26 Immunogenetics: clinical and diagnostic aspects of the Human leukocyte antigen (HLA) system

Christian Seidl, Erhard Seifried

26.1 Genetic principles and immunological functions of the HLA system

The human leukocyte antigen (HLA) system corresponds to the human MHC (major histocompatibility complex) and comprises a group of genes with a key influence on the immunological differentiation between exogenous (non-self) and endogenous (self). Initial descriptions of this antigenic system are based on studies in mice conducted by Peter Gorer and George Schnell in the late 1930ies who already recognized the significance of the H-2 system for transplantation medicine /1, 2/.

The first human antigenic complex was described by Jean Dausset in 1958 using poly transfused patients’ sera. Initially called “Mac”, it is now known as HLA-A2 /3/. Additional types were detected by Jon van Rood (4a and 4b, corresponding to Bw4 and Bw6) later on.

It was early recognized that the HLA system is characterized by special antigenic polymorphism. This led to the first international workshop held in Durham, North Carolina/USA already in 1964 where numerous laboratories from around the world used standardized sera or cell samples to analyze various aspects of the HLA system. The results of the 13 international workshops to date are published in special volumes and provide an important basis of standardized investigation procedures for the determination of HLA types and the definition of clinically relevant functions /4/.

The resolution of the HLA molecular structure and characterization of rules for peptide presentation via HLA molecules has led to a deeper insight into the interaction between the complex of HLA molecule, presented peptide and T cell receptor /5, 6, 7, 8, 9/. The HLA molecule function of presenting extra- and intracellular peptides, and inducing a specific T cell reaction is of essential significance in the specific immune response regulation. This immune function of HLA molecules plays a key role in the identification of self and non self structures and was characterized as MHC (major histocompatibility complex) restriction by Peter Doherty and Rolf Zinkernagel /10/.

Molecular genetic methods allowing direct analysis of genetically encoded polymorphism have become increasingly established in the wake of sequence analysis and investigation of the gene encoding the HLA region. Contrary to serological techniques, these methods enable the detection of minute differences between the different HLA alleles and have proven themselves in clinical practice especially in the field of transplantation diagnostics and risk assessment in autoimmune disease diagnosis.

26.1.1 Structure and function of HLA molecules

The main function of HLA molecules is to present the antigenic peptides to T cells /6, 7/. At the same time, the binding of peptide fragments in the HLA binding groove is a major prerequisite for stable HLA heterodimers on the cell surface. The highly complex peptide binding mechanism occurs in the context of intracellular synthesis and peptide loading onto HLA molecules The peptides can represent self or non self structures, for example, following bacterial or viral infection. The presentation of self peptides by HLA molecules usually does not result in T cell stimulation because the relevant auto reactive T cells were eliminated in the thymus already during maturation (negative selection) or T cell stimulation is not possible due to the lack of a co stimulatory signal (peripheral anergy). As soon as non self peptides are expressed on an antigen presenting cell by HLA molecules, T cells are activated and all body cells presenting these non self peptides via their HLA molecules are eliminated.

HLA class I molecules preferably activate cytotoxic CD8⁺T cells, whereas HLA class II molecules mainly activate CD4⁺helper T cells or inflammatory CD4⁺T cells. This enables important functional aspects of HLA molecules in the regulation of T cell mediated immune response.

• Since T cells bind specifically to membrane bound HLA molecules using CD4 and/or CD8 molecules, they can only recognize non self peptides presented by the corresponding HLA molecules. They do not directly recognize soluble antigens. However, membrane bound antigen recognition is also used to direct the effector function of activated T cells to non self antigen presenting local target cells.
• HLA molecules and the antigens they present determine the type of T cell response. Extracellular antigens (e.g., bacteria) are primarily presented via HLA class II molecules, whereas intracellular antigens (e.g., viruses) from the cytosol are primarily presented via HLA class I molecules. Thus, the non self peptides are recognized by T cell populations exerting completely different functions.
• The immune response to a non self protein is mainly dependent on the capacity of HLA molecules to bind peptide fragments (i.e., fragments of this non self protein) in the antigen binding groove and present these fragments to T cells on the cell surface. However, the binding capacity and way of presenting antigenic peptide fragments of a non self protein vary between the different HLA alleles. Thus, HLA molecules may modulate the immune response to specific antigens.
• T cells directed against self antigens and/or HLA molecules are eliminated in the thymus by negative selection, whereas T cells that recognize non self peptides in the context of self HLA molecules proliferate to become mature T cells (positive selection). Thymus dependent T cell selection is enabled by direct contact between the HLA molecule, peptide and maturing T cell. Differences in the ability of different HLA alleles to present peptide fragments to T cells may influence the thymus dependent T cell selection.
• Natural killer cells (NK cells) are phylogenetically primitive, cytotoxic T cells which lack a T cell receptor and, therefore, act without HLA restriction. However, NK cells have receptors (killer cell immunoglobulin (Ig)-like receptors, KIRs) that have both an activating and inhibiting effect. Inhibition is mediated by the contact between KIR, receptors with a long cytoplasmic domain (domain long, DL) to binding motifs on the α₁-helix of HLA class I molecules. Three binding motifs are distinguished: HLA-C group 1 (Ser77/Asn80), HLA-C group 2 (Asn77/Lys80) and HLA-Bw4 (AS 77–83). Evidence of other binding interactions has been provided for the HLA-A3 and HLA-A11 types.

As a rule, the expression of HLA class I molecules on the cell surface protects the cells against destruction by NK cells. However, reduced expression or the lack of individual HLA class I molecules on the cell surface (e.g., following viral infection or in the presence of tumors) causes the cells to be eliminated by NK cells.

26.1.2 Genetics of the HLA system

The HLA system is located on the short arm of chromosome 6 (6p21.1–6p-21.3) and comprises 3800 kilo base pairs. It is organized into the class I, class II and class III regions (Fig. 26-1 – Genomic organization of the Major Histocompatibility Complex).

26.1.2.1 HLA class I region

The HLA class I region nearest the telomere comprises the gene loci for the classical transplantation antigens HLA-A, HLA-B and HLA-C expressed on all nucleated cells (Tab. 26-1 – Expression of HLA class I and class II types). The gene product of these loci is a polypeptide referred to as heavy α-chain that binds to β₂-microglobulin, a polypeptide encoded on chromosome 15 (15q21–15q22), to form a functionally active HLA molecule. Moreover, the HLA class I gene loci express HLA-E, HLA-F, HLA-G and HLA-H α-chains.

These HLA genes are only transcribed in some tissues, causing specific immune regulation based on the interaction with NK cell receptors. HLA-G is expressed on fetal placental cells that invade the uterus wall. These cells do not express classical HLA class I molecules and, therefore, cannot be recognized as non self. However, contrary to other cells (e.g., tumor cells) that do not express HLA class I molecules, they are not eliminated by NK cells because HLA-G acts as a ligand for the inhibitory receptor ILT-2 on the NK cells. Thus, the HLA-G molecule may play a significant role in the tolerance to the semi allogenic fetus by the maternal immune system /11/. HLA-E, another HLA class I molecule, primarily binds peptides stemming from leader peptides of other HLA class I molecules. The NK cell receptor NKG2A which is integrated in the surface complex CD94 functions as a ligand for the peptide/HLA-E complex. NKG2A has an inhibitory functions and inhibits the stimulation of the cytotoxic activity of NK cells, thus preventing HLA-E expressing cells from being eliminated.

The MIC group, another gene group of the HLA class I region, is located between the HLA-B locus and the HLA-C locus. These genes only have very few structural characteristics in common with the classical HLA class I genes and have different regulation patterns. Two of the five MIC genes, MICA and MICB, are expressed; MICC/D/E are pseudo genes. Contrary to the classical HLA class I genes, they are expressed in fibroblasts and epithelial cells, primarily on intestinal epithelial cells.

Cellular distress (UV light irradiation, heat shock, oxidative stress, carcinogens, infection with Herpesvirus, M. tuberculosis or E. coli) may induce MHC molecule expression on the cell surface. MIC molecules act as ligands for the NK cell receptor NKG2D. More than 50 MICA and 5 MICB alleles have been detected to date. Increased ligation of NKG2D receptors triggered by induced MIC expression leads to NK cell activation. Thus, an NK cell can eliminate a cell expressing MHC class I molecules /12/.

The HFE gene locus is another interesting site in the HLA class I region. The HFE gene also transcribes an HLA class I like α-chain. Individual point mutations in the HFE gene may cause the clinical picture of hereditary hemochromatosis /13/ (Tab. 7.1-5 – Heritable forms of systemic iron overload).

Besides the HLA class I gene loci that express functionally intact HLA molecules, there are numerous gene loci that manifest themselves as pseudo genes or gene fragments and are not expressed (for example HLA-J, HLA-K, HLA-L).

26.1.2.2 HLA class II region

The HLA class II region (formerly referred to as HLA-D region) is subdivided into three subregions: HLA-DR, HLA-DQ and HLA-DP. Each subregion comprises at least one functional α-(A) gene and a β-(B) gene which encode the α-polypeptide chain and the β-polypeptide chain, respectively, of the functional active HLA class II molecule (DR, DQ, DP). The tissue expression of the HLA class II molecules is markedly restricted compared to the HLA class I molecules (Tab. 26-1 – Expression of HLA class I and class II types). HLA class II are constitutively expressed on antigen presenting cells such as B lymphocytes, monocytes, macrophages, dendritic cells and some epithelial and endothelial cells. However, the expression of HLA class II molecules can be induced in most cells (e.g., following stimulation by IFN-γ or IL-4).

Within the HLA-DR region, the number of HLA-DRB genes on a chromosome is not constant, but may vary corresponding to the haplotype constellation (Fig. 26-1 – Genomic organization of the Major Histocompatibility Complex (MHC)). The HLA-DRB1 gene is present in each individual and, together with the HLA-DRA gene, encodes the HLA-DR1 to HLA-DR18 specificities. The HLA-DRB3 gene is present in haplotypes expressing DR11, DR12, DR13, DR14, DR1403, DR1404, DR17 and DR18 and, together with HLA-DRA, encodes the DR52 specificity. The HLA-DRB4 gene is found in the DR4, DR7 and DR9 haplotypes and, together with DRA, encodes the DR53 specificity. The HLA-DRB5 gene, on the other hand, is present in DR15 and DR16 haplotypes and rarely also in the haplotypes of the DR1 group and, together with HLA-DRA, encodes the DR51 specificity. The HLA-DRB1, HLA-DRB3, HLA-DRB4 and HLA-DRB5 genes are the only genes to express β-chains, while the other HLA-DRB genes are pseudo genes. The HLA-DRB1 gene locus is characterized by marked polymorphism, whereas the HLA-DRB3, HLA-DRB4 and HLA-DRB5 gene loci are polymorphic to a lesser extent.

The HLA-DQ region includes two α- and two β-chain genes: HLA-DQA1, HLA-DQB1, HLA-DQA2 and HLA-DQB2 as well as a shortened β-chain gene (HLA-DQB3). Functionally active peptides forming the HLA-DQ molecule are only expressed by HLA-DQA1 and HLA-DQB1. The other gene loci are pseudo genes. Both HLA-DQA1 and HLA-DQB1 are highly polymorphic. The HLA-DQ antigen serotypes are thought to be located on the β-chain. The HLA-DP region comprises 4 gene loci: two α-chain genes, HLA-DPA1 and HLA-DPA2, as well as two β-chain genes, HLA-DPB1 and HLA-DPB2. Analogously to the HLA-DQ region, HLA-DPA1 and HLA-DPB1 are the only gene loci expressing gene products that form the HLA DP molecule. HLA-DPA2 and HLA-DPB2 are pseudo genes.

The gene loci for the HLA-DR, HLA-DQ and HLA-DP types are next to structurally related gene loci in the HLA class II region referred to as DOB, DNA, DMA and DMB, of which DMA and DMB express α- and β-polypeptide chains, respectively. These form a membrane bound molecule that assumes a key function in the intracellular association of peptide and HLA class II molecule during antigen processing /14/.

The TAP genes, which are of functional significance in antigen processing, are also positioned in the HLA class II region. The TAP1 and TAP2 (TAP, transporter associated with antigen processing) gene products form a heterodimeric, membrane bound molecule that allows processed antigens to be transported to the endoplasmic reticulum.

26.1.2.3 HLA class III region

A section referred to as HLA class III region which comprises a multitude of genes with diverse functions is located between the HLA class II and HLA class I regions. No functional or structural similarities have been established between the gene products of HLA class III and those of HLA class I and class II. However, some of these genes may be functionally active in the context of immune response, for example, the genes for complement factors C2, C4 and BF, genes for TNF-α and lymphotoxin (LTα and LTβ) as well as the gene for the heat shock protein 70. The genes encoding C21-hydroxylase are located between the genes C4A and C4B. Homozygosity for deletions and duplications of the C21-B gene locus may cause adrenogenital syndrome (AGS).

26.1.2.4 HLA polymorphism

The HLA-A, HLA-B and HLA-C gene loci of the HLA class I region and the majority of the HLA class II gene loci are characterized by a high degree of polymorphism.

Refer to:

• Tab. 26-2 – Table of HLA class I alleles
• Tab. 26-3 – Table of HLA class II alleles (DRB1).

Molecular genetic analyses of the genomic sequence differences of individual HLA gene loci at DNA level are used to characterize the marked polymorphism of the HLA system. The formerly dominating serological methods of antigen typing now tend to be only used for supplementary analysis within the scope of tissue compatibility tests in blood stem cell transplantation, for example for excluding null alleles that cannot be determined with molecular biological methods /15/. In order to differentiate between molecular genetic and conventional serological methods, the nomenclature specifies the gene locus, followed by an asterisk (*)-separated numerical code, with the first two digits representing the main groups and/or types and the following digits standing for the subgroups and/or subtypes. Further digits specify the sequence differences between alleles that do not cause a change in the amino acid sequence or represent intron variations or non-expressed alleles (null alleles, N) (Tab. 26-4 – Revised nomenclature for HLA serotypes and types determined with molecular biological methods).

As the number of detected alleles is steadily growing, the nomenclature has been revised to introduce a colon (:) for separating the different digit fields (for example: HLA-A*0101 = HLA-A*01:01). Furthermore, the "w" (originally standing for "workshop") has been removed in the molecular biological designation of the HLA-C locus allele (for example: Cw*0103 = C*01:03) but continues to be used in serological determination (antigen designation) for distinction from the complement system (for example: HLA-Cw3).

HLA molecules are co dominantly inherited and expressed. Each individual has not more than two alleles at every HLA gene locus. The entire complex of expressed HLA alleles is referred to as the HLA phenotype. Thus, the HLA phenotype reflects the genetically present allele configuration of the individual HLA gene loci existing on the paternal and maternal chromosome (Fig. 26-2 – Co dominant expression of HLA class I and class II molecules on the surface of cells). Accordingly, the chromosomal presence of HLA types at several HLA gene loci (e.g., HLA-A, HLA-B, HLA-DRB1) is referred to as HLA haplotype, whereas the gene configuration of the two haplotypes is referred to as HLA genotype. The recombination frequency of the HLA system is below 3% in this context /16, 17/. Within a family, the chromosomal factors (haplotypes) are transmitted on from the parents to their children according to Mendel’s laws (Fig. 26-3 – Segregation of HLA haplotypes within a family). Based on two haplotypes in an individual (i.e., not more than 4 haplotypes from the parents two from the father and two from the mother), each parent transmits one haplotype to the children at random configuration. As each child inherits one haplotype from the father and one from the mother, there is a 25% probability that two siblings within a family share identical paternal and maternal haplotypes.

Due to their identical genotypes, the siblings have identical HLA types and, thus, maximum immunological conformance of their HLA systems. The parents only share one haplotype with their children and are HLA haploidentical. HLA conformance is advantageous, but not mandatory in organ transplantation, whereas siblings with identical HLA genotypes are an ideal donor/recipient pair for blood stem cell transplantation.

The high number of alleles at the different HLA gene loci leads to a multitude of possible allele combinations in the population. As a result, unrelated individuals often have different HLA phenotypes.

Another characteristic of the HLA system is the higher incidence of some HLA types in a given population compared to other HLA types. Moreover, there is a marked linkage disequilibrium between the different HLA gene loci (i.e., certain alleles of a gene locus occur in combination with the corresponding alleles of another gene locus more often than would be expected based on the frequency of the individual haplotypes). The deviation of the expected frequency of two HLA haplotypes from the observed is a quantity called delta (Δ) value and expresses the degree of the linkage disequilibrium between two HLA haplotypes /18/.

The distribution of the linkage disequilibrium among different HLA haplotypes can also be dependent on a given population (e.g., European or Oriental). This results in different probabilities regarding the presence of specific HLA haplotype combinations within a population expressed as haplotype frequencies. The differences in HLA haplotype frequency most likely are based on the natural selection of certain HLA haplotypes that allow preferred presentation of pathogens. However, population specific differences in antigen frequency can also influence the extent of disease association or the likelihood of finding an HLA compatible, unrelated blood stem cell donor or HLA compatible unrelated platelet donor.

26.1.3 Molecular structure of HLA class I and class II molecules

Crystallographic determination has provided a precise understanding of the structure and corresponding functional aspects of HLA class I and class II molecules (Fig. 26-4 – HLA class I and HLA class II molecule structure and gene conformation) /5, 8/.

HLA class I molecules consist of a glycosylated, heavy polypeptide chain with a molecular weight of 45 kDa which is referred to as α-chain and is non covalently bound to β₂-microglobulin, a polypeptide with a molecular weight of 12 kDa (Fig. 18.12-1 – HLA antigens on the cell membrane of nucleated cells). The α-chain of the HLA class I molecule consists of three extracellular domains, the α₁- (N-terminal), α₂- and α₃-domains as well as a transmembrane (TM) region and a cytoplasmic (ZP) region. This structural organization of the molecules is also found in the intron/exon regions of the HLA class I gene. Each of the three extracellular domains comprises approximately 90 amino acids. Each α₂- and α₃-domain has a disulfide bond that stabilizes a loop consisting of 63 (α₂) and 86 (α₃) amino acids. The β-sheet structure of the α₃-domain and the β₂-microglobulin show marked homology with the constant regions of immunoglobulins. The β₂-microglobulin which also has a disulfide bond attaches between the α₂- and α₃-domains, stabilizing the tertiary structure of the membrane bound heterodimer. In this conformation, the β₂-microglobulin is completely extracellular, causing the HLA class I molecule to be anchored to the cell membrane only through the α-chain. The α₃-domain of the α-chain is the main binding site for the CD8 molecule of the T cell receptor.

In contrast, the α₁- and α₂-domains provide the peptide binding groove of the HLA molecule and the contact site for the T cell receptor. The bottom of the peptide binding groove consists of a platform of 8 anti parallel β-strands topped by two anti parallel α-helices that form the sides of the peptide binding groove. The ends of the HLA class I peptide binding groove are relatively close to one another and only allow to fit in short peptide fragments of 8–11 amino acids (predominantly 9 amino acids). The peptide binding groove contains the majority of the allelic amino acid variations and includes a number of pockets and irregularities that interact with amino acid residues of the bound peptides. Thus, alterations in the amino acid sequence of the peptide binding groove caused by allelic variations of the HLA molecules can lead to changes in peptide binding affinity, influencing the pattern of presented peptides and, consequently, the nature of immune response.

This simple interaction between the HLA peptide binding groove and the peptide (MHC restriction) represents a fundamental structural mechanism in the genetic regulation of immune response.

HLA class II molecules also have a heterodimeric structure where a heavy α-chain (30–35 kDa) and a light β-chain (26–29 kDa) are non covalently bound /4/. The difference in molecular weight of the HLA class II α/β-chains is primarily due to different glycosylation patterns. Each chain has two extracellular domains (α₁, α₂ and β₁, β₂), a transmembrane region and a cytoplasmic region which can also be found in the intron/exon structure. The membrane bound α₂- and β₂-domains correspond to the α₃-domain and the β₂-microglobulin of the HLA class I molecule and stabilize the conformation of the heterodimer. Both domains (α₂ and β₂) are similarly structured as the constant regions of immunoglobulins.

Unlike the HLA class I molecules, both the β₂- and the β₁-domains are involved in the binding with the CD4 molecule of the C-cell receptor. The peptide binding groove of the HLA class II molecule is formed by the α₁- and β₁-domains and, analogously to the HLA class I molecules, consists of a β-sheet structure topped by two α-helices. Contrary to the HLA class I molecules, however, the HLA class II binding groove has two open ends and allows the binding of longer peptide fragments (11–25 amino acids), with the peptide ends projecting from both ends of the peptide binding groove. As in the HLA class I molecules, the highest number of allelic variations in the amino acid sequence occurs in the peptide binding groove of the HLA class II molecules.

26.1.4 Antigen processing and presentation

The attachment of peptides to HLA molecules is an essential prerequisite for the stability of the HLA heterodimer and already takes place inside the cell during the formation of the HLA molecule. Both the HLA class I and HLA class II molecules utilize a highly complex mechanism for this purpose. This mechanism enables the generation of peptides for peptide fragments of different length depending on the binding capacity of the different HLA molecules. The generated peptides can then attach to the HLA binding groove. Moreover, due to this mechanism, the antigen processing steps are separated within the cell for the different HLA molecule classes /13, 14/. Because of this complex, intracellular antigen processing mechanism of endogenous and exogenous peptides and subsequent antigen presentation, the HLA molecules play a key role in the permanent monitoring of intracellular and extracellular space.

Intracellular, mostly cytoplasmic or nuclear proteins that are predominantly presented by HLA class I molecules are processed in the cytoplasm by way of enzymatic degradation through multi catalytic proteinase complexes known as proteasomes. The peptide fragments generated by the proteasomes are then transferred to the endoplasmic reticulum (ER)by way of ATP-dependent active transport with TAP1/2 heterodimers. In the ER, the HLA molecule is loaded with a peptide. The resulting HLA/peptide complex is transported to the cell surface where it is expressed. Due to the combination of protein fragmentation by proteasomes and peptide transport by TAP molecules, the predominant part of the peptides presented by HLA class I molecules stems from the cytoplasm of the cell. This explains, for example, why most viruses and numerous tumor antigens are expressed by HLA class I molecules.

Proteins can also be transferred to the cytoplasm through a special endocytic mechanism available, for example, in macrophages or dendritic cells, and thus be presented by HLA class I molecules.

Contrary to HLA class I molecules, HLA class II molecules mainly bind peptide fragments of extracellular proteins which enter the cell via endocytosis. While these extracellular proteins are transported via the endosomal compartment to the lysosomal compartment, they are degraded by various proteolytic enzymes. In this context, the binding of peptides with HLA class II molecules primarily takes place in the lysosomal compartment and only to a small extent in the endosomal compartment.

Similarly to the HLA class I molecules, only the peptides ensuring maximum possible stability of the heterodimer bind to the binding groove of the HLA class II molecule. The peptide loaded HLA class II molecules are then expressed at the cell surface and the attached peptide fragments are presented to T cells. Besides their main task of processing and presenting extracellular peptides, HLA class II molecules are also capable of presenting cytoplasmic proteins that enter the endocytic process.

The capacity of HLA class I and class II molecules to present both extracellular and intracellular peptides provides additional protection against various forms of pathogens and indicates the special significance in connection with the regulation of antigen processing and presentation for immune response control.

26.1.5 Soluble HLA antigens

In contrast to the elucidated membrane bound HLA molecules, much less is known about the function of soluble HLA class I (sHLA class I) and HLA class II (sHLA class II) antigens. Besides by proteolysis and membrane shedding, sHLA antigens are generated by alternative splicing from RNA transcripts /19, 20/.

Various studies have proposed an immunomodulatory role for sHLA antigens /21/. It has been shown, for instance, that sHLA class II antigens may interfere with HIV-1 infection. Soluble HLA class II antigens may also cause anergy of auto reactive T cells or induce T cell apoptosis, thus possibly leading to tolerance induction in autoimmune disease. Similarly, sHLA class I antigens can induce apoptosis in allo reactive cytotoxic T cells. Elevated serum levels of sHLA class I antigens were found to be associated with graft-versus-host disease /22/ and episodes of acute rejection after organ transplantation /23, 24, 25/. Soluble HLA class I antigens also bind to NK-cell receptors, thus inhibiting the lytic activity of these cells /26, 27/. Changes in sHLA antigens have also been found in various autoimmune diseases /28/. However, clinical significance of the soluble HLA antigens for diagnosis has not been established.

26.1.6 HLA and pharmacovigilance (abacavir induced hypersensitivity syndrome)

Patients treated with the virustatic agent NRTI abacavir for the first time within the scope of an HIV infection show severe immunopathological hypersensitive response in up to 5% of the cases /29, 30/. Symptoms occur between a few days and 6 weeks after start of therapy and can become life-threatening if therapy is continued. Side effects known as abacavir induced hypersensitivity syndrome occur in approximately 5% of the patients. A diagnostic in vitro test is not available, but the hypersensitivity reaction to abacavir is strongly associated with the presence of the HLA-B*5701 allele.

According to the information available to date, an immunopathological hypersensitive response can be excluded based on negative test results for HLA-B*57:01 because the hypersensitivity reaction to abacavir only occurs in carriers of the HLA-B*57:01 allele. Administration of abacavir is contraindicated in positive evidence of HLA-B*57:01. Hence, testing for HLA-B*57:01 should be performed in the context of initial examination or before therapy is switched to a regime including abacavir.

26.2 HLA molecules in laboratory diagnosis of autoimmune disease

The cause of autoimmune reactions is partly complex and not fully understood. Environmental and hereditary aspects play a role in many cases. Interestingly, a number of autoimmune diseases are clearly associated with certain HLA alleles (Tab. 26-5 – Association between HLA types and diseases (selection)). Autoimmune disease is characterized by long lasting stimulation of auto reactive T cells leading to the elimination of the cells carrying the self antigen and ultimately resulting in tissue damage.

Various mechanisms have been discussed regarding the occurrence of auto reactive T cells in the context of HLA molecules (see Section 25.1 – Autoantibodies in the diagnosis of diseases):

• Inadequate negative selection of auto reactive T cells during thymus development. Auto reactive T cells are usually eliminated by contact with HLA molecules and self antigen during thymus development (Fig. 21.1-7 – Selection of T cells that migrate into the thymus from the bone marrow). However, allelic changes in the structure of HLA molecules may reduce the affinity of T cells to the HLA molecule and/or result in inadequate presentation of self peptides.
• Allele specific changes in the HLA peptide binding groove and the resulting changes in the selection of peptide fragments may lead to the presentation of peptide fragments of invading pathogens which, together with certain HLA types, influence the activity of regulatory T cells that normally prevent autoimmunity.
• Structural similarities between pathogen and HLA molecules may lead to autoimmune disease caused by infection. This mechanism is referred to as molecular mimicry.
• Changes in the membrane expression of HLA molecules may change the balance between Th1 and Th2 cells and the cytokines excreted from these cells. In fact, there are allele specific differences in the conserved sequences in the promoter region of HLA molecules that may cause a different transcription rate. Such mechanisms can be found, for example, for HLA-DRB1 and HLA-DQB1 alleles. Besides functional aspects of HLA molecules, certain autoimmune diseases may also be caused by a linkage disequilibrium between some disease associated HLA types and pathogenically relevant genes (for example, HLA-A3 associated with hemochromatosis, see also Section 26.2.1 – Statistical methods for the assessment of disease association).

26.2.1 Statistical methods for the assessment of disease association

Disease association is determined by comparing the frequency of an allele between patients and healthy controls /31/. In this approach, the number of patients and controls positive for an allele is compared with the chi square test (X²-test based on four field or multi field tables). As a rule, a four field table of the absolute frequency is prepared for each patient and control allele analyzed.

The X²-test expresses the deviation from the expected values, assuming there is no association. The values should in no case be below five and the total of frequencies should be above 50. The degree of freedom of the test is 1 (DoF = 1) for the four-field table. X² is calculated using the following formula:

a, number of patients with the allele; b, number of patients without the allele; c, number of controls with the allele; d, number of controls without the allele; N, total of patients and controls

For numbers of cases below 50, it is recommended to apply the Yates correction to make the X²-value more conservative, according to the formula:

a, number of patients with the allele; b, number of patients without the allele; c, number of controls with the allele; d, number of controls without the allele; N, total of patients and controls

The values obtained for X² can be classified under significance levels (p; probability values). The following applies to uncorrected p values:

Due to the random possibility of significant deviation (type I error), p values (p_u-uncorrected) should always be corrected (p_c-corrected) for the number of comparisons performed, according to the formula:

p_c = 1 – (1 – p_u)ⁿ

n is the number of comparisons (e.g., the number of existing HLA-DRB1 alleles).

The strength of association is determined by calculating the relative risk (RR) and/or the odds ratio (OR) at 5–95% confidence intervals /32, 33/. The relative risk denotes a risk factor for an individual carrying a disease associated allele (e.g., HLA-DRB1*04) to become diseased compared to an individual without this allele. The relative risk is calculated according to Woolf’s formula /34/:

The etiologic fraction (EF) or preventive fraction (PF) can be additionally calculated to estimate the extent to which the analyzed factor contributes to the genetic predisposition of a disease in the population sample /34/. EF values < 0.0 to not more than 0.99 indicate the contribution of the genetic component to the manifestation of the relevant disease; PF values < 1.0 to 0.0 indicate the positive effect of the HLA antigen regarding disease manifestation. Besides calculating the EF and PF values for estimating

the association strength, the positive predictive value (PPV) and/or the negative predictive value (NPV) can be determined when analyzing two or more gene loci, according to the following formula /35/:

26.2.2 Genetic hemochromatosis and HFE gene diagnostics

The HFE 1 gene is located in the major histocompatibility complex on chromosome 6. It has been shown by cloning and sequence analysis that gene variations of the HFE 1 are correlated with predisposition to hereditary hemochromatosis type 1 /36/. This HLA-A3 (–B7)-associated disease induces iron overload (for further information, see Section 7.1.5.1 – Hereditary hemochromatosis) /37/. Homozygous point mutation of the HFE is found in 70–90% of hereditary hemochromatosis patients. The mutation occurs at position 282 of the HFE gene and consists of a single base transition of G to A, resulting in a change of amino acid at codon 282 (Tyr/Cys) and referred to as C282Y. The C282Y mutation causes dissociation with β₂-microglobulin and prevents the expression of an intact HFE molecule on the cell surface. Other two point mutations of clinical relevance have also been described. These gene variants are called H63D (position 187 G → C, codon 63 Asp → His) and S65Y (position 193 A → T and codon 65 Ser → Cys) and may lead to increased disease disposition, especially in the concurrent presence of a C282Y mutation in trans position. Besides a multitude of other HFE variants, a clinically relevant mutation E168X (codon 168, Glu to stop codon) has been described that leads to a stop codon causing premature termination of protein synthesis. There is no functional expression (null allele).

The co-occurrence (compound heterozygosity) of the E168X mutation in trans position and the C282Y, H63D or S65Y variant is also to be considered as increasing the predisposition to the disease /38/. Homozygous E168X carriers have not been observed to date. The HFE mutations can be determined by PCR amplification of the relevant HFE gene regions and subsequent determination of the sequence variation by restriction enzyme digestion or hybridization with sequence-specific oligonucleotides.

26.3 The HLA system and transplantation

26.3.1 Organ transplantation

The transplantation of solid organs is an essential therapeutic option in the treatment of severe organ dysfunction. Decisive criteria in transplantation are the compatibility of AB0 blood group types and, depending on the organ to be transplanted, the histocompatibility in the HLA system (Tab. 26-6 – Mean organ survival following kidney transplantation). Aspects influencing graft survival include the prevention of hyperacute rejection reactions due to allo antibodies naturally occurring in the AB0 system and the avoidance of transplantation in the presence of preformed HLA allo antibodies and partial matching of HLA alleles (Collaborative Transplant Study).

Impressive evidence has been provided in kidney transplantation that points to the necessity of considering the HLA system in organ transplantation despite much progress in immunosuppressive therapy. Graft survival is markedly longer in HLA matched siblings than in haplotype matched parents or in HLA compatible kidneys from unrelated donors. The differences in graft survival become especially evident in long term monitoring. Low resolution HLA typing which refers to the alleles at gene loci HLA-A, HLA-B and HLA-DRB1 reported at two digit level (e.g., HLA-DRB1*01) is used for histocompatibility assessment. To improve histocompatibility, especially in the presence of preformed HLA antibodies, the Tissue Typing Advisory Committee (TTAC) of Eurotransplant recommends to also determine the HLA-C and HLA-DQB1 alleles for consideration in the donor organ allocation.

As a rule, molecular biological determination is used for typing. Patient analyses should be confirmed by second determination from a separately collected blood sample. If the tissue typing results show only one allele at the allocation relevant gene loci (suspected homozygosity) in patients after the first analysis, a family analysis or further (e.g., high resolution) molecular biological analysis should be performed.

HLA compatibility is of utmost significance in kidney transplantation and combined kidney/pancreas transplantation. Moreover, organ allocation to immunized recipients of heart or lung transplants and small intestine transplants should include histocompatibility taking into account HLA antibody specificities. HLA compatibility is of minor significance in other solid organ transplantations and/or non-immunized recipients of heart or lung transplants. This is because the ischemia time which is critical especially in heart and lung transplantation is prolonged due to the determination of the HLA alleles.

Diagnostic testing for preformed HLA allo antibodies is another important aspect in the prevention of acute and chronic rejection reactions besides the matching of HLA alleles between donor (living donor or cadaveric organ) and recipient. Similarly to preformed AB0 blood group antibodies, HLA allo antibodies can lead to (hyper)acute rejection. To avoid such severe side effects, patients waiting for a transplant are tested for preformed HLA allo antibodies at regular intervals (120–180 days).

These tests are performed using serological methods such as the lymphocytotoxicity test (LCT) as well as solid phase testing by ELISA, microsphere particle assay (Luminex) or flow cytometry. Immunized patients are usually tested for HLA antibody specificity using LCT and solid phase testing.

The results of the antibody tests are recorded in the central database of trans regional transplantation organizations (e.g., Eurotransplant) and taken into consideration in the allocation of organs according to the type of organ. Serum samples of all HLA allo antibody positive patients are exchanged between the different transplantation laboratories and countries and made available for serological leukocyte cross-matching based on a 24 hour on-call service. Corresponding serum samples of HLA allo antibody negative recipients are available in the regional laboratories. These elaborate logistics in the diagnostic testing and provision of serum samples for the assessment of organ compatibility enable the directed selection of HLA matched organ recipients and, thus, achieve markedly improved graft survival.

26.3.2 Blood stem cell transplantation

Blood stem cell transplantation is a special challenge to the immunological matching between donor and recipient. Blood stem cell transplantation is the final therapeutic step and usually involves the total removal of the afflicted (leukemic) bone marrow and replacement with a healthy blood stem cell transplant from a donor. The practical consequence is a chimeric patient whose immune system and hematopoietic system have been replaced. Therefore, the transplanted new immune system must be matched with the recipient to the best possible extent to prevent severe, or even lethal, rejection (Tab. 26-7 – Indication for advanced molecular biological HLA typing in preparation of initial allogeneic hematopoietic stem cell transplantation).

In contrast to organ transplantation, where transplant rejection is mediated by the recipient’s immune system host-versus-graft (HVG) reaction, blood stem cell transplants may be activated against the recipient (graft-versus-host (GVH) reaction). Besides the GVH reaction, blood stem cell transplantation may also imply an HVG reaction which is, however, not as pronounced due to intensive conditioning of the patient aimed at the removal of the patient’s own blood stem cells. In such a GVH reaction, endogenous structures like the intestine, skin, mucus layers and liver are attacked by the transplanted, alloimmune system. Depending on the intensity of the attack, this may lead to organ failure and, due to mucosal lesions, severe exudative inflammatory bowel disease followed by infections.

However, moderate GVH reactions are also observed in HLA matched donors and may indeed be beneficial. These reactions can be clinically controlled and protect from recurrence of the underlying disease by causing the transplanted immune cells to attack the patient’s residual leukemic cells (graft-versus-leukemia (GVL) reaction).

The pronounced polymorphism of the HLA system imposes special requirements on the selection of an HLA matched blood stem cell donor. The ideal donor is an HLA genotype matched individual and, based on the genetic rules, can be found in 25% of the siblings of an afflicted patient. In order to determine the genotypes (family tree), the tests include both the siblings and the parents. If no genotype matched donor is detected within the core family, testing can be extended to other relatives (uncles/aunts, cousins) where, however, an HLA matched donor will only be found in rare cases. Hence, upon lack of a donor from the core family, the search for an HLA matched donor will immediately be extended to unrelated, voluntary blood stem cell donors.

The progress in immunogenetic selection and therapy of the side effects of blood stem cell transplantation has led to comparably good transplantation results for both HLA genotype matched siblings/relatives and unrelated donors.

However, one should always keep in mind that the matching in unrelated donors refers only to the HLA phenotype because no family testing was performed. Thus, the required scope of HLA typing is much larger to ensure HLA compatibility. In practice, this implies that high resolution (four digit, e.g., HLA-A*02:01,03:01) molecular biological determination of the alleles at the gene loci HLA-A, HLA-B, HLA-C, HLA-DRB1 and HLA-DQB1 must be performed on all unrelated blood stem cell donors /39/. The enormous polymorphism requires the use of complex and sophisticated molecular biological methods of determination (e.g., sequencing or extended hybridization and/or amplification methods) in this context.

Clinical assessment of the HLA compatibility of the five relevant loci is based on 10/10 matching of the analyzed alleles (2 alleles per HLA locus). In some cases, extended assessment of the HLA compatibility also includes the configuration of HLA class I (especially the HLA-C locus) alleles and the determination of NK cell receptors (killer immunoglobulin-like receptor, KIR) /40, 41/. In such cases, a mismatched donor/recipient HLA structure in the context of NK cell/T cell mediated immune response could result in superior survival following stem cell transplantation /42, 43/.

In all, more than 17 million voluntary blood stem cell donors (including more than 4 million in Germany) are registered worldwide. Thus, unrelated HLA matched donors are found by the Bone Marrow Donor World Wide Registry (BMDW) and the German Central Bone Marrow Donor Registry (ZKRD) for 80–90% of patients afflicted with hematological disease who do not have related donors. The cure rate (more than 5 years without relapse) for leukemic disease is 30–90% depending on the underlying disease, the disease stage and the age group. Success rates above 90% have been achieved in the treatment of genetic defects such as hemoglobinopathies (e.g., homozygous forms of β-thalassemia). Survival rates are similarly high in primary immunodeficiencies (e.g., SCID), especially in blood stem cell transplantations on patients who have not already suffered from severe infections caused by the underlying disease.

26.3.3 Relevance and diagnostics in transfusion medicine

In transfusion medicine, the HLA system types are of significance in the differential diagnostic detection of side effects and in the application of cellular components. In this context, the focus is on unwanted alloimmunization following transfusion with blood components having a residual leukocyte content. Such residual contents are inevitably obtained in manufacturing processes employing differential centrifugation and subsequent separation of blood components (red blood cells, platelets, plasma) from whole blood.

The introduction of in-line leukocyte filtration of cellular blood components (red blood cells and platelets) and the use of equivalent filter systems in plasma production have already markedly reduced the alloimmunization rate. A leukocyte content of 5 × 10⁶ (CILL, concentration of immunogenic leukocyte load) is considered critical for the formation of cytotoxic antibodies.

Modern leukocyte filter systems or apheresis methods for the production of blood components reduce the residual leukocyte content to below 1 × 10⁶/preparation and, thus, are markedly below the critical CILL. However, platelet transfusions, for example, in previously immunized patients may cause HLA antibodies to reappear /44, 45/.

Besides human neutrophil antibodies (HNA), HLA antibodies are also responsible for the immunological form of transfusion-associated acute lung injury (TRALI). In order to control this severe or even lethal side effect, the German Blood Working Group issued a recommendation (vote) and the higher federal authority (Paul Ehrlich Institute) prepared a catalog of measures (step-by-step plan) for donor questioning and testing to prevent the transfusion of blood components containing HLA/HNA antibodies.

26.3.3.1 Platelet transfusion

Besides blood group antigens (especially those of the ABH system), the human platelet antigens (HPA), platelets also express HLA class I types on their cell surface. Because of the small manifestation of AB0 types, AB0 incompatible platelet administration (e.g., A to 0) only rarely leads to shorter recovery times. As a rule, acute and hemolytic transfusion reactions do not occur; therefore, serological erythrocyte cross-matching is not necessary. However, as small amounts of residual red blood cells may remain in the platelet concentrate, rhesus factor D should be considered in the selection of the transfusion material, if possible.

Antibodies to HLA class I antigens and human platelet antigens are clinically relevant in alloimmunization /46/. Alloimmunization impresses clinically by markedly reducing the expected therapy related increment in platelet count following transfusion due to a massive decrease in platelets (refractoriness). Refractoriness is often seen in patients receiving platelets repeatedly or for an extended period of time, for example, following chemotherapy or blood stem cell transplantation. Moreover, women with previous pregnancies are particularly affected.

Refractoriness is the repeated (at least twice) failure to achieve an acceptable increment in platelet count. It is determined calculating the corrected count increment (CCI) as follows:

Refractoriness is assumed in a CCl < 5 × 10⁹/L after 1 hour.

Differential diagnosis in refractoriness should take into account immunological causes based on alloantibodies to HLA or HPA antigens as well as non-immunological factors (e.g., fever, sepsis, antibiotics treatment, disseminated intravascular coagulation and splenomegaly). If HLA and/or human platelet antibodies (HPA)are detectable, an HLA and/or HPA matched platelet donor should be found. For this purpose, a large number of potential HLA and/or HPA typed platelet donors are registered in several transfusion medical departments. Serological leukocyte cross-matching can be performed to assess the success of transfusion, especially in the absence of HLA matched donors and recipients.

26.3.3.2 Neutrophil transfusion

The importance of neutrophil transfusion has increased due to the possibility of donor pretreatment with granulocyte growth factors (G-CSF, neupogen) and the concomitant marked increase in neutrophil content in preparations obtained by apheresis /47, 48/. Neutrophil transfusions are indicated in patients with progressive infections and severe neutropenia of less than 0.5 × 10⁹/L neutrophils and in prophylactic therapy in granulocytopenia with a high risk of occurrence of life threatening bacterial and fungal infection. Moreover, patients with rare congenital neutrophil dysfunction can benefit from neutrophil transfusion in progressive infection /49/.

Granulocyte concentrates typically contain a large amount of red blood cells, should be AB0 and Rh compatible and require prospective serological erythrocyte cross-matching. Moreover, leukocyte cross-matching and testing for leukocyte antibodies before and during treatment should be performed. Leukocyte compatibility is assessed in order to avoid side effects such as febrile non hemolytic transfusion reactions, pulmonary reactions and refractoriness. Granulocytes must be irradiated with a mean dose of 30 Gy before application.

Refractoriness after the administration of granulocyte concentrates can be caused by alloimmunization of the recipient to HLA antigens and granulocyte specific antigens and by non immunological factors such as fever, sepsis and splenomegaly. The increment at 4–8 h after administration of the granulocyte concentrate, (1.5–3.5) × 10⁸ granulocytes/kg of body weight, is suitable for refractoriness assessment. An increment below 5 × 10⁸/L indicates refractoriness following alloimmunization, especially in patients without clinical indications of non immunological causes.

26.3.3.3 Febrile non hemolytic transfusion reaction (FNHTR)

Febrile non hemolytic transfusion reactions are caused by the recipient’s antibodies to contaminated leukocytes in cellular blood components (platelet, red blood cell and granulocyte concentrates) and especially by cytokines released during storage. In clinical symptoms such as fever, shivering and moderate dyspnea, differential diagnostic distinction from hemolytic transfusion event and, in particular, testing for the presence of allo antibodies to HLA and/or HPA (platelets) and/or HNA (granulocytes) are necessary.

26.4 HLA typing and HLA antibodies

HLA types are determined using serological and molecular biological methods. Serotyping is based on the lymphocytotoxicity test developed by Paul Terasaki, where HLA antigens are detected by means of antibodies (antisera). In reverse application, this method is also suited for detecting HLA specific antibodies. Because of its relative ease of use and robustness, it is employed in general clinical testing to detect HLA specific antibodies and used as the standard method for serological leukocyte cross-matching.

Serological methods for determining HLA class I and class II alleles have generally been replaced by molecular biological methods because of the small possibility of detection and the unavailability of sufficient amounts of antisera in serotyping. Similarly, molecular biological methods for the detection of HLA class I alleles are routinely employed in diagnostic investigations on a large scale. The introduction of molecular biological methods have also reduced the importance of cellular typing methods. Mixed lymphocyte culture has been entirely superseded by detailed (high resolution) molecular genetic analysis of HLA class I alleles in the diagnostic assessment of histocompatibility. The main serological and molecular biological test methods are presented in the following.

26.4.1 HLA serotyping by microlymphocytotoxicity test

The microlymphocytotoxicity test (standard NIH method)is based on complement mediated lymphocytolysis by HLA specific antibodies /50/. The antibody can only be bound in the presence of the corresponding antigen on the lymphocyte cell membrane. The microlymphocytotoxicity test is complement dependent (complement dependent cytotoxicity test) and provides the basis for the serotyping of HLA class I/II alleles and the detection of HLA antibodies, where the complement components are activated by immune complexes of IgG (IgG₁, IgG₂, IgG₃) and IgM and subsequently mediate lymphocytolysis by HLA specific antibodies. The membrane of the damaged cell becomes permeable to certain dyes (e.g., eosin or acridine orange/ethidium bromide).

Using immunomagnetic beads and fluorescence dyes T cells or B cells for HLA class I (T cells) or HLA class II (B cells) are determined quickly, neatly and specifically. Monoclonal anti-CD2 antibodies are bound to immunomagnetic polystyrene beads. The employed antisera predominantly stem from polyclonal human sera. However, several monoclonal antisera are also available.

26.4.1.1 Indication

HLA associated autoimmune disease, organ transplantation, blood stem cell transplantation, platelet refractoriness, pretesting of blood stem cell donors, forensic investigation.

26.4.1.2 Specimen

Lymphocytes, usually from anticoagulated peripheral whole blood: heparin blood, acid-citrate-dextrose (ACD) blood or citrate-phosphate-dextrose-adenine (CPDA) blood. Transport at ambient temperature for a maximum of 3 days after collection (as a rule, the sample should be processed within 24 hours). The minimum volume is 5 mL depending on the cell count.

26.4.1.3 Method of determination

Terasaki microtiter plates (72 wells) are coated with oil. Then, 1 μL of antiserum is pipetted into each well. The different antisera are selected depending on the HLA antigens to be determined. The lymphocytes to be analyzed are isolated by density gradient centrifugation or using immunomagnetic beads. Then, 1 μL (approximately 2,000 lymphocytes) of the cell suspension is pipetted into each well. Care must be taken to ensure that antiserum and cell suspension are well mixed. After incubation for 30 min. at ambient temperature, 5 μL of complement are added and the solution is incubated for 60 min. at ambient temperature. Five μL of dye (eosin or fluorescence solution) are added. If eosin is used, another 5 μL of formalin are added to stop the reaction. The reactions are read off under the microscope.

26.4.2 Molecular biological determination of HLA types

The molecular biological methods for HLA determination are based on the direct detection of allele specific or group specific sequence differences at the HLA gene loci analyzed following genomic DNA amplification by polymerase chain reaction (PCR). Because of the structural differences in the HLA molecules, the assessment for determining the polymorphic sequence position usually includes the exon 2 and 3 regions (possibly also exons 1 and 4) for HLA class I alleles and the exon 2 region (possibly also exons 1, 3 and 4) for HLA class II alleles. Detection methods using sequence specific oligonucleotides (SSO) or sequence specific primers (SSP) have proven useful in routine diagnostics. Moreover, direct genomic DNA sequencing (sequenced-based typing, SBT) is employed especially in high resolution (four-digit) HLA typing.

26.4.2.1 SSO (sequence specific oligonucleotide) HLA typing

Detection of allele specific and group specific sequence variations by hybridization with short synthetic DNA probes, so called oligonucleotides. The oligonucleotide probes are selected by utilizing nucleotide differences between the individual alleles or allele groups.

26.4.2.1.1 Indication

HLA associated autoimmune disease, organ transplantation, platelet refractoriness, pretesting of blood stem cell donors, forensic investigation.

26.4.2.1.2 Specimen

Peripheral, venous, EDTA anticoagulated whole blood (approximately 1 mL minimum amount depending on the vitality and number of cells), and all types of nucleated cells including tissue samples suitable for DNA isolation. Store and transport cells at ambient temperature.

26.4.2.1.3 Method of determination

Genomic DNA isolated from tissue or peripheral whole blood is amplified by gene locus specific PCR. Amplification using biotin labeled 5’ modified primer is performed for later reaction evidence. Then, the DNA is denatured into single strands. Using the dot blot method, the amplified DNA fragments are then spotted spatially separated on a nitrocellulose or nylon membrane. The HLA types are differentiated by hybridization using the different oligonucleotide probes.

Another method used is reverse hybridization where, in a reverse approach of the dot blot method, separated pairs of different oligonucleotide probes are spotted immobilized onto strips of nylon membrane or transferred to different wells of an ELISA test plate. Alternatively, the probes can also be spotted onto micro particles (micro spheres or Luminex beads) detectable by different fluorescence emission wavelengths in a flow cytometer (Luminex analyzer). The classical reverse dot blot method has been developed further by using probes spotted closely to one another into the wells of micro titer plates. These spots consist of individual probes or combinations of two or several probes (so called mosaic probes).

Then, the gene locus specific, amplified DNA fragments are added. Hybridization is effected by complementary attachment of oligonucleotide and DNA target sequence. Specificity is increased by subsequent washing steps. Detection is based on a color reaction where, for example, streptavidin coupled to alkaline phosphatase is added after hybridization. Streptavidin binds to the biotin labeled DNA fragments hybridized on the nitrocellulose strips or mosaic spots, forming a purple/brown precipitate after incubation with a chromogenic substrate (BCIP/NBT). Alternatively, FITC labeled SSO probes can be used. In a color reaction measured with an ELISA reader or spot photo processor, they are detected using FITC specific antibody fragments coupled with horseradish peroxidase (POD).

The time required for this kind of typing is between one to several days (dot blot method) or a few hours (reverse dot blot method), depending on the typing format and resolution (two-digit or four-digit typing). The SSO typing method is especially suited for automation, allowing a high sample throughput. In emergency diagnostics, particularly for organ transplantation, however, it is inferior to the SSP typing method.

26.4.2.1.4 Biological influence factors and interference factors

DNA contamination with proteins, contamination with exogenous DNA. Contamination of the DNA with PCR inhibitors such as hemoglobin, heparin or ethanol can lead to substantial interference with the PCR. Insufficiently stringent washing conditions or sequence homologies can result in oligonucleotide cross reactions.

26.4.2.2 SSP (sequence specific primer) HLA typing

This test is based on the amplification of HLA specific gene regions by PCR using sequence specific primers. The primer target sequence includes an allele specific or group specific polymorphic sequence position. This test method utilizes the fact that successful reaction of both primers (especially at the 3’ terminal end) is only possible in the absence of mismatches. Thus, an amplificate is only obtained upon complete agreement of the primers with the target sequence. If the allelic sequence to be determined with the primer is not present, no primer binding is possible due to mismatching and no amplificate can be determined.

26.4.2.2.1 Indication

HLA associated autoimmune disease, organ transplantation (including emergency typing during on call duty), platelet refractoriness, pretesting of blood stem cell donors, forensic investigation.

26.4.2.2.2 Specimen

Peripheral, venous, EDTA anticoagulated whole blood (approximately 1 mL minimum amount depending on the vitality and number of cells), and all types of nucleated cells including tissue samples suitable for DNA isolation. Store and transport cells at ambient temperature.

26.4.2.2.3 Determination

After genomic DNA has been isolated, 50–200 μg of DNA are used per PCR mixture. The number of PCR mixtures needed for HLA typing depends on the resolution (two-digit or four-digit) and the number of analyzed gene loci. However, the primer pairs should be selected such as to enable PCRs under the same temperature conditions. The amplified DNA fragments are then determined by zonal electrophoresis in agarose. Successful amplification will create a DNA fragment of defined length visible as a band. If amplification does not occur, no band will show. Therefore, internal amplification control is performed in each PCR mixture by co amplification of a non polymorphic gene region (e.g., human growth hormone gene).

26.4.2.2.4 Biological influence factors and interference factors

DNA contamination with proteins, contamination with exogenous DNA. Contamination of the DNA with PCR inhibitors such as hemoglobin, heparin or ethanol can lead to substantial interference with the PCR. Primer mismatching may result in sequence homology or insufficient annealing temperature.

26.4.2.3 HLA typing by direct sequencing

Sequencing of PCR fragments after HLA gene locus-specific amplification of genomic DNA. The test is based on the ddNTP (di-deoxynucleotidetriphosphates) chain termination reaction according to Sanger.

26.4.2.3.1 Indication

High resolution typing prior to blood stem cell transplantation or for risk assessment in autoimmune disease associated with HLA subtypes.

26.4.2.3.2 Specimen

Peripheral, venous, EDTA anticoagulated whole blood (approximately 1 mL minimum, depending on the vitality and number of cells), and all types of nucleated cells including tissue samples suitable for DNA isolation. Store and transport cells at ambient temperature.

26.4.2.3.3 Method of determination

Sequencing is performed by PCR using fluorescence labeled chain termination nucleotides (ddNTPs; ddAdenin, ddGuanin, ddCytosin, ddThymin). Dideoxyribonukleosid triphosphates (ddNTPs) are artificial DNA nucleotides which are used for DNA sequencing according to Sanger. Certain techniques are employed to label each ddNTP with a different fluorescence dye. When the sequencing reaction is completed, the resulting fluorescence labeled fragments are separated according to their length by polyacrylamide or capillary gel electrophoresis. As a rule, an automated computer aided evaluation system is used to read the fluorescence emission and determine the base pair sequence resulting from the sequencing reaction. Finally, the HLA alleles are determined by comparing the obtained sequence against those stored in sequence databases.

26.4.2.3.4 Biological influence factors and interference factors

DNA contamination with proteins, contamination with exogenous DNA. Contamination of the DNA with PCR inhibitors such as hemoglobin, heparin or ethanol can lead to substantial interference with the PCR. Primer mismatching may result in sequence homology or insufficient annealing temperature.

26.4.3 HLA antibody detection and serological leukocyte cross-matching

HLA antibodies against HLA class I and class II types may form and usually are IgG antibodies. IgM antibodies have also temporarily been observed following exposure to antigens. HLA antibodies form especially in the context of pregnancy, blood transfusion and transplantation. The separation of HLA specific antibodies from autoantibodies is decisive in diagnostic testing because autoantibodies combined with autologous (i.e., the patient’s) cells and donor lymphocytes may lead to a positive antibody result or a positive cross-matching result. Such antibodies are irrelevant to organ transplantation.

Typically, autoantibodies belong to the IgM class and are lymphocytotoxins which occur naturally or in certain diseases /51/. They lose their cytotoxic activity if their pentameric structure is destroyed using dithiothreitol (DTT). HLA specific IgG allo antibodies are not affected by DTT treatment. Hence, if the presence of autoantibodies is suspected, the patient should be tested by autologous cross-matching and by cross-matching against another individual (e.g., cadaveric donor).

The following criteria can indicate the presence of autoantibodies /52/:

• Slightly positive screening results without discernable specificity
• Evidence of IgM antibodies in DDT screening
• Patient sera with detected autoantibodies. It is recommended to test the serum of such patients for autoantibodies at least once a year.

Thus, it is also possible to discriminate between IgM autoantibodies and HLA specific IgM allo antibodies. HLA specific IgM autoantibodies show a negative reaction in autologous cross-matching and should exhibit specificity for HLA in antibody differentiation.

HLA class I and HLA class II antibodies of the IgG class are clinically relevant /53/. A negative influence on the course of organ transplantation has been observed in pre sensitized patients (Collaborative Transplant Study).

26.4.3.1 Antibody detection by microlymphocytotoxicity test

Principle

The microlymphocytotoxicity test is based on component mediated lymphocytolysis by HLA specific antibodies (antisera) and corresponds to the method used for HLA classes I/II typing (Section 26.4.1 – HLA serotyping by microlymphocytotoxicity test).

26.4.3.1.1 Indication

Organ transplantation, platelet refractoriness, pre-testing of only partially HLA matched blood stem cell donors, transfusion events (febrile non hemolytic transfusion reaction), neutrophil transfusion.

26.4.3.1.2 Specimen

Serum. Transport at ambient temperature for not more than 5 days after the time of collection (as a rule, the sample should be processed within 24 hours), otherwise transport at more than –20 °C. Storage at more than –25 °C for several years. The required volume is 5 mL depending on the method used. Due to the specification of sending aliquots of antibody positive sera to a multitude of laboratories at different national and European transplantation centers within the organ transplantation network (e.g., Eurotransplant), the minimum amount of serum should be 10 mL for patients on the organ transplantation waiting list.

26.4.3.1.3 Method of determination

Corresponds to the standard microlymphocytotoxicity test. However, this test method employs lymphocyte suspensions of different known HLA phenotypes to determine antibody reactivity (so called cell panel). The cell panel (usually 50 different individuals) is either newly created by collecting blood samples from registered donors or can be stored as deep frozen cells. Lymphocyte suspensions of patients suffering from chronic lymphocytic leukemia can be used as an alternative to the lymphocytes of healthy donors.

26.4.3.1.4 Biological influence factors and interference factors

Complement activity, specificity and cross-reactivity of antisera, autoantibodies.

26.4.3.2 Antibody detection by ELISA

Commonly employed methods use soluble HLA antigens which enable the detection of HLA class I and class II antibodies in the patient serum by an enzyme/substrate reaction. In contrast to the microlymphocytotoxicity test, IgG or IgM antibodies can be determined in separate test mixtures regardless of their complement binding capacity.

26.4.3.2.1 Indication

Same as under serological antibody detection.

26.4.3.2.2 Specimen

Same as under serological antibody detection.

26.4.3.2.3 Method of determination

According to the different, commercially available test formats, antibody screening often uses mixtures of different soluble antigens enabling the distinction between antibody positive and antibody negative samples. For antibody differentiation, the different HLA class I and class II antigen extracts of several individuals are spotted into the wells of an ELISA test plate. The addition of patient serum containing specific HLA antibodies leads to the formation of an antigen complex, where the bound anti-HLA IgG Ab or anti-HLA IgM Ab are made visible using a conjugated anti-human IgG or anti-human IgM antibody. The use of alkaline phosphatase and 5-bromo-4-chloro-3-indoxyl phosphate (BCIP), for example, allows easy identification (blue at 630 nm).

26.4.3.2.4 Biological influence factors

Important aspects include the conformity of the soluble antigens, antigen specificity and the absence of antibody cross-reactivity.

26.4.3.3 Antibody detection by micro particles (Luminex microbeads)

Analogously to ELISA, this method is based on microscopically small polystyrene particles, so called micro beads, serving as solid phase. The micro beads are labeled with two fluorescent dyes (red and infrared) which emit in different spectral regions. By combining the two dyes in ten different concentration levels, one hundred spectrally distinct shades of red and infrared are obtained. Each resulting fluorescence intensity defines a population of micro beads serving as the basis for different test formats. Micro bead (Luminex)-based test formats are also employed for HLA typing besides the detection and specification of HLA antibodies Refer to Section 26.4.2.1 –Sequence specific oligonucleotide (SSO) HLA typing.

The micro beads are coated with soluble HLA antigens which enable the detection of HLA class I and/or class II antibodies in the patient serum.

This method utilizes the different color emission spectrum of the micro beads, thus allowing the analysis of up to 100 different parameters in a test mixture (well). Therefore, it is suited for screening as well as differentiation. In particular, Luminex based methods are used to determine the antibody specificities by single antigen format where a specific HLA antigen is added for each different fluorescence intensity. Beside antibodies against HLA-A, B, C, DR and DQ, the test allows the detection of antibodies against HLA-DP and the HLA class II alpha chain (DRA, DQA, DPA) and is increasingly used to differentiate between antibodies leading to unacceptable antigens (giving a contra-indication for transplantation) or risk factors (leading to adapted immunosuppression) in organ transplantation.

26.4.3.3.1 Indication

Same as under serological antibody detection. Because of its especially high sensitivity, single antigen testing is used in the following indications: living donation (spli-liver, kidney), immunized patients with markedly poly specific antibody pattern. In post transplantation for the detecting potential graft specific antibodies.

26.4.3.3.2 Method of determination

Solid phase test formats (micro particle based screening and differentiation assays) are commercially available to differentiate between HLA class I and/or HLA class II antibodies. The test allows also for the detection of complement binding specificities by the addition of C1q. The measured mean fluorescence intensity (MFI) of antibody specificities detected is used to assess the clinical relevance. Criteria for this assessment are complex and should take into account the immunization history as well as the occurrence of natural antibodies. Due to their increasing relevance in transplantation, guidelines or recommendations for the definition of unacceptable antigens based on solid phase testing in combination with lymphocytotoxicity test have been defined (www.euro­trans­plant.org or www.immungenetik.de).

26.5 Statistical rules for the evaluation of HLA antibodies

Antibody screenings and antibody differentiation are evaluated according to statistical rules to determine antibody specificities. Evaluation starts with a four field table showing relations between serum reaction and present antigen specificity i.e., true positive (a), false positive (b), false negative (c ) and true negative (d) reactions. The significance of a determined HLA specificity is calculated using the X² value and the correlation coefficient (c ).

For X² values > 3.84, a positive significance level of p < 0.05 is obtained; r values > 0.85 point to a positive correlation.

The reaction strength of a serum to an HLA specificity can be assessed using the strength index (SI). Calculation of the inclusion index (II) in polyspecific sera allows to assess the proportion of an HLA specificity in the total of HLA specificities of this serum.

The total of positive reactions can also be depicted as the percentage in panel scope (number of used lymphocytes of different individuals) and is referred to as % PRA (percentage of panel reactive antibodies).

Panel reactivity (PRA = [(a + c)/N] × 100 (%)

This value allows to assess the probability of obtaining a positive result from serological cross-matching with any unrelated individual (e.g. cadaveric donor), on condition, however, that a random sample of the selected cell panel reflects the frequencies of the relevant ethnic population (e.g., Europeans).

26.6 Serological leukocyte cross-matching

Based on the microlymphocytotoxicity test (Section 26.4.1 – HLA serotyping by microlymphocytotoxicity test). In the presence of complement as well as antibodies to lymphocyte types, a lymphocyte suspension of the donor will react with the corresponding recipient serum in a way that causes cytotoxic damage to the cell membrane.

26.6.1 Indication

Organ (especially kidney) transplantation or partially HLA mismatched bone marrow transplantation, platelet transfusion in refractory patients receiving HLA matched platelet preparations, neutrophil transfusion.

26.6.2 Specimen

10 mL of recipient serum and donor lymphocytes (isolated from peripheral venous heparin anticoagulated blood; in cadaveric donors alternatively lymphocytes isolated from 3–5 cm³ of spleen or 2–3 lymph nodes).

26.6.3 Method of determination

Cross-matching is performed using non separated lymphocytes (T and B cells) and/or separated T and/or B lymphocytes from peripheral blood or spleen/lymph nodes (in cadaveric donors within the scope of organ transplantation). Non activated T cells express HLA class I antigens, whereas B cells constitutively express both HLA class I and HLA class II antigens. Thus, cross-matching with non separated lymphocytes enables the detection of both HLA class I and HLA class II antibodies. It must be taken into account, however, that the portion of B cells in blood is much smaller than in the spleen or lymph nodes. In addition to the non separated cross-matching mixture, a mixture separated by T cells and B cells can be used to differentiate between HLA class I and HLA class II antibodies.

Standard incubation is 30 min. followed by 45–60 min., both at ambient temperature. During incubation, antibodies in the serum bind to antigens on the cell surface. If a sufficient amount of antibodies is bound and these antibodies belong to the matching class of immunoglobulins (IgM, IgG₁, IgG₃), complement is activated and the cell membrane is destroyed. Hence, incubation times are very important. If incubation is too long, the bound complex will detach again from the cell membrane or will be internalized into the cell.

26.6.4 Biological influence factors and interference factors

It is important in cross-matching to exclude the presence of autoantibodies (IgM) by mixing the relevant sera with dithiothreitol.

26.7 Websites related to immunogenetics

HLA nomenclature and current overview of existing alleles and gene loci

• Anthony Nolan Bone Marrow Thrust: www.anthonynolan.org/HIG/index.html
• IMGT/HLA database: www.ebi.ac.uk/imgt/hla/
• dbMHC database: www.ncbi.nlm.nih.gov/gv/mhc

Guidelines for organ and blood stem cell transplantation – standards for laboratory accreditation:

• Deutsche Gesellschaft für Immungenetik (DGI): www.immungenetik.de
• European Federation for Immunogenetics (EFI): www.efi-web.org
• Deutsche Transplantationsgesellschaft (DTG): www.d-t-g-online.de
• Deutsche Gesellschaft für Hämatologie und Onkologie (DGHO): www.dgho.de
• International Histocompatibility Working Group: https://www.fredhutch.org/en/research/institutes-networks-ircs/international-histocompatibility-working-group.html

Transfusion medicine guidelines and recommendations on hemotherapy:

• Deutsche Gesellschaft für Transfusionsmedizin und Immunhämatologie (DGTI): www.dgti.de

Allocation and clinical outcomes of organ transplantation:

• Eurotransplant (ET): www.eurotransplant.org
• Collaborative Transplant Study (CTS): www.ctstransplant.org

Bone marrow and blood stem cell databases and donor search:

• Zentrales Knochenmarkspenderregister Deutschland (ZKRD): www.zkrd.de
• Bone Marrow Donor World Wide Registry (BMDW): https://wmda.info

References

1. Gorer PA. The antigenic basis of tumor transplantation. J Pathol Bacteriol 1938; 47: 231–52.

2. Schnell G, ed. Biology of the laboratory mouse. Dover, 1941.

3. Dausset J. Iso-leuco-anticorps. Acta Haematol 1958; 20: 156–66.

4. Hansen J, Dupont B, eds. HLA 2002: Immunobiology of the human MHC. IHWG Press, Seattle, USA.

5. Bjorkman PJ, Saper MA, Samraoui B, Bennett WS, Strominger JL, Wiley DC. Structure of the human class I histocompatibility antigen, HLA-A2. Nature 1987; 329: 506–12.

6. Bjorkman PJ, Saper MA, Samraoui B, Bennett WS, Strominger JL, Wiley DC. The foreign antigen binding site and T cell recognition of class I histocompatibility antigens. Nature 1987; 329: 512.

7. Gay D, Maddon P, Sekaly R, Talle MA, Godfrey M, Long E, et al. Functional interaction between human T-cell protein CD4 and the major histocompatibility complex HLA-DR antigen. Nature 1987; 328: 626–9.

8. Brown J H, Jardetzky TS, Gorga JC, Stern LJ, Urban RG, Strominger JL, Wiley DC. Three-dimensional structure of the human class II histocompatibility antigen HLA-DR1. Nature 1993; 364: 33–9.

9. Rammensee H G, Friede T, Stevanovic S. MHC ligands and peptide motifs: first listing. Immunogenetics 1995; 41: 178–228.

10. Zinkernagel R, Doherty P. Restriction of in vitro T cell mediated cytotoxicity in lymphocytic choriomeningitis within a syngeneic or semi-allogeneic system. Nature 1994; 248: 701–2.

11. Rouas-Freiss N, Goncalves RM, Menier C, Dausset J, Carosella ED. Direct evidence to support the role of HLA-G in protecting the fetus from maternal uterine natural killer cytolysis. Proc. Natl. Acad. Sci. USA 1997; 94: 11520–5.

12. Bauer S, Groh V, Wu J, Steinle A, Philips J H, Lanier L L, Spies T. Activation of NK cells and T cells by NKG2D, a receptor for stress-inducible MICA. Science 1999; 285: 727–9.

13. Klein J, Sato A. The HLA system. First of two parts. N Engl J Med 2000; 343: 702–9.

14. Hanson I M, Poustka A, Trowsdale J. New genes in the class II region of the human major histocompatibility complex. Genomics 1991; 10: 417–24.

15. Martin M, Mann D, Carrington M. Recombination rates across the HLA complex: use of microsatellites as a rapid screen for recombinant chromosomes. Hum Mol Genet 1995; 4: 423–8.

16. Thomson M, Neugebauer M, Arnaud J, Borot N, Sevin A, Baur M, Cambon-Thomson A. Recombination fractions in the HLA system based on the data set “provinces Francaises”: indications of haplotype-specific recombination rates. Immunogenetics 1994; 21: 33–43.

17. Baur M P, Danilovs JA. Reference tables for two and three locus haplotype frequencies of HLA-A, B, C, DR, Bf, and GLO. In: Histocompatibility Testing 1980. Terasaki PI, ed. UCLA Tissue Typing Laboratory, Los Angeles, USA 994–1210.

18. Krangel M S. Secretion of HLA-A and -B antigens via an alternative RNA splicing pathway. J Exp Med 1986; 163: 1173–90.

19. Seidl C, Lee JS. Expression of alternatively spliced HLA class II transcripts in lymphoid and nonlymphoid tissues. Immunogenetics 1992; 35: 385–90.

20. Puppo F, Indiveri F, Scudeletti M, Ferrone S. Soluble HLA antigens: new roles and uses. Immunol Today 1997; 18: 154–5.

21. Westhoff U, Doxiadis I, Beelen DW, Schaefer UW, Grosse-Wilde H. Soluble HLA class I concentrations and GVHD after allogeneic marrow transplantation. Transplantation 1989; 48: 891–3.

22. Tilg H, Westhoff U, Vogel W, Aulitzky WE, Herold M, Margreiter R, et al. Soluble HLA class I serum concentrations increase with transplant-related complications after liver transplantation. J Hepatol 1992; 14: 417–9.

23. DeVito-Haynes LD, Jankowska-Gan E, Sollinger HW, Knechtle SJ, Burlingham WJ. Monitoring of kidney and simultaneous pancreas-kidney transplantation rejection by release of donor-specific, soluble HLA class I. Hum Immunol 1994; 40: 191–201.

24. Carbone E, Terrazzano G, Colonna M, Tuosto L, Picolella E, Franksson L, et al. Natural killer cell clones recognize specific soluble HLA class I molecules. Eur J Immunol 1996; 26: 683–9.

25. Adamashvilli I M, McDonald JC, Fraser PA, Milford EL, Pressly TA, Gelder FB. Soluble HLA class I antigens in patients with rheumatoid arthritis and their families. J Rheumatol 1995; 22: 1025–31.

26. Tsuchiya N, Shioto M, Yamaguchi A, Ito K.. Elevated serum levels of soluble HLA class I antigens in patients with systemic lupus erythematosus. Arthritis Rheum 1996; 39: 792–6.

27. Ott M, Seidl C, Westhoff U, Stecker K, Seifried E, Fischer PA, Grosse-Wilde H. Soluble HLA class I and class II antigens in patients with multiple sclerosis. Tissue Antigens 1998; 51: 301–4.

28. Alvarez-Cermeno JC, Villar LM, Nocito M, Bootello A, Gonzalez-Porque P. Intrathecal synthesis of soluble class I antigens in multiple sclerosis. J Neuroimmunol 1992; 36: 77–9.

29. Mallal S, Pillips E, Carosi G, Molina JM, Workman C, Tomazic J, et al. HLA-B*5701 screening for hypersensitivity to abacavir. N Engl J Med 2008; 358: 568–79.

30. Martin AM, Almeida CA, Cameron P, Purcell AW, Nolan D, James I Mc Cluskey, et al. Immune responses to abacavir in antigen-presenting cells from hypersensitive patients. AIDS 2007; 19; 21:1233–44.

31. Svejgaard A, Jersild C, Staub-Nielssen L, Bodmer WF. HL-A Antigens and disease. Statistical and genetical considerations. Tissue Antigens 1974; 4: 95–105.

32. Woolf B. On estimating the relation between blood group and disease. Ann Hum Genet 1955; 19: 251–3.

33. Miettinen, O. Estimability and estimation in case-referent studies. Am J Epidemiol 1976; 103: 226–35.

34. Bengtsson BO, Thomson G. Measuring the strength of associations between HLA antigens and disease. Tissue Antigens 1981; 18: 356–63.

35. Svejgaard A, Ryder LP. HLA and disease association: Detecting the strongest association. Tissue Antigens 1994; 43:18–27.

36. Lebron JA, Bennett MJ, Vaughn DE, Chirino AJ, Snow PM, Mintier GA, et al. Crystal structure of the hemochromatosis protein HFE and characterisation of its interaction with the transferrin receptor. Cell 1998; 93: 111–23.

37. Feder JN, Penny DM, Irrinki A, Lee VK, Lebron JA, Watson N, et al. The hemochromatosis gene product complexes with the transferrin receptor and lowers its affinity for ligand binding. Proc Natl Acad Sci USA 1998; 95: 1472–7.

38. Gottschalk R, Seidl C, Löffler T, Seifried E, Hoelzer D, Kaltwasser JP. HFE codon 63/282 (H63D/C282Y) dimorphism in German patients with genetic hemochromatosis. Tissue Antigens 1998; 51: 270–5.

39. Ottinger HD, Müller C, Riebschläger S, Arnold R, Beelen DW, Blasczyk R, et al. Dritter Deutscher Konsensus zur immungenetischen Spenderauswahl für die allogene Stammzelltransplantation, 2005.

40. Granot N, Storb R. History of hematopoietic cell transplantation: challenges and progress. Haematologica 2020; 105 (12): 2716–29.

41. Farag SS, Fehniger TA, Ruggeri L, Velardi A, Caligiuri MA. Natural killer cell receptors: new biology and insights into the graft-versus-leukemia effect. Blood 2002; 100:1935–47.

42. Hsu KC, Keever-Taylor CA, Wilton A, et al. Improved outcome in HLA-identical sibling hematopoietic stem-cell transplantation for acute myelogenous leukemia predicted by KIR and HLA genotypes. Blood 2005; 105: 4878–84.

43. Cooley S, Weisdorf DJ, Guethlein LA, et al. Donor selection for natural killer cell receptor genes leads to superior survival after unrelated transplantation for acute myelogenous leukemia. Blood 2010; 116: 2411–9

44. The trial to reduce alloimmunization to platelets study group: Leukocyte reduction and ultraviolet B irradiation of platelets to prevent alloimmunisation and refractoriness to platelet transfusion. N Engl J Med 1997; 337: 1861–9.

45. Van Marwijk Kooy M, van Prooijen HC, Moes M, Bosma-Stants I, Akkerman JWN: The use of leukocyte-depleted platelet concentrates for the prevention of refractoriness and primary HLA alloimmunization. A prospective, randomised trial. Blood 1991; 77: 201–5.

46. Kroll H, Kiefel V, Santoso S. Clinical aspects and typing of platelet alloantigens. Vox Sang 1998; 74, suppl 2: 345–54.

47. Peters C, Minkov M, Matthes-Martin S, Pötschger U, et al. Leukocyte transfusion from rhG-CSF or prednisolone-stimulated donors for treatment of severe infections in immunocompromised neutropenic patients. Br J Hematol 1999; 106: 689–96.

48. Price TH, Bowden RA, Boeck M, Bux J, et al. Phase I/II trial of neutrophil transfusions from donors stimulated with G-CSF and dexamethasone for treatment of infections in hematopoietic stem cell transplantation. Blood 2000; 95: 3302–9.

49. Querschnittleitlinien zur Therapie mit Blutkomponenten und Plasmaderivaten. Herausgegeben von der Bundesärztekammer auf Empfehlung ihres Wissenschaftlichen Beirats. 4. Auflage, 2014. www.bundesaerztekammer.de)

50. Terasaki PI, McClelland JD. Microdroplet assay of human serum cytokines. Nature 1964; 204: 998.

51. McCalmon RT Jr, Tardif GN, Sheehan MA, Fitting K, Kortz W, Kam I. IgM antibodies in renal transplantation. Clin Transplant 1997; 11: 558–64.

52. Le Bas-Bernardet S, Hourmant M, Valentin N, Paitier C, Giral-Classe M, Curry S, Follea G, Soulillou JP, Bignon JD. Identification of the antibodies involved in B-cell crossmatch positivity in renal transplantation. Transplantation 2003; 75: 477–82.

53. Worthington JE, Martin S, Al-Husseini DM, Dyer PA, Johnson RW. Posttransplantation production of donor HLA-specific antibodies as a predictor of renal transplant outcome. Transplantation 2003; 75:1034–40.

54. Moers C, Pyrenne J, Paul A, Ploeg R. Machine perfusion or cold storage in deceased-donor kidney transplantation. N Engl J Med 2012; 366: 770–1.