Phil Dyer, Ann-Margaret Little and David Turner

Introduction

HLA: history of clinical application and technical development

Working independently in the mid-20th century, Rose Payne in California, Jon van Rood in Leiden and Jean Dausset in Paris detected antibodies reactive with non-self leucocytes in serum from multigravid women and from multitransfused patients. Linking of these findings to the understanding of the function of the immune system stimulated a rapid expansion of effort to define the immunogenetic basis of allosensitisation. In parallel, clinical transplant teams were attempting to overcome the immune response to allogeneic organ transplantation since they had established effective surgical techniques for organ implantation. The successful transplantation of a kidney between monozygotic twins, by Murray and colleagues in 1954 in Boston, USA, indicated that the alloimmune response is genetically encoded. The aspiration was to identify the gene(s) responsible for alloimmunity and subsequently to avoid their mismatching and so establish clinical tolerance of allogeneic tissues. Full tolerance has still to be achieved due to the plasticity of the immune response.

The culmination of six decades of HLA research is a complete resolution of the major histocompatibility complex of approximately 3.5 megabases of DNA containing at least 25 HLA gene loci coding for in excess of 7000 HLA allele sequences. This is the most polymorphic system in the human genome and exists because of the persistent infections that humans encounter throughout life (Fig. 4.1).

The elucidation of HLA polymorphisms was pursued to support matching, or minimising mismatching, for clinically relevant transplantation since this reduces T-cell activation and minimises the risk of acute cellular rejection. The detection and definition of antibodies to HLA protein allotypes in patients’ serum was soon shown to be an important component of histocompatibility.

To identify circumstances when HAR is likely to occur, Patel and Terasaki developed the complement-mediated lymphocytotoxicity assay (CDC).³ In outline, this assay comprises a two-stage incubation of donor lymphocytes, as a source of donor HLA proteins, with the potential recipient’s serum and addition of heterologous (rabbit) complement. The CDC assay detects lysis of the target donor lymphocytes through binding of HLA reactive antibody, present in recipient serum, with donor cell surface HLA proteins, mediated by complement-dependent cytotoxicity. When a CDC assay detected recipient serum lysis of donor lymphocytes (a ‘positive’ test), then HAR was shown to be highly likely to occur.⁴

The CDC assay was subsequently used to define the specificity of the HLA protein allotypes present on recipient and donor lymphocytes to establish the HLA phenotype – a process known at the time as ‘tissue typing’, a term that is no longer useful. This approach was dependent on the availability of sera containing well-defined reactivity to one, or more, HLA protein allotypes, as typing reagents. Again, the worldwide histocompatibility community collaborated by exchanging such sera, sourced from HLA alloimmunised individuals, most often gravid women. Despite significant challenges, HLA typing serology technology allowed identification of most “broad” HLA protein allotypes. Serological typing still has a role, primarily because it is a rapid assay and monospecific monoclonal antibodies are now available as commercial reagents, and it is used in some centres to support identification of organ donor HLA phenotypes alongside more comprehensive molecular biological testing.

In the early 1990s, the development of the polymerase chain reaction (PCR), which generates millions of identical copies of genomic DNA, revolutionised the identification of HLA alleles. A series of techniques was developed, with two remaining commonly used. Sequence-specific primers (SSPs) are short, artificially designed DNA sequence reagents that are used in a PCR reaction to generate copies of DNA with exclusivity for a unique (HLA) gene sequence. A battery of SSPs is designed and a corresponding number of PCR reactions are set up utilising DNA extracted from cells from the individual for whom an HLA phenotype is to be identified. Identification of positive PCR amplifications is visualised by gel electrophoresis and indicates presence of a specific polymorphic HLA gene sequence. Subsequently, HLA alleles can be allocated to the DNA tested and an HLA type to the individual. PCR-SSP is used to type for all HLA loci but the resolution achieved is sometimes limited.

The second current molecular HLA typing technique employs xMAP® technology, commercially available as the Luminex platform.⁵ In essence, commercially available kits of Luminex microbeads (5.6 μm), with inherent unique fluorochromasia, are coated with oligonucleotide probes that will hybridise exclusively to the DNA hypervariable region encoding specific HLA alleles. Multiple bead populations, each bearing known HLA-specific oligonucleotide probes, are identified in a multiplex assay by an internal and specific fluorochrome dye and visualised in a flow cytometer with laser light stimulation. Hybridisation to target HLA DNA sequences is revealed by a fluorochrome-tagged streptavidin conjugate that binds to biotin incorporated in the primers used to amplify the HLA locus of interest.

Conventional gene sequencing technologies can also be used to detect HLA gene polymorphisms, especially in the context of haemopoietic progenitor stem cell transplantation when complete identity between the stem cell donor and the patient HLA is preferable to limit rejection and graft versus host disease.

For the detection and definition of HLA reactive allosensitisation, histocompatibility testing laboratories have developed specific and sensitive assays, with the current technology of choice again employing the xMAP® technology. In essence, commercially available microbeads with multiple or single HLA proteins attached to their surface are incubated with patient serum and any HLA reactive antibody bound is detected by addition of a fluorochrome-labelled secondary antibody. The Luminex platform is used to identify individual microbeads by their inherent fluorochromasia and positive binding of antibody to HLA proteins on the beads by flow cytometry.

HLA genes and proteins: structure and genetics relevant to transplantation

There are two types of HLA genes within the human MHC: HLA class I and HLA class II. HLA class I genes encode a single-polypeptide heavy chain that associates with β-2-microglobulin to form an HLA class I protein. HLA class II genes encode both an α (DRA) and a β chain (DRB1, 3, 4, 5), which associate non-covalently to form an HLA class II protein. Both HLA class I and class II proteins are expressed at the cell surface and are bound to a short processed peptide within a peptide-binding site formed by the membrane distal domains of the HLA protein (Fig. 4.2).

HLA class I proteins are expressed on all nucleated cells and are also present on some non-nucleated cells such as platelets. HLA class II proteins have a restricted cell expression found on the surface of cells involved in antigen presentation, such as dendritic cells and macrophages. However, HLA class II protein expression can be up-regulated on other cell types, including endothelial cells, in response to stimulation by immune mediators such as γ-interferon at the time of immune activation.⁷

Three types of HLA class I proteins (HLA-A, HLA-B and HLA-C) and three types of HLA class II proteins (HLA-DR, HLA-DQ and HLA-DP) have a role in the generation of immune reactions that influence transplant outcome.

The HLA genes are the most polymorphic expressed genes within our genome. Polymorphism within HLA genes is not random and the variation found within DNA sequences is predominantly non-synonymous. Most of the protein variation observed impacts on the function of the HLA protein either by influencing the structure of peptides bound and/or influencing interactions with T-cell and natural killer cell receptors.

The well-defined WHO nomenclature system is key to the description of HLA allele and protein variation. For description, HLA allele names are broken down into numerical fields that are separated by colons (Fig. 4.3).

For organ transplantation, it is usually sufficient for donor and recipient typing to be at low resolution for HLA-A, -B, -C, -DRB1 and -DQB1. In the UK there is a requirement for histocompatibility laboratories to report deceased organ donor HLA types at this level to facilitate a consistent national process of organ allocation.

Not all defined HLA alleles occur at the same frequency.

The WHO system for assigning names to HLA alleles does not consider allele frequency and therefore efforts have been made to establish databases that contain information on the frequency of HLA alleles in different populations.Within a given population there will be a number of common HLA alleles and a number of rare HLA alleles. The organ allocation system within the UK defaults rare HLA types to their most common structurally related HLA type to reduce the bias in organs being provided to the patients with the most common HLA types. For example, DR9, which has a frequency of 2.6% in kidney patients, is structurally related and defaulted to DR4, which has a phenotype frequency of 29.8%, so that DR9-positive patients have access to the kidney donor pool more frequently.

HLA genes are inherited following Mendelian genetics and as they are co-located within the MHC, they are inherited ‘en bloc’ as a haplotype. It follows that any two siblings have a 1 in 4 chance of inheriting both the same or different HLA haplotypes from their parents.

Some combinations of HLA alleles are found on haplotypes at a higher frequency than would be predicted from random association of alleles.¹¹ This observation is called linkage disequilibrium and may result from one or more events, such as: