but DQcx is the onlyhighly polymorphic a chain. This pattern of differ-ential variability strongly implies that it is generatedby natural selection. The obvious explanation for thisselection is an effect of polymorphism on immuneresponse differences to clinically important patho-gens, and some form of frequency dependent selectionin relation to this is still the most plausible mechanism

(*) (s)- DP - l- DO - DO - - DR

DP12 DPW2 DP DPW1 DZa DO OX Dx DODOOao DRO ORpq OR(53) DRa DRA

Fig4 Molecular map ofthe human HLA-D region. The blocks indicate the various genes in the DP, DO, DQ and DRsubregions and the arrows above them the directions of transcription. U indicate genes known to be expressed; M those knownto be pseudogenes; and D those which are potentially expressed, but where the expression is not definitely established (basedon Trowsdale etal 198526).

copyright. on F

ebruary 13, 2020 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.40.9.948 on 1 S

eptember 1987. D

ownloaded from

952for the generation of the extensive polymorphism ofthe HLA region.12

Sequence comparisons between the products of theHLA region and also between them and immuno-globulin and other genes of the immunoglobulinsupergene family are beginning to provide a consis-tent overall picture of the pattern of evolution of theregion. The greatest homology between parts of theHLA molecules and immunoglobulins is about20-25% at the amino acid level, which suggests anevolutionary separation of about 750 million years.The differences between HLA-D region and ABCproducts are not much less, and the approximate per-centage of shared amino acid sequence between HgIIa and ,B chains is also only 30%. This suggests that aseries of successive duplications first of all separatedthe HLA region from the immunoglobulins and otherrelated products, and then within the HLA regionseparated HgI and HgII products, and subsequentlywithin HgII a and chains. All of these events arelikely to have happened more than 700 million yearsago. Within the HLA-D region the differencesbetween subregion genes, either ax or ,B, are about45%, while within a subregion the difference betweenthe two DP a chains is about 25%, but between thetwo DQa chains only about 7%. This suggests a fur-ther hierarchy of duplication events, of which themost recent, that giving rise to the division with theDQ subregion, may have occurred only some 15 mil-lion years ago. Sequence variation between alleles,especially neutral substitutions, can given some indi-cation of their age. The data suggest that some allelesmay be five or more million years old and thereforepredate Homo sapiens, but postdate divergence of thehominids from the great apes. There are, however,vestiges of similarities between alleles even in manand mouse, suggesting that some allelic differencesmay, indeed, be much older than this but have natu-rally diverged during mammalian evolution.13

Immune response and disease associations

The discovery by McDevitt et al that specific immuneresponse genes could be found in the mouse H2region3 stimulated the search for HLA and diseaseassociations based on immune response differenceshaving a role in diseases with an immune or auto-immune aetiology. There is now an extraordinarilylong list of HLA and disease associations.'4 Table 1

gives examples of some of the more striking presentlyknown associations. In all these cases an HLA anti-gen has been found much more often in patients withthe disease than in appropriate controls. This indi-cates either a direct effect of the antigen in question,perhaps on immune response, or a susceptibility con-trol by an allele at a closely linked locus that is in

BodmerTable I Some HLA and disease associations

Rheumatoid arthritis 80% DR4Ankylosing spondylitis - 100% B27Juvenile diabetes - 100% DR3 or DR4Narcolepsy - 100% DR2

Other diseases Graves' disease DR3*Myasthenia gravis DR3Multiple sclerosis DR2Psoriasis CW6Coeliac disease DR3Haemachromatosis A3Reactive arthritis B27

100% means almost 100%.*The listed antigens are those with the strongest disease association.

strong gametic association with the allele controllingthe associated antigen. Most of the diseases involved,including in particular rheumatoid arthritis, anky-losing spondylitis, juvenile insulin dependent diabetesmellitus (IDDM), Grave's disease, myasthenia gravis,and coeliac disease are clearly autoimmune diseases.The aetiology of diseases such as haemachromatosisand psoriasis is less clear cut, though in psoriasis thereis some argument for an immune aetiology. Nar-colepsy has become the most striking of the associ-ations, and it is notable that multiple sclerosis sharesthe same antigen. This suggests an immune aetiologytriggered by a virus, which may have more limitedtropism in narcolepsy than in multiple sclerosis.Other mechanisms, however, certainly cannot beruled out, since it has been suggested that HLA-DR2is associated with a differential pattern of sleep in nor-mal persons.'5There are relatively few published examples of a

direct effect of HLA on immune response. Law etalshowed that men who had undergone vasectomy andwho had produced particular types of autoantibodyto sperm had a significantly higher prevalence of theantigen A28 than corresponding controls,'6 and thisobservation has been confirmed by Hancock etal.'7A recent interesting example of a difference inimmune response is the suggestion that antigens ofvarious Mycobacterium species gave a higherresponsiveness in a skin test in HLA-DR4 positivesubjects than in others.'8 This would be exactly thetype of association that might be expected to occur inselection acting on HLA variants through differentialeffects on resistance to important pathogens. Thelater price to pay for this protection is, then, the asso-ciation with autoimmune diseases such as rheumatoidarthritis and IDDM.The association between IDDM and HLA illus-

trates many features of the analysis ofHLA effects ondisease susceptibility. The initial association found bySingal and Blajchman'9 was with HLA-B15, and onlylater was the association with the HLA-D regiondeterminants DW3 and DW4 and then their sero-

copyright. on F

ebruary 13, 2020 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.40.9.948 on 1 S

eptember 1987. D

ownloaded from

The HLA system: structure andfunctionTable 2 HLA-DR genotypes of43 diabetic patients(IDDM) (x means not 3, 4, or 1)

DR type 3/x 3/3 3/4 4/4 4/1 4/x All other

Noobserved 2 1 17 9 10 4 0Expected 59 03 06 03 04 52 303

Based on data from Winearls et al 1984.23 The expected numbers arecalculated from the frequencies seen in a local control population.

logical equivalents, DR3 and DR4, established.20 Thedistribution of HLA types in affected sibling pairsprovided direct evidence for linkage between theHLA system and the gene controlling susceptibility toIDDM.2' In contrast with population associationdata, which depend on a strong gametic associationbetween the antigen observed in patients (which has ahigher prevalence than in controls) and a presumedallele at a closely linked disease susceptibility locus,the sibling pair family analysis is a direct search forlinkage that does not depend on gametic associ-ation.22

Table 2 shows the distribution of genotypes at theHLA-DR locus for 43 patients with IDDM, based onstudies of their families.23 Patients all have eitherDR3 or DR4, and there is a striking excess ofDR3/DR4 and DR4/DR1 heterozygotes, whichagrees with earlier data, at least for DR3/DR4.21Another feature of the data (table 3) is the fact thatthe association between B1 5 and DR4 is muchstronger in patients with IDDM than it is in a controlpopulation. This observation accounts for the factthat the first association between HLA and IDDMwas with B15. This was initially puzzling in relation tothe subsequently established association with DR4, asthere was no strong association between B 15 andDR4 in the normal population. The data clearlyindicate that the primary effect of the HLA system onIDDM is not due to DR4 alone. The susceptibilitycould either be due to an allele at another locusclosely linked to both the DR and B loci, which showsstrong gametic asociation with both Bi 5 and DR4, orit could be due to DR4 together with an allele atanother locus that is associated with B 15. The hetero-

Table 3 B15, DR4 haplotype combinations in diabeticpatients and controls

Haplotye B15, DR4 BIS,x

Patients 16 2Controls 1 6

X' = 98

x means not DR4.Haplotype B15, DR4 means that B5 and DR4 are inherited togetheron the same chromosome.Data from Winearls et al 1984.23

953

zygote effect, as pointed out by Svejgaard et al24 andBodmer25 can be explained by an effect of a particularcombination of a and ,B chains that is only found, say,in DR3/DR4 heterozygotes. As DQa is the only poly-morphic a chain locus, this would imply that the com-binations must be DQa/# combinations,23"2 which isconsistent with the population association data onboth B1 5 and DR4. Attempts are now being made torefine the data on HLA and disease associations bystudying restriction fragment length polymorphismsthat can be identified using clones of the various HLAregion genes.26 Using this approach, Bell et al27found suggestive evidence for the role of HLA-DQfchain polymorphism in the association between DR3and myasthenia gravis.The associations between HLA and cancers are

much less clear cut than those with diseases with adefinite immune aetiology.28 The most obvious possi-bility concerns effects of HLA variation on immuneresponse to viruses which are aetiologically associatedwith cancer. Suggestive associations have been foundbetween HLA-DR5 and the acquired immunedeficiency syndrome (AIDS) virus, human immuno-deficiency virus (HIV)29 30 and between Burkitt'slymphoma, the original source of the Epstein-Barrvirus, and HLA-DR7.3' A weak population associ-ation between HLA-B locus antigens and Hodgkin'sdisease has been confirmed by a sibpair family anal-ysis on patients with Hogkin's disease, showing asevere distortion of the Mendelian HLA segregationexpected in the absence of an HLA effect.28 It isimportant to emphasise that only a very small pro-portion, perhaps up to 3%, of cases of Hodgkin's dis-ease occur in families with two or more affected per-sons. It is within such families that HLA typing canestablish whether the affected pairs of siblings withHodgkin's disease are identical, share only one HLAchromosome but not the other, or have neither chro-mosome in common. The expected incidence of thesethree classes based on the assumption of Mendeliansegregation and in the absence of an HLA effect is1:2:1. It is the departure from this expectation thatshows that there may be a gene in the HLA regionwhich confers susceptibility to Hodgkin's disease. Thedata suggest that it may be possible to find an HLAvariant that is highly associated with Hodgkin's dis-ease in the population, and HLA-DP is an obviouspossibility.32 Hodgkin's disease is interesting in that itshows how a gene, which may only confer a smallchance, say I in 20, of getting the disease, may never-theless, if there is virtually no chance of the diseasedeveloping without the gene, produce an unequivocalinherited susceptibility, even when there is nosignificant familial clustering of the disease.33

The effects of the HLA system may not be directlyon disease susceptibility itself but on the pattern of

copyright. on F

ebruary 13, 2020 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.40.9.948 on 1 S

eptember 1987. D

ownloaded from

954

expression of disease. Thus Oliver et al34 suggest thatthe extent to which germ cell testicular tumoursmetastasise may depend on the patient's HLA type.Patients who were DR7 showed a higher incidence ofstage IV disease, with haematogenous spread to lung,liver, brain or bone than those with other types. Fur-ther data, both from families and population studies,are needed to substantiate this interesting possibilityof a genetic influence on the pattern of tumour metas-tasis.The bases for the effects of the HLA system on

susceptibility to disease need to be established byappropriate functional studies. Transfection ofcloned genes into a cell in which their expression canreadily be detected now provides enormous scope forinvestigating the relation between function and struc-ture. The cloned genes can be changed in defined pos-itions and the effects of this on expressioninvestigated, both at the serological level and in termsof function-for example, in antigen presentation.Thus in our laboratory DP cosmid clones have beentransfected into mouse L cells and expressed both interms of recognition by monoclonal antibodies, andfunctionally, through presentation of antigen."3 Thespecific sequence at the DNA and protein level thathas a role in an immune response underlying an HLAand disease association must eventually be identified.This may mean reconstructing in vitro the basis for anautoimmune attack using appropriate T cell clones,which either recognise the autodeterminant in cellulardestruction, or, mediated through T cells, generate adamaging autoimmune response. Manipulation ofcloned genes expressed in suitable cells will providethe basis for testing whether, for example, a givensequence suspected of underlying an HLA and dis-ease association is critical for the presentation of aparticular antigen. The triggers for the developmentof autoimmunity, be they viruses, other infections, orexposure to antigens in the environment through dietor in other ways, also need to be identified.32 36

Monoclonal antibodies and the distribution ofHLAdeterminants in normal and pathological sue

Because of their specificity, unlimited availability,and their potentially extraordinarily high reactivity,monoclonal antibodies are making a majorcontribution to the analysis of the HLA system. Oneof the earliest monoclonal antibodies against anyhuman determinant was W6/32 which reacts with allHLA-A, B, and C molecules.37 This and other similarantibodies, including also some against f2-micro-globulin, have been widely used for studies on thetissue distribution of the HLA-A, B, C or HgI deter-minants.38 Monoclonal antibodies against HLA-Dregion or HgII determinants have made an important

Bodmercontribution to the analysis of HLA-D region serol-ogy and biochemistry, including the sorting out of thethree sets of products DP, DQ, and DR.3940 In addi-tion to the monomorphic antibodies, there are nowmany polymorphic antibodies for both HLA-A, B,C41 and HLA-D region42 specificities. Monoclonalantibodies to the D region specificities are also help-ing define new determinants, and aiding the sero-logical identification of differences previously onlydefined by cellular techniques. New approaches to theproduction of specific HLA monoclonal antibodies,as well as the development of more sensitive tech-niques based on enzyme linked immunoabsorbentassay (ELISA),43 should eventually make HLA typ-ing using these approaches the routine standard. Thusmouse L cells transfected with human HLA-DP geneshave been used to make a DP specific polymorphicmonoclonal antibody in C3H mice.44 As the L cellsare derived from C3H mice, HLA specificitiesexpressed on their surface should be the only deter-minant, apart from a possible tumour antigen, thatthe C3H mouse can react to.

For some time, red blood cells seemed to be theonly major human normal tissue that did not expressHLA-A, B, C or HgI determinants. Subsequently,especially following the use ofmonoclonal antibodies,sperm and most trophoblasts were also found to lackexpression of HgI determinants on their surface.38This absence of expression by sperm has been the sub-ject of some controversy but has recently been clearlyconfirmed by Kuhlmann et al.45 Its functionalimportance is not yet clear, though it may simplyrepresent a gradual slowing down and stopping ofturnover of the HLA-A, B, C gene products as thesperm mature. This is also assumed to be the case forred blood cells as reticulocytes have been reportedstill to express HgI determinants.Immune T cell attack depends on the presence of

HgI histoglobulins on the surface of the target cells.46The lack of HLA-A, B, and C determinants on thetrophoblast, which carries paternal as well as mater-nal genes (and so should express antigens foreign tothe maternal environment) therefore protects the tro-phoblasts from cellular immune attack against anyantigen. This explains in simple terms why the fetussurvives as an allograft-one of the classical con-undrums of immunology.47 48 Avoidance of rejectionof the fetus is most likely to be due to protectionagainst cell mediated immune attack; HLA and otherantibodies are commonly produced as a result offetal-maternal stimulation and, in most cases, have nountoward consequences for the fetus. Some apparentexpression of HLA-A, B, C related determinants onextra villous trophoblasts has been described by Red-man et al,49 but the importance of this is unclear.A detailed survey of the distribution of HLA-A, B,

copyright. on F

ebruary 13, 2020 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.40.9.948 on 1 S

eptember 1987. D

ownloaded from

The HLA system: structure andfunctionC antigens on normal human organs has beenreported by Daar etal.S° The survey shows that inaddition to red cells, sperm, and trophoblast, certainother cells, such as hepatocytes and neurones, haveeither very low or absent expression of HgI. The lackof HLA-A, B, C antigens from sperm may protectthem from cellular immune attack in the female,before fertilisation, and so may have a functionalbasis that is similar to the lack of these determinantsfrom trophoblasts.5' There is no obvious similarrationale for the low expression or lack of these deter-minants on other tissues, though it may have impli-cations for transplantation-for example of theliver-or using neurones.The lack of HgI determinants on choriocarci-

nomas52 53 presumably explains why this tumour(which, since it carries genes introduced from themale as does the trophoblast, should express foreignantigens), is unaffected by HLA incompatibility.54The choriocarcinoma is protected from T cell immuneattack just like its normal progenitor the trophoblast.These cells, interestingly, switch off HLA-A, B, Cgene expression in coordination, while still syn-thesising and secreting fl2-microglobulin.The first human tumour cell line anomalously

shown to lack completely surface expression of HLA-A, B, C or HgI products was the Daudi cell linederived from a Burkitt's lymphoma. Genetic experi-ments with Daudi showed that its primary geneticdefect was the expression of f32-microglobulin,55 andsubsequent biochemical studies confirmed that HLA-A, B, and C products were synthesised, but in theabsence of #2-microglobulin were not functionallyexpressed on the cell surface.52 The lack of expressionof HgI in a tumour derived cell line Daudi, whicharose from a cell type that normally expresses theirdeterminants on the surface, was suggested by Arce-Gomez etal"` to be a change that could have beenselected for during tumour progression through anadvantage of resistance to T cell immune attack. Suchresistance is presumably only an advantage totumours that express tumour specific antigens in aform that can induce T cell immunity. In the case ofthe Burkitt lymphoma derived cell line the Epstein-Barr virus determinants would be a natural target forsuch immunity. The fact that only few Burkitt's lym-phomas seem to have deranged expression of HLA-A, B, C or fl2-microglobulin determinants52 may beexplained by the observation of Rooney etal56 thatmany Burkitt's lymphomas may lack the Epstein-Barr virus determinants recognised by cytotoxic Tcells, and so may have used this alternative route toescape from host T cell immunity directed against thetumour.These ideas naturally extend to the question of

whether carcinomas may be found that lack HgI

955

expression and, if so, whether this represents anescape from immune response to tumour specific anti-gens selected for during tumour progression. In initialstudies a surprisingly high proportion of carcinomaderived cell lines were shown to have low or absentHgI expression, suggesting that these might corre-spond to clinical situations in which tumours had aspecific antigen that generated a T cell immuneattack.38 52 Other examples of this phenomenon havesubsequently been described.5057A most interesting ramification of these ideas is the

observation that certain adenovirus strains switch offHgI antigen expression in the cells which they infector transform.58 59 Though there is some controversyabout the extent to which this reduced expressionfavours tumour growth, it seems most likely thatthese adenoviruses have evolved a mechanism thatallows them to hide inside a cell and so escape T cellimmune attack with respect to virally induced orexpressed determinants.These observations on variations of HgI expression

on tumours suggest the possibility of a correlationwith prognosis. Some unpublished observations fromour laboratory support this idea in the case of colo-rectal carcinomas. Thus in a screen of sections fromjust over 50 cases derived from a typical sample ofcolorectal carcinomas seen at St Mark's Hospital twowere found with complete lack of reactivity with theHLA-A, B, C monomorphic antibody, W6/32.(Richman P, personal communication). In contrast,we have found that only three of eight colorectalcarcinoma derived cell lines express detectableamounts of HgI on their surface. There is therefore asuggestion that those that do express are either moredifferentiated or less "aggressive" tumours in terms oftheir growth characteristics and consequently prog-nosis (Spurr N, Makin C, and Bodmer W,unpublished observations). The clear implication isthat the cell lines are derived from tumours with amore aggressive phenotype and therefore with apoorer prognosis, which is consistent with the rela-tively low yield of tumours that give rise to growingcell lines. It would then be those tumours with apoorer prognosis which tend more often to lack HgIsurface expression. Perhaps as the tumours progress,they are more likely to develop changes on their cellsurface associated with their malignancy that may berecognised as tumour specific antigens by the immunesystem. As a result, they are then subject to strongerselection by the development of immunity to T cellattack through reduced or absent functional surfaceexpression of HLA-A, B, C determinants.The cellular and serological basis on which HLA-D

region or HgII determinants were defined suggested alimited tissue distribution, specifically on peripheralblood B lymphocytes and also monocytes or macro-

copyright. on F

ebruary 13, 2020 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.40.9.948 on 1 S

eptember 1987. D

ownloaded from

956phages. An early cellular study also showed the pres-ence of these molecules on the surface of endothelialcells.60 This distribution was confirmed by initialstudies with monoclonal antibodies, which in addi-tion hinted at their unexpected presence in other non-lymphoid tissues, in particular the kidney.38 System-atic surveys of the distribution of HLA-D regionHgII determinants subsequently showed their wide-spread, though not universal, presence on endothelialcells and an occasional patchy distribution on epi-thelial cells, including on colorectal carcinomas.61 62The clue to the probable clinical importance of this

distribution of HgII determinants came first from theobservations of Lampert et a163 and Mason et at64that graft versus host disease induced the expressionof HgII products in adjacent epithelial tissues. Thiswas followed by the observation of Bottazo et at65that HgII determinants were expressed on thyroid fol-licular cells in patients with Graves' disease. As aresult of this came a general indication that auto-immune reactions were associated with induction ofexpression of HgD products on affected epithelialcells. They pointed out that such cells expressingHLA-D are able to present antigen to lymphocytes,and that this may be a contributory factor to thedevelopment of autoimmunity.32 At about the sametime that these observations were made it was shownby Wallach et al66 and subsequently by many othersthat y-interferon could induce the expression of HLA-D region determinants on a wide variety of cell types,including in particular, epithelial cells. As y-interferonis released by T cells during immune stimulation,this can explain nearly all the occasions when epi-thelial cells have been observed to express HgIIdeterminants. This expression may be an importantadaptation that enhances T cell response to viraldeterminants expressed on infected epithelial cells.The occasional triggering of autoimmunity may be anunfortunate side effect of this mechanism for enhanc-ing the body's immune response to viral infections.The first observation of unexpected expression of

HgII determinants on tumours was the descriptiongiven by Winchester et al67 of their presence on cer-tain human melanomas, an observation that was sub-sequently confirmed by many others. There is no evi-dence that the normal melanocyte expresses HgII(Bobrow L, personal communication). Perhaps theexpression of HgII on melanomas is induced initiallyby y-interferon, or some other similar differentiationinducer, and is associated with the expression of othercell surface molecules, such as growth factor recep-tors. This is supported by the indication thaty-interferon may induce the expression of the surfacereceptor for tumour necrosis factor.68 Surface expres-sion of a growth factor receptor may then provide astrong selective advantage for tumour progression,

Bodmerand the expression of HgII products may simply be aby-product of this change. Interestingly, Ruiter et al69have observed that melanomas also often lack expres-sion of HgI determinants, and they have suggestedthat there is a better prognosis associated with mela-nomas that have a high expression of HgI and lackHgII than those that express HgII and have a low orabsent expression of HgI. The changed expression ofthe HLA-A, B, C or HgI products may be connectedwith escape from T cell immunity, as seems likely tobe the case for other tumours.Monoclonal antibodies which react differentially

with the HLA-DP, DQ, and DR products have beenused to show that in leukaemias and lymphomas atleast coordinate expression of all three sets of prod-ucts does not always occur.70-72 These observationshave recently been extended to colorectal carcinomasfor which Ghosh et at73 have shown that DR fol-lowed by DP, is most abundantly expressed, DQbeing present in much smaller quantities. Theseauthors state that HgII expression did not correlatewith the Dukes stage of the tumours, though all theirseven Dukes' A expressed HgII, while only 16 of 25 ofthe remainder did so. Observations in our laboratorysuggest that increasing expression of HgII deter-minants on colorectal adenocarcinomas may be asso-ciated with a tendency to be better differentiated(Richman P, personal communication). These pat-terns of differential expression of the HLA-D sub-region products suggest some differences in function.As evolution is a somewhat messy patching up pro-cess, however, a clear cut functional distinctionbetween these products should not necessarily beexpected.32

Monoclonal antibodies are proving to be essentialfor the clear cut characterisation of HLA products innormal tissues, in tumours, and in other diseasestates. Variations in expression clearly may haveimportant functional implications and justifyinvestigation of HLA expression as a part of normalhistopathological investigation. For this purpose, it isextremely valuable to have antibodies that areeffective in routinely fixed wax embedded sections.Results with an antibody against the HLA-DR sub-region a chain, 1 B5, have been described by Epenetosetat74 and show the value of such a reagent forroutine immunohistology. It should now be possibleto make a wider range of these reagents that will helpin the investigation of the expression of HLA regiondeterminants, especially in different sorts of tumoursand their metastases.

Conclusion

The HLA region constitutes about 1/1000th of thehuman genome, corresponding to a size of about 3 x

copyright. on F

ebruary 13, 2020 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.40.9.948 on 1 S

eptember 1987. D

ownloaded from

The HLA system: structure andfunction 957

106 base pairs. The total coding sequences within theregion, which include 6 D region a chains, 10 fichains, about 25 class I chains, and the genes for C4,factor B, C2 and 21-hydroxylase code (includingpseudogenes) for a total of about 14 500 amino acids.The proportion of the DNA sequence which codes, orwhich potentially codes for a protein-the codingratio-is therefore 3 x 14 500/3 x 106, or about 1:70.This is similar to coding ratios found for other knowngenes, but nevertheless leaves room for further func-tionally expressed genes which could perhaps have arole in disease associations,' 1 and the tumour necro-sis factor genes are a possible example. A completecharacterisation of the set of expressed genes in theHLA region would immediately limit the possibleexplanations for the effects of HLA variation on awide range of inherited susceptibilities to some majorchronic disease.The structures and functions of molecules are inex-

tricably interrelated. Thus if we are to understand thefunction of the HLA products at a molecular level itis essential to establish their three dimensional struc-ture using the techniques of x-ray crystallography,and through this and other approaches work out thefunctional basis for their interaction with the T cellreceptor and other key molecules with a role in theimmune response. Such structural knowledge,together with the power of recombinant DNA andcellular techniques, should enable us to unravel therole of the HLA genes in transplant rejection,immune response, disease associations, and tumourprogression.

References

I Payne R, Tripp M, Weigle J, Bodmer WF, Bodmer JG. ColdSpring Harbor Symposia of Quantitative Biology 1964;29:285-95.

2 Albert ED, Baur MP, Mayr WR. In: Albert ED, etal, eds.Histocompatibility testing. Berlin: Springer-Verlag, 1984.

3 Barnstable CJ, Jones EA, Bodmer WF. Genetic structure ofmajor histocompatibility regions. In: Lennox ES, ed. Inter-national review of biochemistry. Defense and recognition.IIA:22. Cellular aspects. Baltimore: University Park Press,1979:151.

4 Crumpton MJ, ed. HLA in medicine. Br Med Bull 1987;43:i-vi.5 Morris PJ, Fuggle SV, Ting A, Wood KJ. HLA and organ trans-

plantation. Br Med Bull 1987;43:184-202.6 Spies T, Morton CC, Nedospasov SA, Fiers W, Pious D,

Strominger JL. Genes for the tumour necrosis factor a and fare linked to the human major histocompatibility complex.Proc Natl Acad Sci USA 1986;83:8699-702.

7 Steinmetz M, Minard K, Horvath S, et al. A molecular map oftheimmune response region from the major histocompatibilitycomplex of the mouse. Nature 1982;300:35-42.

8 Moller G. Immunological reviews No 84. Copenhagen: Munks-gaard, 1985a.

9 M6ller G. Immunological reviews No 85. Copenhagen: Munks-gaard, 1985b.

10 Auffray C, Strominger JL. Molecular genetics of the humanmajor histocompatibility complex. Adv Hum Genet 1986;15:197-247.

11 Bodmer WF. Human genetics: the molecular challenge. ColdSpring Harbor Symposia of Quantitative Biology. Molecularbiology of Homo Sapiens (In press).

12 Bodmer WF. Evolutionary significance of the HL-A system.Nature 1972;237:139-45.

13 Bodmer WF, Trowsdale J, Young J, Bodmer J. Gene clusters andthe evolution of the major histocompatibility system. PhilTrans R Soc Lond B 1986;312:303-15.

14 Tiwari JL, Terasaki PI. HLA and disease associations. New York:Springer Verlag, 1985.

15 Schulz H, Geisler P, Pollmaecher T, et al. HLA-DR2 correlateswith rapid eye movement sleep latency in normal humansubjects. Lancet 1986;ii:803.

16 Law HY, Bodmer WF, Mathews JD, Skegg DCG. The immuneresponse to vasectomy and its relation to the HLA system.Tissue Antigens 1979;14:1 15-39.

17 Hancock RJT, Duncan D, Carey S, Cockett ATK, May A.Anti-sperm antibodies, HLA and semen analysis. Lancet1983;ii:847-8.

18 OttenhoffTHM, Torres P, de las Aguas JT, et al. Evidence for anHLA-DR4-associated immune response gene for myco-bacterium tuberculosis. Lancet 1986;ii:310-2.

19 Singal DP, Blajchman MA. Histocompatibility (HL-A) antigens,lymphocytotoxic antibodies and tissue antibodies in patientswith diabetes mellitus. Diabetes 1973;22:429-32.

20 Bertrams J, Baur MP. Insulin dependent diabetes mellitus. In:Albert ED, et al, eds. Histocompatibility testing. Berlin:Springer Verlag, 1984:348-58.

21 Svejgaard A, Platz P, Ryder LP. Insulin-dependent diabetes mel-litus. In: Terasaki PI, ed. Histocompatibility testing. LosAngeles: UCLA Tissue Typing Laboratory, 1980:638-56.

22 Thomson G, Bodmer WF. The genetic analysis of HLA anddisease associations. In: Dausset J, Svejgaard A, eds. HLA anddisease. Copenhagen: Munksgaard, 1977:84-93.

23 Winearls BC, Bodmer J, Bodmer WF, et al. A family study of theassociation between insulin dependent diabetes mellitus, auto-antibodies and the HLA system. Tissue Antigens1984;24:234-46.

24 Svejgaard A, Platz P, Ryder LP. HLA and disease 1982-a sur-vey. Immunol Rev 1982;70:193-218.

25 Bodmer WF. In: Albert E, etal, eds. Histocompatibility testing.1984. Berlin: Springer Verlag, 1984:11-22.

26 Trowsdale J, Young JAT, Kelly AP, etal. Structure, sequenceand polymorphism in the HLA-D region. Immunol Rev1985;85:5-43.

27 Bell J, Rassenti L, Smoot S, et al. HLA-DQ beta-chain poly-morphism linked to myasthenia gravis. Lancet 1986;i:1058-60.

28 Dausset J, Colombani J, Hors J. Major histocompatibility com-plex and cancer, with special reference to human familialtumours (Hodgkin's disease and other malignancies). CancerSurveys 1982;1:1 19-47.

29 Pollack MS, Safai B, Dupont B. HLA-DR5 and DR2 are sus-ceptibility factors for acquired immunodeficiency syndromewith Kaposi's sarcoma in different ethnic subpopulations.Disease Markers 1983;1:135-9.

30 Pollack MS, Falk J, Gazit E, et al. Classical and AIDS Kaposi'ssarcoma. In: Albert ED, etal, eds. Histocompatibility testing1984. Berlin: Springer-Verlag, 1984:403-6.

31 Jones EH, Biggar RJ, Nkrumag FK, Lawler SD. Study of theHLA system in Burkitt's lymphoma. Hum Immunol1980;3:207-10.

32 Bodmer WF. HLA today. Hum Immunol (In press).33 Bodmer WF. Genetic susceptibility to cancer. In: Forner JG,

Rhoads JE, eds. Accomplishments in cancer research. 1985 prizeyear, General Motors cancer researchfoundation. JB Lippincott&Co, 1986:198-211.

34 Oliver RTD, Stephenson CA, Parkinson MC, etal. Germ celltumours of the testicle as a model ofMHC influence on humanmalignancy. Lancet 1986;i: 1506.

35 Austin P, Trowsdale J, Rudd C, Bodmer W, Feldmann M,

copyright. on F

ebruary 13, 2020 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.40.9.948 on 1 S

eptember 1987. D

ownloaded from

958 BodmerLamb J. Functional expression of HLA-DP genes transfectedinto mouse fibroblasts. Nature 1985;313:61-4.

36 Bodmer WF. Models and mechanisms for HLA and disease asso-ciations. J Exp Med 1980;152:353s-7s.

37 Barnstable CJ, Bodmer WF, Brown G, etal. Production ofmonoclonal antibodies to group A erythrocytes. HLA andother human cell surface antigens-new tools for geneticanalysis. Cell 1978;14:9-20.

38 Brodsky FM, Parham P, Barnstable CF, Crumpton MJ,Bodmer WF. Monoclonal antibodies for analysis of the HLAsystem. Immunol Rev 1979;47:3-61.

39 Crumpton MJ, Bodmer JG, Bodmer WF, Heyes JM, Lindsay J,Rudd CE. Biochemistry of class II antigens: workshop report.In: Albert ED et al, eds. Histocompatibility testing 1984. Berlin:Springer-Verlag, 1984:29-37.

40 Bodmer JG, Bodmer W. Monoclonal antibodies in HLA deter-minants. Br Med Bull 1984;40:267-75.

41 Fauchet R, Bodmer JG, Kennedy LJ, etal. HLA-A, B, C mono-clonal antibodies. In: Albert ED etal, eds. Histocompatibilitytesting 1984. Berlin: Springer-Verlag, 1984:211-16.

42 Bodmer JG, Kennedy LJ, Aizawa M, et al. HLA-D region mono-clonal antibodies. In: Albert ED etal, eds. Histocompatibilitytesting 1984. Berlin: Springer-Verlag:217-36.

43 Durbin H, Bodmer WF. A sensitive micro-immunoassay usingf-galactosidase/anti-f-galactosidase complexes. J ImmunolMethods 1986.

44 Heyes J, Austin P, Bodmer J, etal. Monoclonal antibodies toHLA-DP transfected mouse L cells. Proc Nat! Acad Sci USA1986;83:3417-21.

45 Kuhlmann D, Dohr G, Pusch HH, etal. Absence of HLA class Iand class II antigens as well as fi2-microglobulin from normaland pathological human spermatozoa. Tissue Antigens1986;27:179-84.

46 Doherty PC, Blanden RV, Zinkernagal RM. Transplant Rev1976;29:89-124.

47 Goodfellow PN, Barnstable CJ, Bodmer WF, Snary D,Crumpton MJ. Expression of HLA system antigens on pla-centa. Transplantation 1976;22:595-603.

48 Barnstable CJ, Bodmer WF. Immunology and the fetus. Lancet1978;i:326.

49 Redman CWG, McMichael AJ, Stirrat GM, Sunderland CA,Ting A. Class I major histocompatibility complex antigens onhuman extra-villous trophoblast. Immunol 1984;52:457-68.

50 Daar AS, Fuggle SV, Fabre JW, Ting A, Morris PJ. The detaileddistribution of HLA-A, B, C antigens in normal humanorgans. Transplantation 1984a;38:287-92.

51 Bodmer WF. HLA structure and function: a contemporary view.Tissue Antigens 1981;17:9-20.

52 Travers PJ, Arklie JL, Trowsdale J, Patillo RA, Bodmer WF.Lack of expression of HLA-ABC antigens in choriocarcinomaand other human tumour cell lines. National Cancer InstituteMonograph No 60. Research frontiers in aging and cancer1982:175-80.

53 Yamashita K, Nakamura T, Shimizu T, Katagiri M, Ikeda H.Expression ofHLA class I and class II antigens in human cho-riocarcinoma cell lines. Int J Gynaecol Obstet 1986;24:301-7.

54 Lawler SD. HLA and trophoblastic tumours. Br Med Bull1978;34:305-8.

55 Arce-Gomez B, Jones EA, Barnstable CJ, Solomon E,Bodmer WF. The genetic control of HLA-A and B antigens insomatic cell hydrids: Requirement for fi2-microglobulin. TissueAntigens 1978;11:96-112.

56 Rooney CM, Row M, Wallace LE, Rickinson AB. Epstein-Barrvirus-positive Burkitt's lymphoma cells not recognised byvirus-specific T-cell surveillance. Nature 1985;317:629-31.

57 Sanderson AR, Beverley PCL. Interferon, #2-microglobulin andimmunoselection in the pathway to malignancy. ImmunologyToday 1983;4:211-13.

58 Schrier PI, Bernards R, Vaessen RTMJ, Houweling A,van der Eb AJ. Expression of class I major histocompatibilityantigens switched off by highly oncogenic adenovirus 12 intransformed rat cells. Nature 1983;305:771-5.

59 Paabo S, Nilsson T, Peterson PA. Adenoviruses of subgenera B,C, D and E modulate cell-surface expression of major histo-compatibility complex class I antigens. Proc Natl Acad SciUSA 1986;83:9665-9.

60 Hirshberg H, Evensen SA, Henricksen T, Thorsby-E. Transplant1975;19:495-504.

61 Daar AS, Fuggle SV, Fabre JW, Ting A, Morris PJ. The detaileddistribution of MHC class II antigens in normal humanorgans. Transplantation 1984b;38:293-8.

62 Daar AS, Fuggle SC, Ting A, Fabre JW. Anomalous expressionof HLA-DR antigens on human colorectal cancer cells. JImmunol 1982;129:447-9.

63 Lampert LA, Suitters AJ, Chisholm PM. Expression of Ta antigenon epidermal keratinocytes in graft-versus-host disease. Nature(Lond) 1981;293:149-50.

64 Mason DW, Dallman M, Barclay AN. Graft-versus-host diseaseinduces expression of Ia antigen in rat epidermal cells and gutepithelium. Nature 1981;293:150-1.

65 Bottazzo GF, Pujol-Borrell R, Hanafusa T, Feldmann M. Roleof aberrant HLA-DR expression and antigen presentation ininduction of endocrine autoimmunity. Lancet 1983;ii: 1115-9.

66 Wallach D, Fellous M, Revel M. Preferential effect of gammainterferon on the synthesis ofHLA antigens and their MRNASin human cells. Nature 1982;229:833-6.

67 Winchester RJ, Wang CY, Gibofsky A, Kunkel HG, Lloyd KO,Old LJ. Expression of IA-like antigens on cultured humanmalignant melanoma cell lines. Proc Natl Acad Sci USA1978;75:6235-9.

68 Aggarwal BB, Eessalu TE, Hass PE. Characterisation of recep-tors for human tumour necrosis factor and their regulation byy-interferon. Nature 1985;318:665-7.

69 Ruiter DJ, Bermann W, Welvaart K, etal. Immunohistochemicalanalysis of malignant melanomas and nevocellular nevi withmonoclonal antibodies to distinct monomorphic determinantsof HLA antigens. Cancer Res 1984;44:3930-5.

70 Newman RA, Delia D, Greaves MF, Navarrete C, Fainboim L,Festenstein H. Differential expression of HLA-DR and DR-linked determinants of human leukaemia and lymphoid cells.Eur J Immunol 1983;13:172.

71 Carrel S, Schmidt-Kessen A, Giuffre L. Recombinant interferon-y can induce the expression of HLA-DR and -DC on DR-negative melanoma cells and enhance the expression of HLA-ABC and tumour-associated antigens. Eur J Immunol1985;15:1 18-23.

72 Bodmer JG, Heyes JM, Lindsay J. Study of monoclonal anti-bodies to the HLA-D region products DQwl and DRw52. In:Albert ED etal, eds. Histocompatibility testing 1984. Berlin:Springer-Verlag, 1984:432-8.

73 Ghosh AK, Moore M, Street AJ, Howat JMT, Schofield PF.Expression of HLA-D sub-region products on human colo-rectal carcinoma. Int J Cancer 198038:459-64.

74 Epenetos AA, Bobrow LG, Adams GE, Collins CM,Isaacson PG, Bodmer WF. A monoclonal antibody thatdetects HLA-D region antigen in routinely fixed, waxembedded sections of normal and neoplastic lymphoid tissues.J Clin Pathol 1985;38:12-17.

Requests for reprints to: Sir Walter Bodmer, FRS, ICRFLaboratories, PO Box 123, Lincoln's Inn Fields, LondonWC2A 3PX, England.

copyright. on F

ebruary 13, 2020 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.40.9.948 on 1 S

eptember 1987. D

ownloaded from