The fundamentals of neutrophil antigen and antibody investigations

Authors


  • 4A-S12-01

Yoke Lin Fung, Critical Care Research Group, The University of Queensland and The Prince Charles Hospital, Brisbane, QLD 4032, Australia
E-mail: ylfung@uq.edu.au

Abstract

Background  The neutrophil population makes up the largest proportion (40–70%) of white blood cells and is crucial to our innate immune response. An unexpectedly significant proportion of neutrophils (∼28%) is located in the pulmonary circulation. This is an immune system front line wherein neutrophils guard an important interface between the external environment and the body. Neutrophils migrate to sites of infection and/or inflammation, where they phagocytose and destroy foreign microorganisms and mediate inflammation and eventual healing. Therefore, a decrease in neutrophil numbers often results in increased susceptibility to infection. Sustained or transient immune neutropenia occurs when neutrophil-reactive alloantibodies or autoantibodies destroy neutrophils, or significantly compromise their function. Neutrophil-reactive antibodies are capable of activating the neutrophil’s microbicidal arsenal with resultant damage to the surrounding tissue.

Antigens and Antibodies  There are currently eight well characterized human neutrophil antigens (HNA): HNA-1a, HNA-1b, HNA-1c, HNA-2a, HNA-3a, HNA-3b, HNA-4a and HNA-5a. Only HNA-1a, HNA-1b and HNA-1c are specific to neutrophils, whilst the others are also expressed on other cell types. It is important to remember that neutrophils also express HLA Class I antigens, and that there is some evidence that adequately activated neutrophils may express HLA Class II antigens. This provides an interface between the innate and adaptive immune response. Therefore, both HNA and HLA Class I and II antibodies are capable of reacting with neutrophils in vivo and in vitro.

Routine Antigen and Antibody Investigation  Antigen genotyping protocols are now available for HNA-1a, HNA-1b, HNA-1c, HNA-3a, HNA-3b, HNA-4a and HNA-5a. The serological examination of neutrophils is reliant on harvesting a representative, pure population of neutrophils, which is difficult because neutrophils are short lived (6–8 h) and fragile. The International Granulocyte Immunobiology Workshop recommends the use of both the granulocyte immunofluorescence test (GIFT) and granulocyte agglutination test (GAT), complemented by the monoclonal antibody immobilization of granulocyte antigen (MAIGA) assay. The GIFT uses paraformaldehyde-fixed neutrophils and detects the binding of antibodies to surface HNA and HLA antigens. Serological reaction in the GAT is reliant on viable neutrophils. It demonstrates the ability of antibodies to sensitize neutrophils and stimulate them to undergo chemotaxis towards other sensitized neutrophils, thus forming microscopic agglutinates. The MAIGA is useful in confirming antibody specificities as it employs monoclonal antibodies to known HNA epitopes. To screen for neutrophil-reactive antibodies, serum or plasma is exposed to a screening panel of neutrophils in the GIFT and GAT. To differentiate between HNA and HLA antibodies, the simple strategy of using platelet pools to absorb out HLA antibodies can be used. Specificity is determined using the GIFT and GAT, and a phenotyped and genotyped neutrophil panel and confirmed with the MAIGA.

Application The combination of the GIFT, GAT and MAIGA was originally developed for the reliable investigation of immune neutropenias. However, in the last decade these techniques have become fundamental to the investigation of immune mediated TRALI where neutrophils are a key target and effector cell.

Introduction

Antibodies directed against neutrophils can cause their destruction and result in alloimmune or autoimmune neutropenia. From the 1960s to the late 1970s, it was cases of alloimmune neonatal neutropenia (ANN) and autoimmune neutropenia of infancy which triggered investigations leading to the development of reliable serological techniques for detecting antibodies to neutrophils. Subsequently, the first recognized human neutrophil antigens were serologically characterized [1–3].

Antibodies directed against neutrophils can also stimulate or activate neutrophils; this mechanism is often implicated in the pathogenesis of transfusion-related acute lung injury (TRALI) [4]. Neutrophil-reactive HNA and HLA Class I antibodies implicated in TRALI are thought to agglutinate and/or activate neutrophils in the pulmonary vasculature, resulting in damage to the immediate surrounding tissue and consequently symptoms of respiratory distress [5]. There is now a growing evidence base for neutrophils as key effector cells in TRALI [4], and this has been the main driver for the current interest in new mass screening approaches for neutrophil-reactive antibodies.

Neutrophil antigens and antibodies

Today, there are five known HNA systems encompassing eight well characterized antigens (Table 1), which are designated according to the internationally agreed granulocyte antigen nomenclature system [6].

Table 1.   HNA systems and their characteristics
Ag SystemHNA-1 (15-25)HNA-2 (26-33)HNA-3 (34-42)HNA-4 (43,44,46)HNA-5 (44,45,47)
AgHNA-1aHNA-1bHNA-1cHNA-2aHNA-3aHNA-3bHNA-4aHNA-5a
Old nomenclatureNA1NA2SHNB15b MARTOND
Gylcoprotein (Cluster)FcγRIIIb (CD16b)gp 50-64 (CD177)CTL-2 (undefined)αM (CD11b)αL (CD11a)
Anchor/linkglycosyl-phosphatidylinositol link (GPI)GPItransmembranetransmembranetransmembrane
ExpressionNeutrophilsNeutrophils (monocytes of pregnant women)Neutrophils, granulocytes, monocytes, lymphocytes, platelets, kidney, placental tissue, spleen, lymph node tissue, endothelial cellsGranulocytes, monocytes, T-lymphocytesGranulocytes, monocytes, lymphocytes

Antigens within the HNA-1 system are exclusive to neutrophils and are located on the FcRIIIb receptor (CD16b) which, uniquely for the Fc receptor family, is anchored to the membrane by glycosyl-phosphatidylinositol (GPI), rather than through a transmembrane link [7]. Consequently, when bound by IgG, CD16b must co-operate with other transmembrane proteins such as FcRII to generate an intracellular signal to activate neutrophil phagocytosis, degranulation or the respiratory burst [8,9]. Neutrophils express CD16b relatively late in their maturation at the metamyelocyte stage [10]. In normal, resting neutrophils, 50% of CD16b occurs on the microvilli and the remainder is located in the Golgi complex and in many small vesicles [11,12]. Pathological or physiological activation of neutrophils results in proteolytic cleavage of CD16b with its replacement by translocation from intracellular compartments. Therefore, CD16b and HNA1 expression is dynamic [12–14].

The frequency of HNA-1a is higher in most Asian populations, and ranges between 0·300 and 0·680 [15]. Within the Asian population, the Chinese Tajik and Asian Indians have a lower HNA-1a expression of 0·350 and 0·300, respectively [15]. The frequency of HNA-1b is higher in Caucasians (0·627 to 0·718) [15]. The lower frequency HNA-1c has also been described [16]. Antibodies to the HNA-1 system antigens are commonly associated with both alloimmune and autoimmune neutropenia [17–20] and have also been implicated in TRALI events [21,22]. The differing population frequencies for HNA-1a and HNA-1b, which both appear to be highly antigenic, should be borne in mind when investigating a potential ANN in a neutropenic newborn from a cross cultural family. There is evidence of differing functional capacities between alleles of the HNA-1 system, such as the reduced in vitro Fc-mediated phagocytosis by HNA-1b neutrophils [23,24]. This is important to remember, because the binding of cognate antibodies to these antigens in vivo can result in severe functional impairment as well as neutropenia [25].

HNA-2a appears to be exclusive to neutrophils [26] except for one report which described transient HNA-2a expression on the monocytes of women during their first trimester of pregnancy [27]. This is consistent with reported increased HNA-2a expression on neutrophils during pregnancy [28]. HNA-2a is located on a 58–64 kDa glycoprotein, CD177, which is GPI linked and located on the plasma membrane and intracellular membranes of the small vesicles and specific granules of neutrophils [29]. It is apparently expressed on only a sub-population of neutrophils [30]. The frequency of HNA-2a in the population is high (Table 1) and antibodies to this antigen have been implicated in immune neutropenias [31,32], transfusion reactions and TRALI [21,33].

Recently, HNA-3a and its allele, HNA-3b were characterized [34], and this has enabled genotyping for this system [35]. In the German population, the HNA-3a antigen frequency is 97%, with gene frequencies of 0·792 and 0·207 for HNA-3a and HNA-3b, respectively [35]. Antibodies to HNA-3a are implicated in immune neutropenia [36,37], but in the last two decades have gained notoriety for their involvement in severe and often fatal TRALI events [38–41]. These antibodies are predominantly leucoagglutinating, causing bulky cellular aggregates in vitro by binding to neutrophils via their F(ab) ends [34]. This may echo their in vivo behaviour and explain their association with the most severe TRALI reactions. Consequently, these antibodies are best detected using the granulocyte agglutination test (GAT) [42].

HNA-4a is located on the αM subunit (CD11b) of the C3bi receptor (CD11b/CD18) and is expressed on granulocytes, monocytes and natural killer cells (Table 1) [43,44]. HNA-5a is located on the αL (CD11a) subunits of the LFA-1 complex, expressed on all leukocytes [44,45]. Antibodies to both HNA-4a and HNA-5a were originally identified in sera respectively from a multiparous blood donor [43] and a multi-transfused male patient [45] indicating that the antigens are immunogenic. However, the clinical relevance of antibodies to these antigens is still unclear in spite of the functional importance of their carrier glycoproteins. HNA-4a and HNA-5a antibodies have not yet been implicated in transfusion reactions, although ANN cases have been associated with each antigen [46,47].

Investigation

The most widely used, accredited approach for detecting neutrophil-reactive antibodies is the combined use of the granulocyte immunofluorescence test (GIFT) and GAT. Both of these techniques have recently been described in detail [48]. In the GIFT, antibodies binding to neutrophils are labelled with a conjugated anti-human globulin which can then be visualized either by fluorescence microscopy or more usually, using flow cytometry [3,49]. GAT reactions rely upon a very different principle, because they depend on the ability of IgG antibody to sensitize and stimulate neutrophils, which results in migration to and agglutination of adjacent neutrophils [50]. Thus, viable and functional neutrophils are critical to successful GAT results. Results from the International Granulocyte Immunobiology Workshop have repeatedly demonstrated that the GAT is vital for the effective detection of antibodies to HNA-3a [42].

The fact that neutrophils live only for 6–8 h and are fragile, make it essential that fresh blood is used for neutrophil isolation. The principle of neutrophil isolation is that unwanted cell types are physically separated and removed, leaving behind a purified, functional pool of neutrophils [3,51,52]. In our laboratory, we centrifuge EDTA or citrated whole blood (1000 g for 10 min) and remove platelet-rich plasma. Dextran (5% solution) is mixed with the remaining leucocyte-rich mixture and this is incubated at 37°C for 30 min at an angle to promote rapid red cell sedimentation. The leucocyte-rich plasma is then transferred into a fresh conical tube, and underlayed with 1·077 and 1·114 density gradient solutions. After centrifugation at 1600g for 15 min, remaining red cells sediment at the bottom of the tube, granulocytes are retained at the lower gradient interface and mononuclear cells are retained at the top gradient interface. Harvested granulocytes are washed with buffer prior to their use in the GIFT, GAT or monoclonal antibody immobilization of granulocyte antigen (MAIGA).

Screening

The detection of neutrophil-reactive antibodies usually involves testing serum or plasma with a panel of neutrophils using the GIFT and GAT. Initial testing may be done using a small random panel of un-typed, normal neutrophils. The specificity of any samples with positive reactions can then be characterized using a typed neutrophil panel to test samples absorbed and unabsorbed with a platelet pool to remove contaminating HLA antibodies.

Determining the HNA antibody specificity

This requires testing the sample against a genotyped or phenotyped panel of neutrophils by GIFT and GAT. The typed neutrophil panel should include cells confirmed positive and negative for the various HNAs. To phenotype neutrophils it is essential to use well characterized anti-HNA sera. Such sera will have been serologically well defined only after extensive testing with multiple typed panel cells, although this is often difficult to acquire. The availability of published genotyping protocols for HNA-1 [53], HNA-3 [35], HNA-4 [54] and HNA-5 [55], and commercial HNA genotyping kits have made it possible for more laboratories to develop very well characterized panels of HNA typed neutrophils.

The MAIGA provides an additional means of defining antibody specificity, using monoclonal antibodies (MoAb) specific for the known carrier glycoproteins (Table 1) displaying the genetic polymorphisms of the target antigens [48,56]. Positive MAIGA results can confirm antibody specificity but negative MAIGA results do not preclude the presence of a specific antibody. Besides the technically challenging nature of this assay, a range of monoclonal antibody clones must be used, where possible, for each carrier glycoprotein. This avoids the possibility of the human polyclonal antibody binding at an epitope identical to that targeted by the monoclonal capture antibody and thus hindering its binding.

Differentiating HNA from HLA

As neutrophils also express HLA, it is essential to differentiate between positive serology caused by HNA and HLA antibodies. Because platelets express large amounts of HLA, a simple approach is to use a pool of platelets to absorb out HLA antibodies in the sample [57]. However, because HLA types vary greatly between individuals, it is important to use an adequate, representative platelet pool from multiple donors. Briefly, the test sample is incubated with a pool of washed, packed platelets, allowing time for the platelets to absorb any HLA antibodies. After incubation, the test sample is centrifuged and the platelets removed, and the absorbed sample is tested for HNA antibodies. Positive results with both unabsorbed and platelet absorbed sera, indicates an HNA specific antibody, whilst positive results only in the unabsorbed sera indicates HLA specificity. Another approach is to use an HLA antibody in the MAIGA assay to detect HLA antibodies.

Conclusion

The validated techniques (GIFT, GAT and MAIGA) for the serological investigation of neutrophil antibodies are technically demanding and labour intensive, and in their present form do not support mass screening. The tests have evolved through many technical incarnations to be increasingly sensitive and specific for the detection and characterization of antibodies in clinical scenarios in which an immune mechanism is well understood. It has to be said that the primary and secondary autoimmune neutropenias in adults, drug-induced immune neutropenias and TRALI do not fit easily into this category. A number of groups are accelerating their efforts to develop fast and stable platforms for mass screening of samples for HNA antibodies in donors and patients who can have complex serology from a range of unknown or known immunological stimuli. These new testing platforms can generally be characterized as cell-based or antigen-based platforms [48]. A number of these technologies are promising but all still require significant validation to ensure that they have the requisite sensitivity and specificity for their purpose and each has its own unique limitations [48].

The ability to competently perform GIFT, GAT and MAIGA assays will enable laboratories to investigate in detail, any clinical scenario where neutrophil-reactive antibodies may be involved. These include alloimmune neonatal neutropenia, chronic benign autoimmune neutropenia of infancy, febrile non-haemolytic transfusions, TRALI and some adult autoimmune neutropenias. The Granulocyte Immunobiology Working Party of the ISBT conducts annual International Granulocyte Immunobiology Workshops for granulocyte reference laboratories (http://www.isbtweb.org/working-parties/). Laboratories intending to develop neutrophil antigen and antibody detection abilities may benefit by collaborating with and being mentored by one of these granulocyte reference laboratories.

Disclosures

No potential conflict of interests to declare.