Immunogenetics (2019) 71:161–170

https://doi.org/10.1007/s00251-018-1082-2

REVIEW

Genetics of antigen processing and presentation

Adrian Kelly 1 & John Trowsdale 1

Received: 2 August 2018 / Accepted: 24 August 2018 / Published online: 13 September 2018
# The Author(s) 2018

Abstract
Immune response to disease requires coordinated expression of an army of molecules. The highly polymorphic MHC class I and
class II molecules are key to control of specificity of antigen presentation. Processing of the antigen, to peptides or other moieties,
requires other sets of molecules. For classical class I, this includes TAP peptide transporters, proteasome components and
Tapasin, genes which are encoded within the MHC. Similarly, HLA-DO and -DM, which influence presentation by HLA class
II molecules, are encoded in the MHC region. Analysis of MHC mutants, including point mutations and large deletions, has been
central to understanding the roles of these genes. Mouse genetics has also played a major role. Many other genes have been
identified including those controlling expression of HLA class I and class II at the transcriptional level. Another genetic approach
that has provided insight has been the analysis of microorganisms, including viruses and bacteria that escape immune recognition
by blocking these antigen processing and presentation pathways. Here, we provide a brief history of the genetic approaches, both
traditional and modern, that have been used in the quest to understand antigen processing and presentation.

Keywords MHC . HLA . Antigen processing . Antigen presentation . Genetics

Some history history of these developments is presented in a book on the

natural history of the MHC (Klein 1986).
The early history of the genetics of antigen processing and Inbred strains were pivotal in the next development, where
presentation followed from the work on histocompatibility researchers used them to study the genetics of tumour rejec-
in mice, initially by Peter Gorer, who worked at the Lister tion. As early as 1903, it was discovered that tumours that
Institute and Guy’s hospital in London. His discovery of the grew well when transferred within the same strain were
MHC in turn was inspired by three developments. First was rejected in a different one. Then, in 1922, Little and Johnson
the curious pastime of inbreeding mouse strains, a fashionable showed that transplantation of normal tissue was subject to the
hobby which spread from China. It reached America in the same strain specificity as tumours. A third stimulus was the
early 1900s, and after a while, pioneering geneticists realised development of blood group research, largely attributed to
the advantage of inbred strains for research. Many of the Landsteiner.
mouse strains were started over 100 years ago. C57BL, one JB Haldane suggested that tumour resistance factors may
of the original strains, was designated as the mice were black be akin to blood group antigens, but it was Gorer who per-
and were number 57. BALB/c mice were white, on the other formed experiments to test the idea that antigens were shared
hand, and were designated by their originator, Halsey J. Bagg, by both malignant and normal tissues. This led to the formu-
as Bagg albino, or BALB/c for short. A fascinating early lation of an immunological theory of transplantation, which
was later systematised by Peter Medawar. George Snell was
studying similar phenomena, and after collaborating with
This article is part of the Topical Collection on Biology and Evolution of Gorer, he proposed calling the tumour-resistance factors
Antigen Presentation Histocompatibility genes. His approach started to reveal some
of the complexity of histocompatibility.
* John Trowsdale Early work leading to the discovery of the human HLA
[email protected] complex developed in the 1950s and was dependent on the
1 study of antibodies against alloantigens on white blood cells
Department of Pathology, University of Cambridge,
Cambridge CB21QP, UK by three laboratories: Jean Dausset in Paris, Rose Payne and
162 Immunogenetics (2019) 71:161–170

Walter Bodmer in Stanford, and Jon van Rood in Leiden. It were proposed to be soluble molecules secreted by suppressor
was realised that some patients, and women who had borne T cells. It was a shock to find that discrete I-J genes did not
several children, tended to make such antibodies, which were exist once the I region of the mouse had been cloned
independent of ABO blood groups and erythrocytes. At the (Kronenberg et al. 1983).
time, human organ transplantation was becoming widespread, The genetics of antigen processing and presentation was
and it gradually became accepted that human leukocyte anti- advanced dramatically in the late 1980s when two further
gens were the equivalent of mouse H-2 antigens. The hope major technical developments helped to get to grips with the
was that careful matching, as in ABO, could lead to organ complexity. One was the DNA cloning revolution and the
transplants that were not rejected. It was soon realised that second was the determination of the structure of MHC mole-
the HLA system was more complex than ABO and progress cules from crystals. Initial cloning of H-2 and HLA antigen
depended on exchange of cells and antisera. The International genes led quickly to assembly of maps of the MHC in humans
Histocompatibility Workshops, which have been held every and mice. These dramatically simplified the picture to just a
few years since 1964, were critical in interpreting and integrat- handful of class I and class II loci, albeit with a profound level
ing information obtained with a variety of techniques from of polymorphism. The early maps of mouse MHC were pains-
different laboratories. Analysis of data from these workshops takingly assembled from overlapping cosmids (Steinmetz and
indicated that a single genetic region was pivotal, namely, in Hood 1983). Many different human haplotypes have been
humans, HLA, the Major Histocompatibility Complex. An analysed by these techniques (Horton et al. 2008; Shiina
associated protein chain, β2microglobulin, was identified as et al. 2004), as well as using more modern, high-throughput
a component of HLA antigens and was later mapped outside approaches (Norman et al. 2016).
the complex to chromosome 15 (Goodfellow et al. 1975). The nature of the polymorphism was mysterious, as
Attempts to develop in vitro assays to study graft rejection class I and class II chains encompassed many amino acid
led to the mixed lymphocyte reaction (MLR), which also changes seemingly scattered throughout the first two do-
turned out to be controlled by the MHC region. However, mains of class I and the first domains of both chains of
the results did not correlate completely with the serologically class II. The breakthrough came with the crystal structure
defined determinants. Work in both human and mouse indi- of the first MHC antigen, HLA-A2 by Pamela Bjorkman,
cated that these determinants, as well as MLR responsiveness, who was a PhD student at the time in the Wiley/Strominger
were both part of the H2 and HLA complexes but were sep- laboratories (Bjorkman et al. 1987a, Bjorkman et al.
arated genetically. Meanwhile, a different set of experiments 1987b). The realisation that class I and class II molecules
showed that levels of antibody response to short synthetic possessed a groove which bound peptides immediately
polypeptides were controlled by the MHC. For example, swept away other models of antigen recognition, such as
C57 mice responded well to the branched synthetic polypep- those invoking independent receptors for antigen and his-
tide (T, G)-A-L, but CBA animals were poor responders. tocompatibility molecules on T cells.
These effects were traced to the H-2 region leading to the The development of molecular immunology through the
so-called immune response (Ir) genes (Benacerraf and creation of mutants, DNA sequencing, protein structure and
McDevitt 1972). gene discovery then paved the way for uncovering the various
These many years of work, all indicated that the MHC components of the antigen processing and presentation path-
was a major hub controlling a number of immunological ways, as outlined below (Table 1).
phenomena. Indeed, additional experiments showed that
the MHC also controlled susceptibility to viruses. Work
in Canberra showed that cytotoxic T cells simultaneously Mutant cell lines
recognised viral antigens and MHC molecules, in a phe-
nomenon that came to be known as MHC restriction Panels of HLA homozygous typing cell lines have been im-
(Zinkernagel and Doherty 1974). portant tools in the analysis of human tissue types through
The nature of the MHC became even more complex when HLA classes I and II. These lines were derived by Epstein
it was proposed that T cell suppression was also controlled by Barr Virus transformation of lymphoid cells from consanguin-
the class II region (Green et al. 1983). There followed a decade eous mating, usually first cousin. They have been used for
of controversy over the existence of this phenomenon (Bloom over 40 years by the HLA community and are still in use today
et al. 1992). It was eventually accepted, and the responsible T (Turner et al. 2018).
cells were called regulatory to distinguish them from the con- Analysis of the involvement of the HLA region in anti-
fusing history of suppression (Sakaguchi et al. 2007). The gen processing and presentation was further enhanced by
genetics behind the controversy concerned the I-J gene which the generation of deletions and other mutations in
was supposed to map to the I region of the mouse MHC and lymphoblastoid cell lines (Fig. 1). One series of mutants
control the function of suppressor T cells. The I-J antigens was made by γ ray mutagenesis of the cell line B-LCL721
Immunogenetics (2019) 71:161–170 163

Table 1 Some human antigen

processing and presenting Gene Comments Chromosome
components. Alternative names
and gene designations are given HLA-A Classical class I 6
in parentheses HLA-B Classical class I 6
HLA-C Classical class I 6
HLA-E Non-classical class I 6
HLA-F Non-classical class I 6
HLA-G Non-classical class I 6
MICA Class I –related 6
MICB Class I –related 6
CD1a Presents lipids 1
CD1b Presents lipids 1
CD1c Presents lipids 1
CD1d Presents lipids 1
CD1e
MR1 Presents metabolites to MAIT cells 1
HLA-DR Classical class II 6
HLA-DQ (HLA-MB, DC) Classical class II 6
HLA-DP (HLA-SB) Classical class II 6
HLA-DO (HLA-DNA, DZA + HLA-DOB) Non-classical class II 6
HLA-DM (RING6 + RING7)) Non-classical class II 6
Invariant chain (CD74, Ii) 5
β2microglobulin (B2M) Component of mature class I 15
LMP2 (PSMB9, RING12) Inducible proteasome subunit 6
LMP7 (PSMB8, RING10) Inducible proteasome subunit 6
MECL1 (PSMB10) Inducible proteasome subunit 16
PA28a (PSME1) Assembly of immunoproteasome 14
PA28b (PSME2) Assembly of immunoproteasome 14
TAP1 (RING4, PSF1) Peptide transporter subunit 6
TAP2 (RING11, PSF2) Peptide transporter subunit 6
ERp57 (PDIA3, GRP58) Oxidoreductase in TAP complex 15
ERAP1 Amino peptidase 5
Tapasin (TAPBP) Peptide loader 6
TAPBPR (TAPBPL) Peptide editor 12
BIP (HSPA5) ER chaperone 9
GILT (IFI30, IP30) Thiol reductase 19
PDI (P4HB) ER chaperone (redox-regulated) 19
Calreticulin (CALR) ER chaperone 19
Calnexin (CANX) ER chaperone 5
UGT1 (UGT1A1) Glucuronosyltransferase 2
CIITA Master class II transcription factor 16
NLRC5 Master class I transcription factor 16

by the DeMars group (DeMars et al. 1984; Kavathas et al. More details on the class I HLA types are available on the
1980). Cells were subjected to 2 cycles of mutagenesis IPD-IMGT/HLA database.1 To our knowledge, not all class II
followed by immune-selection with monoclonal antibodies genes have been typed in these cells. Mutant cell lines were
for MHC antigen loss. B-LCL721 contains the following isolated by subjecting these cells to irradiation. After 5 days
two MHC haplotypes: growth, the survivors were selected with complement and a
variety of monoclonal antibodies. Analysis of these mutants
HLA-A*01 HLA-B*08 HLA-DRB1*03 HLA-DPB1*04
1
HLA-A*02 HLA-B*51 HLA-DRB1*01 HLA-DPB1*02 https://www.ebi.ac.uk/cgi-bin/ipd/imgt/hla/fetch_cell.cgi?10972
164 Immunogenetics (2019) 71:161–170

showed that they contained large homozygous or hemizygous with EMS an HLA-DR3+ cell line, T5-1 (Mellins et al. 1988;
deletions over the MHC region (Fig. 1). One such line Pious et al. 1985). Similarly, in mouse studies designed to in-
LCL721.174, and its derivative 174xCEM.T2, has been used vestigate the nature of self and non-self-recognition, Klaus
extensively to investigate TAP transporter function and MHC Karre generated a murine cell line RMA-S that lacked surface
class I peptide loading. 174xCEM.T2, more commonly re- H-2 expression but was rejected in tumour transplant models
ferred to as T2, was generated by fusing LCL721.174 with (Karre et al. 1986). Analysis of this cell line was instrumental in
CEMR.3, a T-LCL, and selecting for loss of CEMR.3-derived providing early models for peptide loading by MHC class I
copies of chromosome 6 (Salter et al. 1985). Both molecules (Townsend et al. 1989). In addition, Roberto
174XCEM.T2 and LCL721.174 therefore carry the same Accolla irradiated Raji cells to produce a mutant line RJ2.2.5
chromosome 6 MHC deletion. selecting against HLA-DR expression with an antibody D1–
A similar procedure was used by Andreas Ziegler, using 12(Accolla 1983), facilitating identification of CIITA.
the lymphoma cell line BJAB.B95.8.6 (Spring et al. 1985). These mutant cell lines and their derivatives have been in-
The wild-type cells were typed with the resolution available at strumental in a long series of experimental approaches to uncov-
that time as: er various components of antigen processing, and they continue
to be important tools in investigating their mechanism of action.
HLA-A*1 HLA-B*35 HLA-C*4 HLA-DR*5 HLA-
DQw3 HLA-DP*4
HLA-A*2 HLA-B*13 HLA-C* HLA-DR* HLA-DQw1 TAP transporters
HLA-DP*2
A key insight into antigen processing was the finding that
The Pious laboratory used a variation of this approach to influenza A-specific cytotoxic T lymphocytes (CTL) recog-
isolate mutants in class II (8.1.6 and 9.28.6) by mutagenizing nise discrete epitopes of the nucleoprotein molecule. These

Fig. 1 Composition of the mutant 721.84.5

721.175 721.82.4 721.174
lines derived in the DeMars
laboratory (DeMars et al. 1984; A 1 2 A 1 2 A 1 2
Kavathas et al. 1980) and the fur- B 8 5 B 8 5 B 8 5
DR 3 1 moab
ther derivative T2 (Salter et al. DR 3 1 CC 6.4
DR 3 1
DQ 2 1 DQ 2 1 DQ 2 1 721.174 x CEM
1985). The shaded boxes show DP 4 2 DP 4 2 DP 4 2 cell fusion
the proposed extents of the dele- GLO 2 1 GLO 2 1 GLO 2 1
tions. See text for more details of moab
721.180.2 Chromosome
typing. These mutant cells have CC 11.23 loss
an- moab an- moab
been used in numerous antigen DR: L243 DR: L243
processing and presentation stud-
ies. The Figure adapted from A 1 2 A 1 2 A 1 2 A 1 2
Demars et al. 1984 B 8 5 B 8 5 B 8 5 B 8 5
DR 3 1 an-A2: DR 3 1 DR 3 1 DR 3 1
an-A2:
DQ 2 1 moab BB 7.2 DQ 2 1 DQ 2 1 DQ 2 1
moab BB 7.2
DP 4 2 DP 4 2 DP 4 2 DP 4 2
GLO 2 1 GLO 2 1 GLO 2 1 GLO 2 1

721.134 721.45.1 721.144 T2

721.61

A 1 2 A 1 2 A 1 2
B 8 5 B 8 5 B 8 5
an-A2:
DR 3 1 an-BB: DR 3 1 DR 3 1
moab BB 7.2
DQ 2 1 DQ 2 1 DQ 2 1
DP 4 2 DP 4 2 DP 4 2
GLO 2 1 GLO 2 1 GLO 2 1
721.19 LCL 721 721.127

an-B5:
721.53
A 1 2
B 8 5
DQ 2 1
DR 3 1
DP 4 2
GLO 2 1
Immunogenetics (2019) 71:161–170 165

data implied a mechanism for generation of viral protein frag- E. coli and P. aeruginosa). About half of them exhibit
ments and their transport to the cell surface for presentation to granulomatous skin lesions. This may be due to suboptimal
CTL (Townsend et al. 1985). The peptide-binding groove in cytokine and cellular responses, leading to tissue damage
HLA molecules was a strong contender for the peptide carrier. and favouring subsequent bacterial infection and further
There was a problem though in that MHC molecules were recruitment of phagocytic cells. Indeed, antibiotics that
integrated into the cell membrane, so how did peptide frag- impair neutrophils appear to be beneficial.
ments, which lacked signal sequences, access the endoplasmic Patients with defects in TAP do not suffer from severe
reticulum (ER) to be loaded? How were the peptides viral infection. This is unexpected but may relate to the fact
transported to the cell surface on MHC molecules? that they have a normal humoral (antibody) response and
The solution came from MHC mapping and analysis of the significant numbers of T cells, particularly γδT cells, as
mutants referred to above. Several groups simultaneously well as NK cells and neutrophils. Additionally, presenta-
identified genes encoding transporter molecules mapping to tion of some viral antigens is TAP-independent. Another
the MHC. These were transporters of the so-called ABC symptom in these patients is necrotizing granulomatous
(ATP-binding cassette) type, which were subsequently named lesions particularly around the nose, resulting in loss of
TAP, for Transporters Associated with Antigen Processing the septum and associated cartilage. Presentation is vari-
(Deverson et al. 1990; Monaco et al. 1990; Spies et al. 1990; able though. The skin ulcers were found to contain macro-
Trowsdale et al. 1990). These data suggested immediately that phages and NK cells.
the transporters, encoded as a heterodimer of two proteins, In terms of biological effects, the proportion of T cells may
TAP1 and TAP2, were responsible for pumping peptides into be reduced in patients with TAP defects, but there is wide
the ER (Kelly et al. 1992). To confirm this, the TAP1 sequence variation in phenotype. They tend to have a higher proportion
restored HLA class I expression when transfected into a cell of γδT cells. Interestingly, numbers of NK cells are normal,
line with a mutant TAP1 gene, LCL721.134 (Spies and but they have no cytotoxic activity against the standard class I-
DeMars 1991), and the TAP2 sequence restored function negative targets, unless activated by cytokines.
and co-precipitation of TAP1 and TAP2 when transfected into
cells deficient in TAP2 (Kelly et al. 1992). However, this was
not the case when using LCL721.174 cells, which maintained
a large deletion over the class II region, covering both com- Proteasome components—LMPs
ponents of the heterodimer, TAP1 and TAP2(Spies and
DeMars 1991). Studies of rats and chickens were particularly At the same time that TAPs were discovered, clues to the
informative as the TAP genes are polymorphic in these spe- proteolytic machinery, responsible for breaking the antigenic
cies. It turned out that specific allelic versions of TAP supply proteins into peptides, were also provided, by studying mu-
peptides appropriate for class I molecules whose genes are tants over the MHC region. The biochemical prelude to this
linked in cis on the haplotype. Inappropriate pairing of TAP discovery came in the early 1980s when Monaco and
alleles with class I alleles they serve with non-binding pep- McDevitt described a series of up to16 low-molecular weight
tides led to reduced expression of class I, which could be proteins, LMPs, that mapped to the MHC. Initially, the func-
detected at the cell surface (Deverson et al. 1990; Kaufman tions of these proteins were not obvious, but they formed a
2015; Powis et al. 1996). complex by co-precipitation and varied between different
mouse haplotypes, in other words were polymorphic
(Monaco and McDevitt 1984). It was subsequently
Natural mutants in TAP established that genes for two interferon-inducible, catalytic
proteasome subunits, LMP2 (PSMB9; β1i) and LMP7
A number of individuals have been identified with (PSMB8: β5i), were closely linked to the TAP1 and TAP2
immune-deficiencies due to mutations in TAP (de la Salle genes in both human and mouse MHC (Glynne et al. 1991;
et al. 1994; Zimmer et al. 2005). These belong to the cat- Kelly et al. 1991a). As with the TAPs, finding proteasome
egory of type I bare lymphocyte (BLS) syndrome, where components encoded in the MHC immediately suggested that
HLA class I, but not class II, is affected. These patients the proteasome was responsible for producing the antigenic
generally have a severe reduction of class I at the cell peptides. Proteasomes are ubiquitous cellular components re-
surface, but in spite of this, most of these individuals reach sponsible for continual turnover of proteins in general.
adulthood. They are rare and are almost exclusively HLA However, the MHC-encoded proteasome genes were interfer-
homozygous due to first cousin parentage. They generally on inducible and, when expressed, replaced constitutive com-
display chronic infections of the respiratory tract (purulent ponents in a subset of the main structures. Another interferon-
rhinitis, pansinusitis, otitis media) with bacteria inducible subunit PSMB10 (β2i) is encoded outside the MHC
(H. influenzae, S. pneumoniae, S. aureus, K. pneumoniae, on chromosome 16q22.1.
166 Immunogenetics (2019) 71:161–170

ERp57 and ERAP1 Pasquier, was at chromosome position 12p13.3. A gene in this
region shared sequence homology with Tapasin, dubbed
A number of chaperones are associated with class I as it TAPBPR (Teng et al. 2002). The product of this gene was
matures and picks up peptide in the peptide-loading com- subsequently shown to be a novel component of the MHC
plex. These chaperones serve multiple proteins in the ER class I presentation pathway, which acts as a peptide exchange
and are not dedicated to class I. However, ERp57 is a catalyst (Hermann et al. 2015; Neerincx and Boyle 2017).
component of the peptide-loading complex. The protein TAPBPR interacts with class I in a similar manner to Tapasin
aids folding of nascent glycoproteins by facilitating di- but is not complexed to TAP (Jiang et al. 2017). TAPBPR links
sulphide bond isomerisation. UDP-glucose:glycoprotein glucosyltransferase 1 onto MHC
ERAP1 is an amino peptidase. It plays a role in determin- class I (Neerincx et al. 2017). Tapasin is monomorphic, but
ing the length and sequence of peptides bound and presented TAPBPR is relatively polymorphic and there are a number of
by class I allotypes. HLA-B27 is strongly associated with common alleles and splice variants in humans (Porter et al.
ankylosing spondylitis (Chen et al. 2014), and genetic studies 2014). The relationship between the products of these alleles
indicated that certain ERAP1 allotypes are more associated and MHC Class I loci and alleles is not known. As with
with ankylosing spondylitis than others, due to their contribu- Tapasin, it appears that some alleles of class I are dependent,
tion to peptide trimming (Reeves et al. 2014a, Reeves et al. and others independent, of the molecule.
2014b). There has been some controversy over this effect as it
was difficult to analyse genetically. It was proposed to involve
specific combinations of different ERAP1 allotypes, making
association studies difficult (Reeves et al. 2015; Robinson and HLA-DM and DO
Brown 2015).
The existence of an MHC-encoded factor that was essential
for MHC class II antigen presentation was first suggested
TAPASIN through the characterisation of mutagenised B lymphoblastoid
cells selected for loss of HLA-DR3 reactivity with a DR3-
The gene for Tapasin, another component of the antigen pro- specific antibody, 16:23(Mellins et al. 1990). These cells had
cessing machinery, is located just outside the MHC region normal levels of cell surface DR but were unable to present
(Ortmann et al. 1997). The product of this gene, which is an native antigen to T cells. They possessed class II molecules
additional member of the immunoglobulin gene family, links that were predominantly filled with the CLIP peptide and fell
the peptide transporter to the nascent class I molecule in a apart in the presence of SDS. The defect was eventually
complex, the peptide-loading complex (Blees et al. 2017). mapped to HLA-DM (Morris et al. 1994), a class II-related
Optimisation of the MHC class I peptide cargo is proposed molecule previously identified in both mouse and human in
to be dependent on Tapasin (Williams et al. 2002). However, screens for expressed genes encoded on cosmids spanning the
class I molecules vary in their dependence on the protein, MHC region (Kelly et al. 1991b; Cho et al. 1991). HLA-DM
an effect that has been mapped to specific positions on the acts as a peptide editing chaperone stabilising class II whilst
HLA molecule (Park et al. 2003) as well as to its confor- exchanging CLIP for antigenic peptide. DM activity is regu-
mational flexibility (Garstka et al. 2011). For example, the lated by a second non-classical class II molecule, now called
HLA-B*44 allele, which differs exclusively at position HLA-DO. HLA-DO alpha and beta chains were first identi-
156, B*44:02 (156Asp), is Tapasin-dependent, whereas fied due to their high homology with human and murine clas-
other alleles are not (Badrinath et al. 2012). The reason sical class II molecules (Larhammar et al. 1985; Tonnelle et al.
for this is not known but could relate to the fact that some 1985; Trowsdale and Kelly 1985). It proved much harder to
virus products compromise antigen presentation by decipher the role of DO even though the molecule was iden-
interacting with the Tapasin protein. tified some 6 years before DM. DOA and DOB were not
located together in the MHC, and the chains showed different
mRNA expression patterns initially suggesting that they
TAPBPR would not associate as a pair. DO resides in late endosomal
compartments but requires association with DM in the endo-
There is evidence that the jawed vertebrate genome has under- plasmic reticulum for correct assembly (Liljedahl et al. 1996).
gone duplication at least twice in its evolutionary history, ac- The in vivo role of DO is still debatable, but mechanistically, it
cording to the 2R hypothesis. Accordingly, it was demonstrat- functions as a negative regulator of DM activity (van Ham
ed that there are traces of at least four clusters of genes related et al. 1997; Denzin et al. 1997). Unlike the transient interac-
to those in the MHC proper on chromosome 6 (Kasahara tions seen between DM and DR, the DM/DO interaction is
1999). One paralogous region, identified by Louis Du very stable, suggesting that DO acts as a competitive inhibitor
Immunogenetics (2019) 71:161–170 167

of DM, a view consistent with the DM/DO crystal structure factors. A further indication of the integration along haplo-
(Guce et al. 2013). types came from data relating to the way that class I molecules
‘educate’ by interacting with receptors on NK cells.
Hydrophobic leader sequences from class I molecules supply
Haplotypes and linkage disequilibrium peptides that bind HLA-E, which instructs CD94/NKG2A
receptors on NK cells. Leader sequences of HLA-B molecules
It has been known for some time that the MHC comprises a are dimorphic: Those with − 21 methionine (− 21 M) provide
large region in linkage disequilibrium (LD), or polymorphic peptides that bind HLA-E, whereas − 21 threonine do not.
frozen blocks (Dawkins et al. 1999). In some populations, for Those haplotypes with -21 M rarely encode the ligands for
example, combinations of alleles, such as HLA-A1-B8-DR3, other NK receptors, namely Bw4+ HLA-B and C2+ HLA-C.
are more commonly found together than would be expected From these data, it was proposed that there are two schools of
based on frequencies of individual alleles in the population. HLA haplotypes; one focused on supplying CD94/NKG2A
Several mechanisms may be invoked to explain this. First, ligands, and the other, KIR ligands. Individuals with − 21 M
there may have been insufficient time for recombination after had NKG2A+ cells that were more effective than those with
relatively recent expansion of families, in some cases in iso- only − 21 T (Horowitz et al. 2016). Similarly, HLA-A leader
lated populations. Another possibility is that recombination sequences determine expression levels of HLA-E, again
‘cold spots’ in the MHC sequence restrain the level of ex- influencing interaction with CD94/NKG2A on NK cells and
change of alleles between haplotypes. This may be facilitated in turn HIV replication (Ramsuran et al. 2018).
by a reduced level of pairing at meiosis in a region such as the
MHC, where the variation between haplotypes compromises
homologous interaction. Indeed, overall recombination is low Regulation of transcription and translation
in the MHC.
An attractive explanation for the high LD is clustering of Analysis of mutants has also been productive for identifying
alleles that encode proteins that work well together and main- regulators of antigen processing and presenting genes. For
tenance of functionally coordinated sets of alleles. As many years, it was difficult to track down a master regulator
discussed above, the discovery of both TAP transporters and of class I. On the other hand, mutants affecting class II were
LMP proteasome components mapping to the MHC was isolated from bare lymphocyte patients (Reith and Mach
largely serendipitous and was inspired to some extent by the 2001). The class II master regulator CIITA is a nucleotide-
notion that the MHC was a cluster of genes with inter-related binding domain and leucine-rich repeat receptor (NLR) pro-
functions. Furthermore, it was proposed that linkage of poly- tein. This large gene was initially found by expression cloning
morphic transporters with polymorphic MHC class I genes in a class II-deficient cell line (Steimle et al. 1993). Mutants
permitted functional coordination, such that appropriate bind- invoking loss of function of CIITA generally result in severe
ing peptides, for the class I molecules linked in cis, were immunodeficiency, a form of bare lymphocyte syndrome.
pumped by the relevant TAP. This turned out to be the case Individuals with BLS have impaired antibody and T cell re-
for rats and chickens, where genes for class I and TAP are in sponses due to the lack of MHC class II expression. Class II
close proximity (Powis et al. 1992; Tregaskes et al. 2016). In transcription requires, in addition to CIITA, at least three ad-
the human MHC TAP transporters, genes are separated from ditional factors: RFX5, RFX-AP and RFX-ANK. The genes
the class I genes they serve by the class III region, so this encoding these factors were identified by comparing panels of
functional integration may not be fully operational in our spe- rare BLS patients and mutant cell lines selected for loss of
cies. Another consideration is that each individual has two class II expression. This identified four complementation
haplotypes, the various components of which must cooperate groups. Transient heterokaryon fusions between these cell
to some extent. lines restored class II expression if the partners harboured
Recent data add weight to the notion of a high degree of defects in different components, thereby defining the four
allelic integration on individual haplotypes. It is becoming groups. A combination of genetic complementation and pro-
appreciated that, in addition to protein-coding loci, other poly- tein characterisation eventually identified the genetic lesions,
morphic genomic features are embedded in the MHC com- as reviewed in Reith and Mach (2001).
plex. One clue to this came from the finding that a single It was only in the last decade that a master transcription
nucleotide polymorphism (SNP) 35 kb upstream of HLA-C regulator for class I was discovered (Kobayashi and Elsen,
associated with levels of mRNA transcript and cell-surface 2012; Meissner et al. 2010; Neerincx et al. 2013). Like the
expression (Kulkarni et al. 2011). Binding of microRNA class II transcription activator, CIITA, NLRC5 is an NLR
hsa-miR-148 to a variable 3′ untranslated region in HLA-C protein and the two molecules share some homology. In cell
genes regulated their expression. The conclusion was that ex- lines defective in class I expression, such as mouse melanoma
pression levels may be regulated both by cis- and trans-acting B16F10, over-expressed NLRC5 could restore class I
168 Immunogenetics (2019) 71:161–170

expression. In human cells, over-expression of NLRC5 result- Open Access This article is distributed under the terms of the Creative
Commons Attribution 4.0 International License (http://
ed in induction of non-classical class I genes HLA-E, -F and - creativecommons.org/licenses/by/4.0/), which permits unrestricted use,
G, which normally exhibit a restricted tissue expression. distribution, and reproduction in any medium, provided you give appro-
CIITA-dependent activation of HLA class II requires the priate credit to the original author(s) and the source, provide a link to the
enhanceosome complex, which binds to a motif, SXY, in the Creative Commons license, and indicate if changes were made.
promoter of the gene. NLRCF probably interacts with a sim-
ilar module upstream of class I genes.
Confirmation of the role of NLRC5 in regulation of MHC
class I was provided by studying Nlrc5-deficient knock-out References
mice (Staehli et al. 2012). The data showed that MHC class I
Accolla RS (1983) Human B cell variants immunoselected against a
was downregulated, exhibiting the greatest effect in cells in single Ia antigen subset have lost expression of several Ia antigen
the immune system. However, treatment of these mice with subsets. J Exp Med 157:1053–1058
inflammatory stimuli, such as interferon or LPS, resulted in Apps R, Qi Y, Carlson JM, Chen H, Gao X, Thomas R, Yuki Y, Del Prete
significant class I expression, indicating that Nlrc5-indepen- GQ, Goulder P, Brumme ZL, Brumme CJ, John M, Mallal S, Nelson
G, Bosch R, Heckerman D, Stein JL, Soderberg KA, Moody MA,
dent mechanisms must exist to regulate MHC class I. Denny TN, Zeng X, Fang J, Moffett A, Lifson JD, Goedert JJ,
Recent work suggests that there is still more to learn about Buchbinder S, Kirk GD, Fellay J, McLaren P, Deeks SG, Pereyra
variation in HLA expression. For example, the Anderson lab- F, Walker B, Michael NL, Weintrob A, Wolinsky S, Liao W,
oratory identified an elaborate system regulating forms of Carrington M (2013) Influence of HLA-C expression level on
HIV control. Science 340:87–91
HLA-C, which is specific to natural killer cells, and could
Badrinath S, Saunders P, Huyton T, Aufderbeck S, Hiller O, Blasczyk R,
relate to NK cell licencing or education (Li et al. 2018). Bade-Doeding C (2012) Position 156 influences the peptide reper-
Other work suggests that, in addition to allelic differences, toire and tapasin dependency of human leukocyte antigen B*44
levels of class I should be taken into account in relation to allotypes. Haematologica 97:98–106
autoimmunity, infection and transplantation (Apps et al. 2013; Benacerraf B, McDevitt HO (1972) Histocompatibility-linked immune
response genes. Science 175:273–279
Kaur et al. 2017; Li et al. 2018; Petersdorf et al. 2014). Bjorkman PJ, Saper MA, Samraoui B, Bennett WS, Strominger JL,
Wiley DC (1987a) The foreign antigen binding site and T cell
recognition regions of class I histocompatibility antigens.
Nature 329:512–518
Screening for novel components Bjorkman PJ, Saper MA, Samraoui B, Bennett WS, Strominger JL, Wiley
DC (1987b) Structure of the human class I histocompatibility anti-
New screening techniques have the potential to identify novel gen, HLA-A2. Nature 329:506–512
factors involved in MHC biology. For example, forward ge- Blees A, Januliene D, Hofmann T, Koller N, Schmidt C, Trowitzsch S,
Moeller A, Tampe R (2017) Structure of the human MHC-I peptide-
netic screens involving retroviral insertional mutagenesis or
loading complex. Nature 551:525–528
random CRISPR/Cas9 targeting of a haploid human cell line Bloom BR, Salgame P, Diamond B (1992) Revisiting and revising sup-
KBM7 implicate TXNDC11 as a novel factor required for pressor T cells. Immunol Today 13:131–136
MHC class I endoplasmic reticulum-associated protein degra- Chen L, Fischer R, Peng Y, Reeves E, McHugh K, Ternette N, Hanke T,
dation (ERAD) (Timms et al. 2016). In addition, a genome- Dong T, Elliott T, Shastri N, Kollnberger S, James E, Kessler B,
Bowness P (2014) Critical role of endoplasmic reticulum aminopep-
wide RNAi screen was used to identify pathways regulating tidase 1 in determining the length and sequence of peptides bound
MHC class II antigen presentation (Paul et al. 2011). and presented by HLA-B27. Arthritis Rheumatol 66:284–294
Cho SG, Attaya M, Monaco JJ (1991) New class II-like genes in the
murine MHC. Nature 353:573–576
Dawkins R, Leelayuwat C, Gaudieri S, Tay G, Hui J, Cattley S, Martinez
Conclusion P, Kulski J (1999) Genomics of the major histocompatibility com-
plex: haplotypes, duplication, retroviruses and disease. Immunol
Immunogenetics has played a predominant role in uncovering Rev 167:275–304
many of the fundamentals of antigen processing and presen- de la Salle H, Hanau D, Fricker D, Urlacher A, Kelly A, Salamero J,
Powis SH, Donato L, Bausinger H, Laforet M et al (1994)
tation. In this review, we have sketched out some of the Homozygous human TAP peptide transporter mutation in HLA
history and background to the genetics, focussing mainly class I deficiency. Science 265:237–241
on the human system. Other contributions to this volume of DeMars R, Chang CC, Shaw S, Reitnauer PJ, Sondel PM (1984)
Immunogenetics may be consulted for more in-depth cov- Homozygous deletions that simultaneously eliminate expressions
of class I and class II antigens of EBV-transformed B-
erage of individual components of antigen processing and lymphoblastoid cells. I. Reduced proliferative responses of autolo-
presentation. gous and allogeneic T cells to mutant cells that have decreased
expression of class II antigens. Hum Immunol 11:77–97
Funding information JT is supported by the European Research Council Denzin LK, Sant'Angelo DB, Hammond C, Surman MJ, Cresswell P
under the European Union’s Horizon 2020 research and innovation pro- (1997) Negative regulation by HLA-DO of MHC class II-
gramme (Grant agreement No. 695551). restricted antigen processing. Science 278:106–109
Immunogenetics (2019) 71:161–170 169

Deverson EV, Gow IR, Coadwell WJ, Monaco JJ, Butcher GW, Howard Kobayashi KS, van den Elsen PJ (2012) NLRC5: a key regulator of MHC
JC (1990) MHC class II region encoding proteins related to the class I-dependent immune responses. Nat Rev Immunol 12:813–
multidrug resistance family of transmembrane transporters. Nature 820
348:738–741 Kronenberg M, Steinmetz M, Kobori J, Kraig E, Kapp JA, Pierce CW,
Garstka MA, Fritzsche S, Lenart I, Hein Z, Jankevicius G, Boyle LH, Sorensen CM, Suzuki G, Tada T, Hood L (1983) RNA transcripts
Elliott T, Trowsdale J, Antoniou AN, Zacharias M, Springer S for I-J polypeptides are apparently not encoded between the I-A and
(2011) Tapasin dependence of major histocompatibility complex I-E subregions of the murine major histocompatibility complex.
class I molecules correlates with their conformational flexibility. Proc Natl Acad Sci U S A 80:5704–5708
FASEB J 25:3989–3998 Kulkarni S, Savan R, Qi Y, Gao X, Yuki Y, Bass SE, Martin MP, Hunt P,
Glynne R, Powis SH, Beck S, Kelly A, Kerr LA, Trowsdale J (1991) A Deeks SG, Telenti A, Pereyra F, Goldstein D, Wolinsky S, Walker B,
proteasome-related gene between the two ABC transporter loci in Young HA, Carrington M (2011) Differential microRNA regulation
the class II region of the human MHC. Nature 353:357–360 of HLA-C expression and its association with HIV control. Nature
Goodfellow PN, Jones EA, Van Heyningen V, Solomon E, Bobrow M, 472:495–498
Miggiano V, Bodmer WF (1975) The beta2-microglobulin gene is Larhammar D, Hammerling U, Rask L, Peterson PA (1985) Sequence of
on chromosome 15 and not in the HL-A region. Nature 254:267– gene and cDNA encoding murine major histocompatibility complex
269 class II gene A beta 2. J Biol Chem 260:14111–14119
Green DR, Flood PM, Gershon RK (1983) Immunoregulatory T-cell Li H, Ivarsson MA, Walker-Sperling VE, Subleski J, Johnson JK, Wright
pathways. Annu Rev Immunol 1:439–463 PW, Carrington M, Bjorkstrom NK, McVicar DW, Anderson SK
Guce AI, Mortimer SE, Yoon T, Painter CA, Jiang W, Mellins ED, Stern (2018) Identification of an elaborate NK-specific system regulating
LJ (2013) HLA-DO acts as a substrate mimic to inhibit HLA-DM by HLA-C expression. PLoS Genet 14:e1007163
a competitive mechanism. Nat Struct Mol Biol 20:90–98
Liljedahl M, Kuwana T, Fung-Leung WP, Jackson MR, Peterson PA,
Hermann C, van Hateren A, Trautwein N, Neerincx A, Duriez PJ,
Karlsson L (1996) HLA-DO is a lysosomal resident which requires
Stevanovic S, Trowsdale J, Deane JE, Elliott T, Boyle LH (2015)
association with HLA-DM for efficient intracellular transport.
TAPBPR alters MHC class I peptide presentation by functioning as
EMBO J 15:4817–4824
a peptide exchange catalyst. Elife 4
Horowitz A, Djaoud Z, Nemat-Gorgani N, Blokhuis J, Hilton HG, Beziat Meissner TB, Li A, Biswas A, Lee KH, Liu YJ, Bayir E, Iliopoulos D,
V, Malmberg KJ, Norman PJ, Guethlein LA, Parham P (2016) Class van den Elsen PJ, Kobayashi KS (2010) NLR family member
I HLA haplotypes form two schools that educate NK cells in differ- NLRC5 is a transcriptional regulator of MHC class I genes. Proc
ent ways. Sci Immunol 1:eaag1672 Natl Acad Sci U S A 107:13794–13799
Horton R, Gibson R, Coggill P, Miretti M, Allcock RJ, Almeida J, Forbes Mellins E, Arp B, Ochs B, Erlich H, Pious D (1988) A single amino acid
S, Gilbert JG, Halls K, Harrow JL, Hart E, Howe K, Jackson DK, substitution in the human histocompatibility leukocyte antigen DR3
Palmer S, Roberts AN, Sims S, Stewart CA, Traherne JA, Trevanion beta chain selectively alters antigen presentation. J Exp Med 168:
S, Wilming L, Rogers J, de Jong PJ, Elliott JF, Sawcer S, Todd JA, 1531–1537
Trowsdale J, Beck S (2008) Variation analysis and gene annotation Mellins E, Smith L, Arp B, Cotner T, Celis E, Pious D (1990) Defective
of eight MHC haplotypes: the MHC haplotype project. processing and presentation of exogenous antigens in mutants with
Immunogenetics 60:1–18 normal HLA class II genes. Nature 343:71–74
Jiang J, Natarajan K, Boyd LF, Morozov GI, Mage MG, Margulies DH Monaco JJ, Cho S, Attaya M (1990) Transport protein genes in the mu-
(2017) Crystal structure of a TAPBPR-MHC I complex reveals the rine MHC: possible implications for antigen processing. Science
mechanism of peptide editing in antigen presentation. Science 358: 250:1723–1726
1064–1068 Monaco JJ, McDevitt HO (1984) H-2-linked low-molecular weight poly-
Karre K, Ljunggren HG, Piontek G, Kiessling R (1986) Selective rejec- peptide antigens assemble into an unusual macromolecular com-
tion of H-2-deficient lymphoma variants suggests alternative im- plex. Nature 309:797–799
mune defence strategy. Nature 319:675–678 Morris P, Shaman J, Attaya M, Amaya M, Goodman S, Bergman C,
Kasahara M (1999) The chromosomal duplication model of the major Monaco JJ, Mellins E (1994) An essential role for HLA-DM in
histocompatibility complex. Immunol Rev 167:17–32 antigen presentation by class II major histocompatibility molecules.
Kaufman J (2015) Co-evolution with chicken class I genes. Immunol Rev Nature 368:551–554
267:56–71 Neerincx A, Boyle LH (2017) Properties of the tapasin homologue
Kaur G, Gras S, Mobbs JI, Vivian JP, Cortes A, Barber T, Kuttikkatte SB, TAPBPR. Curr Opin Immunol 46:97–102
Jensen LT, Attfield KE, Dendrou CA, Carrington M, McVean G, Neerincx A, Castro W, Guarda G, Kufer TA (2013) NLRC5, at the heart
Purcell AW, Rossjohn J, Fugger L (2017) Structural and regulatory of antigen presentation. Front Immunol 4:397
diversity shape HLA-C protein expression levels. Nat Commun 8: Neerincx A, Hermann C, Antrobus R, van Hateren A, Cao H, Trautwein
15924 N, Stevanovic S, Elliott T, Deane JE, Boyle LH (2017) TAPBPR
Kavathas P, Bach FH, DeMars R (1980) Gamma ray-induced loss of bridges UDP-glucose:glycoprotein glucosyltransferase 1 onto MHC
expression of HLA and glyoxalase I alleles in lymphoblastoid cells. class I to provide quality control in the antigen presentation pathway.
Proc Natl Acad Sci U S A 77:4251–4255 Elife 6
Kelly A, Powis SH, Glynne R, Radley E, Beck S, Trowsdale J (1991a)
Norman PJ, Hollenbach JA, Nemat-Gorgani N, Marin WM, Norberg SJ,
Second proteasome-related gene in the human MHC class II region.
Ashouri E, Jayaraman J, Wroblewski EE, Trowsdale J, Rajalingam
Nature 353:667–668
R, Oksenberg JR, Chiaroni J, Guethlein LA, Traherne JA, Ronaghi
Kelly A, Powis SH, Kerr LA, Mockridge I, Elliott T, Bastin J, Uchanska-
M, Parham P (2016) Defining KIR and HLA class I genotypes at
Ziegler B, Ziegler A, Trowsdale J, Townsend A (1992) Assembly
highest resolution via high-throughput sequencing. Am J Hum
and function of the two ABC transporter proteins encoded in the
Genet 99:375–391
human major histocompatibility complex. Nature 355:641–644
Ortmann B, Copeman J, Lehner PJ, Sadasivan B, Herberg JA, Grandea
Kelly AP, Monaco JJ, Cho SG, Trowsdale J (1991b) A new human HLA
AG, Riddell SR, Tampe R, Spies T, Trowsdale J, Cresswell P (1997)
class II-related locus, DM. Nature 353:571–573
A critical role for tapasin in the assembly and function of multimeric
Klein J (1986) Natural history of the major histocompatibility complex.
MHC class I-TAP complexes. Science 277:1306–1309
Wiley, New York
170 Immunogenetics (2019) 71:161–170

Park B, Lee S, Kim E, Ahn K (2003) A single polymorphic residue within complex class II region controlling the class I antigen presentation
the peptide-binding cleft of MHC class I molecules determines spec- pathway. Nature 348:744–747
trum of tapasin dependence. J Immunol 170:961–968 Spies T, DeMars R (1991) Restored expression of major histocompatibil-
Paul P, van den Hoorn T, Jongsma ML, Bakker MJ, Hengeveld R, ity class I molecules by gene transfer of a putative peptide transport-
Janssen L, Cresswell P, Egan DA, van Ham M, Ten Brinke A, er. Nature 351:323–324
Ovaa H, Beijersbergen RL, Kuijl C, Neefjes J (2011) A genome- Spring B, Fonatsch C, Muller C, Pawelec G, Kompf J, Wernet P, Ziegler
wide multidimensional RNAi screen reveals pathways controlling A (1985) Refinement of HLA gene mapping with induced B-cell
MHC class II antigen presentation. Cell 145:268–283 line mutants. Immunogenetics 21:277–291
Petersdorf EW, Gooley TA, Malkki M, Bacigalupo AP, Cesbron A, Staehli F, Ludigs K, Heinz LX, Seguin-Estevez Q, Ferrero I, Braun M,
Du Toit E, Ehninger G, Egeland T, Fischer GF, Gervais T, Schroder K, Rebsamen M, Tardivel A, Mattmann C, MacDonald
Haagenson MD, Horowitz MM, Hsu K, Jindra P, Madrigal A, HR, Romero P, Reith W, Guarda G, Tschopp J (2012) NLRC5
Oudshoorn M, Ringden O, Schroeder ML, Spellman SR, Tiercy deficiency selectively impairs MHC class I-dependent lymphocyte
JM, Velardi A, Witt CS, O'Huigin C, Apps R, Carrington M, killing by cytotoxic T cells. J Immunol 188:3820–3828
I n t e r n a t i o n a l H i s t o c o m p a t i b i l i t y Wo r k i n g G r o u p i n Steimle V, Otten LA, Zufferey M, Mach B (1993) Complementation
Hematopoietic Cell T (2014) HLA-C expression levels define cloning of an MHC class II transactivator mutated in hereditary
permissible mismatches in hematopoietic cell transplantation. MHC class II deficiency (or bare lymphocyte syndrome). Cell 75:
Blood 124:3996–4003 135–146
Pious D, Dixon L, Levine F, Cotner T, Johnson R (1985) HLA class II Steinmetz M, Hood L (1983) Genes of the major histocompatibility com-
regulation and structure. Analysis with HLA-DR3 and HLA-DP plex in mouse and man. Science 222:727–733
point mutants. J Exp Med 162:1193–1207 Teng MS, Stephens R, Du Pasquier L, Freeman T, Lindquist JA,
Porter KM, Hermann C, Traherne JA, Boyle LH (2014) TAPBPR iso- Trowsdale J (2002) A human TAPBP (TAPASIN)-related gene,
forms exhibit altered association with MHC class I. Immunology TAPBP-R. Eur J Immunol 32:1059–1068
142:289–299 Timms RT, Menzies SA, Tchasovnikarova IA, Christensen LC,
Powis SJ, Deverson EV, Coadwell WJ, Ciruela A, Huskisson NS, Smith Williamson JC, Antrobus R, Dougan G, Ellgaard L, Lehner PJ
H, Butcher GW, Howard JC (1992) Effect of polymorphism of an (2016) Genetic dissection of mammalian ERAD through compara-
MHC-linked transporter on the peptides assembled in a class I mol- tive haploid and CRISPR forward genetic screens. Nat Commun 7:
ecule. Nature 357:211–215 11786
Powis SJ, Young LL, Joly E, Barker PJ, Richardson L, Brandt RP, Melief Tonnelle C, DeMars R, Long EO (1985) DO beta: a new beta chain gene
CJ, Howard JC, Butcher GW (1996) The rat cim effect: TAP allele- in HLA-D with a distinct regulation of expression. EMBO J 4:2839–
dependent changes in a class I MHC anchor motif and evidence 2847
against C-terminal trimming of peptides in the ER. Immunity 4: Townsend A, Ohlen C, Bastin J, Ljunggren HG, Foster L, Karre K (1989)
159–165 Association of class I major histocompatibility heavy and light
chains induced by viral peptides. Nature 340:443–448
Ramsuran V, Naranbhai V, Horowitz A, Qi Y, Martin MP, Yuki Y, Gao X,
Townsend AR, Gotch FM, Davey J (1985) Cytotoxic T cells recognize
Walker-Sperling V, Del Prete GQ, Schneider DK, Lifson JD, Fellay
fragments of the influenza nucleoprotein. Cell 42:457–467
J, Deeks SG, Martin JN, Goedert JJ, Wolinsky SM, Michael NL,
Tregaskes CA, Harrison M, Sowa AK, van Hateren A, Hunt LG, Vainio
Kirk GD, Buchbinder S, Haas D, Ndung'u T, Goulder P, Parham P,
O, Kaufman J (2016) Surface expression, peptide repertoire, and
Walker BD, Carlson JM, Carrington M (2018) Elevated HLA-A
thermostability of chicken class I molecules correlate with peptide
expression impairs HIV control through inhibition of NKG2A-
transporter specificity. Proc Natl Acad Sci U S A 113:692–697
expressing cells. Science 359:86–90
Trowsdale J, Hanson I, Mockridge I, Beck S, Townsend A, Kelly A
Reeves E, Colebatch-Bourn A, Elliott T, Edwards CJ, James E (2014a)
(1990) Sequences encoded in the class II region of the MHC related
Functionally distinct ERAP1 allotype combinations distinguish in-
to the ‘ABC’ superfamily of transporters. Nature 348:741–744
dividuals with ankylosing spondylitis. Proc Natl Acad Sci U S A
Trowsdale J, Kelly A (1985) The human HLA class II alpha chain gene
111:17594–17599
DZ alpha is distinct from genes in the DP, DQ and DR subregions.
Reeves E, Colebatch-Bourn A, Elliott T, Edwards CJ, James E (2015)
EMBO J 4:2231–2237
Reply to Robinson and Brown: it is the combination of ERAP1
Turner TR, Hayhurst JD, Hayward DR, Bultitude WP, Barker DJ,
allotypes that identifies individuals with ankylosing spondylitis.
Robinson J, Madrigal JA, Mayor NP, Marsh SGE (2018) Single
Proc Natl Acad Sci U S A 112:E1817
molecule real-time DNA sequencing of HLA genes at ultra-high
Reeves E, Elliott T, James E, Edwards CJ (2014b) ERAP1 in the patho- resolution from 126 international HLA and immunogenetics work-
genesis of ankylosing spondylitis. Immunol Res 60:257–269 shop cell lines. HLA 91:88–101
Reith W, Mach B (2001) The bare lymphocyte syndrome and the regu- van Ham SM, Tjin EP, Lillemeier BF, Gruneberg U, van Meijgaarden
lation of MHC expression. Annu Rev Immunol 19:331–373 KE, Pastoors L, Verwoerd D, Tulp A, Canas B, Rahman D,
Robinson PC, Brown MA (2015) ERAP1 biology and assessment in Ottenhoff TH, Pappin DJ, Trowsdale J, Neefjes J (1997) HLA-DO
ankylosing spondylitis. Proc Natl Acad Sci U S A 112:E1816 is a negative modulator of HLA-DM-mediated MHC class II peptide
Sakaguchi S, Wing K, Miyara M (2007) Regulatory T cells—a brief loading. Curr Biol 7:950–957
history and perspective. Eur J Immunol 37(Suppl 1):S116–S123 Williams AP, Peh CA, Purcell AW, McCluskey J, Elliott T (2002)
Salter RD, Howell DN, Cresswell P (1985) Genes regulating HLA class I Optimization of the MHC class I peptide cargo is dependent on
antigen expression in T-B lymphoblast hybrids. Immunogenetics 21: tapasin. Immunity 16:509–520
235–246 Zimmer J, Andres E, Donato L, Hanau D, Hentges F, de la Salle H (2005)
Shiina T, Inoko H, Kulski JK (2004) An update of the HLA genomic Clinical and immunological aspects of HLA class I deficiency. QJM
region, locus information and disease associations: 2004. Tissue 98:719–727
Antigens 64:631–649 Zinkernagel RM, Doherty PC (1974) Immunological surveillance against
Spies T, Bresnahan M, Bahram S, Arnold D, Blanck G, Mellins E, Pious altered self components by sensitised T lymphocytes in lymphocytic
D, DeMars R (1990) A gene in the human major histocompatibility choriomeningitis. Nature 251:547–548