Chicken macrophages can be universally studied in tissues by flow cytometry using the MRC1L-B antibody (9, 10, 60, 65, 66). Tissue-resident macrophages exhibit diverse functionality and can be defined based on their location in the organ. For example, in the chicken spleen, several macrophage subpopulations exist. These include periarteriolar lymphoid sheaths, resident macrophages, ellipsoid-associated macrophages, and red-pulp macrophages. The ability to segregate different macrophage subsets is difficult as specific markers for each population have yet to be identified. Splenic MRC1L-B⁺ macrophages universally express MHC class II, CD40, and CD80 and lack FLT3 expression (10, 67). Although mAbs against chicken CD83 and CD86 have been described, no convincing staining of mononuclear phagocytes has been demonstrated. In the liver, MRC1L-B⁺ macrophages can be segregated into MRC1L-B^Low TIM4^Hi cells and MRC1L-B^Hi TIM4^Low and MRC1L-B^Hi TIM4⁻ cells. Transcriptome analysis indicates that MRC1L-B^Low TIM4^Hi represent Kupffer cells, which are highly phagocytic compared to the MRC1L-B^Hi liver-resident macrophages (63). The TIM4 mAb JH9 stains a small population of CD3⁺ and Bu1⁺ cells in the liver and bursa, respectively (63).

Functional assays involving the assessment of phagocytosis can be integrated into flow cytometry experiments. Using commercially available fluorescent beads, which can be labeled with antigens such as LPS or inactive avian influenza virus, pH-sensitive pHrodo-labeled bioparticles, such as Salmonella or E. coli, or CFSE-labeled dead cells, can be utilized to determine the efficiency and specificity of chicken macrophage phagocytosis or effectorcytosis (10, 68–70). Performing phagocytosis assays at 4°C, a temperature commonly referred to as “on ice,” should be used to assess specific binding or adhesion of particles to cell surfaces without allowing active internalization (phagocytosis) to occur. This approach helps researchers differentiate between particles that are merely attached to the cell membrane and those that have been engulfed by the cell. It should be noted that 4°C control may not always be optimal for in vitro model antigen uptake studies. For example, 4°C control does not prevent bone marrow-derived macrophages from phagocytosing pH-sensitive pHrodo-labeled bioparticles. Instead, an actin polymerization inhibitor, cytochalasin D, inhibited the uptake of these bioparticles (70). The differential expression of surface markers, such as CD40 and MHC class II, can be an indicator of cell activation. In chickens, LPS-treated bone marrow-derived macrophages upregulate CD40 expression and downregulate MHC class II expression (71). In the chicken lung, MRC1L-B⁺ that phagocytosed LPS- or avian influenza-coated beads had significantly higher CD40 expression compared to cells that had taken up uncoated beads. The same study also demonstrated an increase in MHC class II expression by cells that phagocytosed LPS-coated beads (72). This observation is still to be determined for other tissue-resident macrophages, and more research is required to understand how infection alters the expression of these markers in a tissue-specific manner.

It has been well known that a small population of chicken splenic macrophages stain for chicken B-cell marker chB6 (Figure 4C) (47). Therefore, the BAFF-R mAb is a more specific reagent for chicken B cells (73). Of note, MRC1L-B may be sensitive to enzymatic digestion. To assess the impact of isolation techniques on MRC1L-B expression, researchers should examine immunohistology sections of their tissue samples to visualize the level/abundance of cells expressing the marker. Specifically, comparing the effects of non-enzymatic approaches to enzymatic methods can provide insights into how different isolation techniques influence MRC1L-B expression.

Generally, in mammals, DCs are defined by their expression levels of MHC class II, CD11c, and co-stimulatory molecules CD40 and CD86 (74). Using this general phenotype, researchers have sought to phenotype chicken DCs in this manner by flow cytometry. Vu Mahn et al. found that MHC class II⁺ putative CD11c⁺ MRC1L-B⁻ splenocytes express gene transcripts associated with mammalian cDC1 cells (75). The antibody against DEC-205, a marker for murine DC, was generated for chickens. Although useful in immunohistology, this antibody does not provide a strong staining pattern in flow cytometry (76). Recently, reagents against chicken FLT3, XCR1, and CSFR2 were developed, which aid to study chicken cDC without dependence on transgenic chickens (67). The anti-chicken FLT3 monoclonal, designated ROS-AV184, was found to label two cell populations in the spleen, FLT3^Hi and FLT3^Low cells. The FLT3^Hi cells, known as cDC1 cells, lack expression of the macrophage marker, MRC1L-B, and exhibit slightly lower levels of CD45 and MHC class II expression compared to the FLT3^Low cells. The FLT3^Low cells express MRC1L-B, making them macrophages. If receptor–ligand interaction is of high affinity, this offers an opportunity to analyse protein expression on cells through flow cytometry with fluorochrome-labeled ligands instead of antibodies. Recently, Wu et al. demonstrated that chicken XCL1-AF647 binds to XCR1 on FLT3⁺ cDC1 cells. In addition, to detect CSF2R expression, a CSF2-AF647 protein was generated and found not to stain chicken cDC1 cells (67). Together, this demonstrates that chicken cDC1 cells can be distinguished by staining for FLT3 or XCR1. However, if researchers do not have mAb FLT3, DC can be enriched by including MRC1L-B, CD3, and chB6 or BAFF-R staining with MCH class II to remove T and B cells and macrophages from MHC class II⁺ cell population (Figures 4A,B).

The maturation status of chicken cDC1 can be defined by their expression of CD1.1 using mAb CB3 clone (77). In the blood, a majority of XCR1⁺ cDC1 are MHC class II^Low and CD1.1^Hi, whereas small subpopulations have the MHC class II^Hi CD1.1^Hi or MHC class II^Hi CD1.1⁻ phenotype. In the spleen, these subpopulations are present from 1 week of hatch, with the MHC class II^Hi and CD1.1⁻ cells becoming the most abundant by 2-week post-hatch. It is hypothesized that the splenic XCR1⁺ MHC class II^Hi CD1.1^Hi cDC1 is derived from the blood XCR1⁺ MHC class II^Hi CD1.1^Hi cDC1 pool that lose CD1.1 expression as they mature and develop in the spleen (12).

The earliest B-cell-specific surface marker expressed on pre-bursal B cells is Bu1/chB6, with the first chB6⁺ cells becoming detectable simultaneously around embryonic day (ED) 10 in the embryonic bursa and spleen (79). As chB6 is strongly expressed at all stages of B-cell development except in differentiated plasma cells (47), it has become the most used marker for chicken B cells. The protein is a typical type I transmembrane protein with a highly glycosylated extracellular region and no recognizable similarity to known mammalian molecules. chB6 is recognized by several commercially available mAbs like AV20 and BoA1 (a cross-reactive guinea fowl antibody). It is an alloantigen with two alleles, Bu1a/chB6a and Bu1b/chB6b, which are recognized by anti-chB6a (clone L22) and anti-chB6b (clone 11G2) (80, 81). If using the allotype-specific antibodies to address all B cells, it is important to determine the presence of the alleles in the chicken line; otherwise, only a fraction of B cells might be stained.

Shortly after chB6 expression becomes detectable, pre-bursal B cells begin to express another pan-B-cell marker, the BAFF receptor (BAFF-R), recognized by mAb anti-BAFF-R clone 2C4 (73) and at ED14, chB6+ cells in the spleen are all BAFF-R+ (82).

Whilst most cells before their migration to the bursa are Ig-negative, very few pre-bursal cells have completed a productive BCR rearrangement and express surface Ig, detectable with anti-Ig Light chain or anti-IgM staining (82, 83). All pre-bursal B cells in the spleen express relatively high levels of CXCR4 and CXCR5 (82, 84); hence, migration of pre-bursal B cells can be mediated by their attraction toward CXCL12 and CXCL13L1-L3, respectively. Potentially also connected to their migratory behavior, pre-bursal B cells express sialyl-Lewis-X/CD15s but not Lewis-X/CD15 (85, 86). CD15s is a tetrasaccharide carbohydrate that is usually attached to O-glycans on the surface of cells and can mediate the interaction with selectins as the first step of leukocyte emigration from blood vessels. The chicken molecules can be detected with cross-reacting mAbs for the human molecules, e.g., mouse anti-human CD15s clone CSLEX1 and mouse anti-human CD15 clone W6D3 (85, 86).

Between ED9 and ED12, a small number of pre-bursal stem cells migrate to the bursa anlage and colonize the lymphoid follicles, where they start to strongly proliferate and diversify their BCR by gene conversion. From ED14 to 18, bursal B cells show a homogeneous expression of chB6, BAFF-R, MHC class II, CD40, CXCR4, and CXCR5 (36, 82, 84, 87, 88) and the initially small percentage of surface IgM-positive cells increases up to 50%. During their differentiation in the bursa, B cells lose CD15s expression and become CD15-positive (85, 86). Around hatch, the first B cells emigrate from the bursa into the periphery. The small fraction (ca. 5%) of emigrating cells amongst bursal B cells can be addressed as Ig Light chain⁺, MHC class II⁺, chL12⁺, and CXCR4^Low cells (see Figure 5A) (82, 89). ChL12, or the OV antigen, is recognized by mAb 11A9. The nature of the antigen is not known, but it should always be considered that it is an alloantigen, which is not recognised by 11A9 in every chicken line (90).

Post-emigration from the bursa, the immature B cells seed B-cell areas in secondary lymphoid organs, such as the peri-ellipsoidal white pulp in the spleen or the B-cell areas in caecal tonsils. Due to the special structure of the avian spleen, there is no histological discrimination between marginal zone and follicular B cells (91) and, to date, no markers have been described that would assign splenic B-cell subpopulations. All chicken B cells are CD5⁺ (92) hence, CD5-based discrimination of chicken B1 and B2 cells performed in mice and with reservations in humans is not possible. Splenic B cells are quite homogeneously BAFF-R⁺, Ig Light chain⁺, MHC class II⁺, and CD40⁺ (36, 83, 93). Frequently, chB6 expression is not completely homogenous; instead, especially in the spleen, a small immunoglobulin Light chain⁺ fraction expressing higher amounts of chB6 and a higher FSC can be observed (Figures 5B,C) (84). The BCR on the vast majority of cells (>95%) is an IgM isotype, with very few cells expressing a class-switched BCR of IgY or IgA isotype (Figure 5B). Noteworthy, not all commercially available anti-chicken IgY antibodies stain both soluble and membrane-bound IgY. Whilst clone 4D12 works optimally for ELISA and immunohistology, it does not stain membrane-bound IgY. However, anti-chicken IgY clone G1 stains both IgY variants (see Figures 5D,E) (94).

In the spleen and similarly in all larger secondary and tertiary lymphoid accumulations, encapsulated germinal centres (GCs) can be identified, consisting predominantly of GC B cells. It is important to be aware that in regular spleen preparations (mincing through a strainer), GCs and hence GC B cells will not be present. The GCs strongly stick to the residual artery tree and do not pass through the sieve without further measures. Imamura and colleagues have shown that chB6⁺ GC B cells can be obtained when the splenic artery tree with adjacent GCs is freed of the red pulp and separately digested with collagenase (95). Large GCs with a comparatively thinner capsule, which are not attached to an artery, are found in the caecal tonsils (96) hence, caecal tonsil preparations potentially contain GC B cells. To date, no markers are available to address these cells specifically. However, next to a high expression of chB6, immunohistochemistry of GCs revealed a weak positivity for CD57 (using the cross-reactive anti-human CD57 clone HNK-1) (97). In addition, HNK-1 works in flow cytometry, and as PWP B cells are chB6⁺ CD57⁻, it could serve as a marker to identify GC B cells in cell suspensions. Indeed, despite the lack of dead cells and doublet exclusion, older flow cytometric analysis of caecal tonsil cell suspensions found a chB6^high CD57⁺ cell population, which could readily represent GC B cells (97).

As with GC B cells, due to the lack of markers, it is currently not possible to identify memory B cells by flow cytometry. Here, new techniques like single-cell sequencing of B cells from different tissues and BCR sequencing will certainly help to identify these differentiation stages and potential new markers. Interestingly, scRNA sequencing of chicken blood leukocytes has identified several different B-cell subclusters (18). Though these have so far not been functionally assigned, it highlights the great value of this technique to identify chicken B-cell subpopulations.

Histologically, plasma cells, the final differentiation stage of B cells, can be identified by their typical morphology with a cartwheel nucleus structure. They have been identified in the spleen and in mucosa-associated lymphoid tissues (HALT, BALT, and GALT). Immunohistochemistry has also shown that in contrast to all other known chicken B-cell stages, plasma cells do not express chB6 or show only very weak staining (47). Downregulation of chB6 has also been shown by flow cytometry when B cells were differentiated toward a plasma cell phenotype in vitro by the presence of CD40L (93). Based on single color immunohistochemistry, it is also suggested that splenic plasma cells express CD57, and basic flow cytometry revealed a small population of large cells in the spleen and caecal tonsils expressing chB6^low CD57⁺ (97). However, this observation should be verified with further experiments, including additional markers. A rich source for plasma cells may be the Harderian gland, a lacrimal gland in the eye orbit. The gland reacts to intra-ocular vaccination, and it is described that leukocytes, including those with a plasma cell phenotype, can be isolated from the gland (98–101). Immunohistochemistry of the gut reveals a multitude of IgA-positive plasma cells in the lamina propria (LP), and with enzymatic digestion, it is possible to isolate IgA surface-positive cells (98).

Another approach to identify B-cell differentiation stages could be staining for transcription factors. Two key transcription factors for plasma cell differentiation are IRF4 and Blimp1/PRDM1. Whilst IRF4 induces plasma cell differentiation by directing immunoglobulin class switching, proliferation, and survival, BLIMP1 acts as a transcriptional repressor that represses B-cell features (102). For the rat anti-human-IRF4, clone 3E4 cross-reaction with porcine cells has been demonstrated (103) and preliminary data suggest cross-reaction with chicken cells (personal communication, Dr. W. Gerner). At least for one anti-Blimp1 antibody (mAb rabbit anti-human Blimp1, clone C14A4), cross-reactivity with the chicken protein in Western blots was demonstrated (104), and in mammals, this antibody was used in flow cytometry.

In contrast to mammals, where B-cell production in the bone marrow can be a lifelong process, the bursa Fabricii, and hence the chicken’s primary B-cell organ, involutes with sexual maturity (78). As analysis of older birds and studies with bursectomised birds have clearly shown the establishment of a bursa-independent dividing B-cell pool post-hatch, it is postulated that after bursa involution, the peripheral B-cell pool is maintained by post-bursal stem cells (89). Whilst after bursal emigration all B cells are chL12⁺, with increasing age, a chB6⁺/chL12⁻ B-cell population becomes detectable in the spleen, which might represent these cells (105). As chL12 detects an alloantigen, it may not be useful for birds lacking the allele. With the availability of new techniques and markers, these cells can now be further analyzed.

As antigen-presenting cells, all B cells are MHC class II⁺, but CD40L stimulation and mimicking T cells help further increase MHC class II (93). Whilst freshly isolated B cells from bursa, PBL, and spleen are CD80⁻, in vitro activation of B cells leads to strong upregulation of CD80 and also strongly increases CD25 expression (106, 107). As mentioned, regular spleen cell preparations do not contain GC B cells, so it is possible that, like human B cells, CD80 is a marker for activated and dividing chicken GC B cells. Up to one-third of freshly isolated splenic B cells are already CD57⁺ (97). This fraction is doubled by in vitro stimulation. Overall, activation of B cells can lead to an increase in already existing marker expression on all cells (MHC class II, CD25), complete de novo expression on a subset of B cells (CD80), or expression on an increased fraction of cells (CD57) (Figure 5F).

Avian T-cell subsets expressing the variants of the αβ TCR can be identified by staining with the commercial clones TCR2 (TCR αVβ1) and TCR3 (TCR αVβ2), respectively (116). TCR2⁺ and TCR3⁺ subsets differ in ontogeny and tissue distribution, with TCR2⁺ cells in general being more abundant than TCR3⁺ cells (117). Functional differences between TCR2⁺ and TCR3⁺ subsets are poorly described, but interestingly, TCR2⁺ cells but not TCR3⁺ cells migrate to the chicken intestine, hence being of importance to mucosal IgA production (118, 119). Several subsets of αβ T cells can be identified by staining for the co-receptors CD4 and CD8 (120). As opposed to chicken CD4 and CD8β, the chicken CD8α gene is polymorphic (121). Several antibodies exist (CT-8, 3–298, EP72, AV12-14) recognising chicken CD8α, but the 3–298 clone may be superior in being the only commercial reagent recognising the majority, if not all, of CD8α variants (122). In contrast to CD8α, only a single commonly used CD8β antibody exists (EP42). Within the αβ T-cell population in, e.g., peripheral blood, the following subsets exist: CD4⁺ CD8αα⁺, CD4⁺ CD8αβ⁺, and CD4⁻ CD8αβ⁺, in addition to the less well-characterized subsets of CD4⁻ CD8αα⁺ (123) and CD4⁺CD8αβ⁺ cells (124).

Chicken cytotoxic T lymphocytes (CTL) recognize peptides presented by MHC I molecules and show cytotoxic activity (125, 126). Interestingly, CD3⁺CD8⁺ cells in peripheral blood usually express the CD8αβ isoform, but a CD8αα-positive subset also exists and may expand, e.g., during viral infection (127). A common CTL assay in mammals is based on the detection of transient expression of lysosomal-associated membrane glycoproteins (LAMPs) on the cell surface. Cytotoxic activity via the perforin–granzyme pathway occurs in pre-formed lytic granules surrounded by lipid bilayers containing LAMPs that are fused with the plasma membrane. Hence, the LAMP-1 (CD107a) degranulation assay described above can also be used in studies of chicken cytotoxic T cells. Within the αβ T-cell population (αVβ1+ splenocytes), both CD8αα⁺ and CD8αβ⁺ showed potential to degranulate in vitro upon phorbol myristate acetate (114).

In addition to the granzyme/perforin killing pathway, evidence exists of a Fas/FasL pathway in chickens (128) and an anti-human FasL antibody (CD178, clone SB93a) was shown to cross-react with the chicken FasL by immunohistochemistry (129) but its suitability for flow cytometry is to be determined. A common parameter reported in relation to CTL responses is the production of IFN-γ. A number of monoclonal antibodies directed against chicken IFN-γ exist. Some only recognize the recombinant protein they were raised against, and others perform well in ELISA but are not suited for flow cytometry. Induction of IFN-γ production in splenocytes or PBMC by mitogens or specific antigens is reported to be detected by using antibody clones: 2B7, 11G5, 7E3, 12F12 (130, 131), mAb80 (132), and a rabbit polyclonal anti-chIFN-γ reagent (133). Clones 12F7, 12D4 (130) and EH9 (134) seem less suitable for intracellular staining (ICS) and flow cytometry. The anti-chicken IFN-γ antibody (clone 5C.123.02/08) from the commercial chicken IFN-γ Invitrogen ELISA kit works for intracellular staining (ICS) of the recombinant protein expressed by CHO cells (133). However, although some report staining of the native protein using these antibodies, others observe only weak staining with the ELISA reagents (135) or fail to reproduce even a dim signal (136). Additional clones MT6C2 and MT7C1 from Mabtech, as well as four Chinese clones, were found not suitable for ICS (136, 137). Unfortunately, the two superior antibody clones for intracellular staining and flow cytometry, mAb80 and 5G11, are not commercially available.

Several reagents recognising chicken CD4 are available, e.g., the clones CT4, EP96, AV29, and 2–35. In other species, CD4 may also be expressed by monocytes, but this is not the case in chickens (138). However, a small CD3⁻ CD4⁺ NK population (with slightly higher FSC/SSC than resting lymphocytes) is sometimes identified in peripheral blood (54) and hence at least TCRαβ or CD3 in combination with CD4 should be used to identify chicken Th cells. Interestingly, some lines of chickens have a high abundance of CD4⁺CD8⁺ double-positive TCRαβ⁺ lymphocytes, and there is a genetic influence on levels in peripheral blood but not necessarily in the intestine (121, 123, 138). The double-positive subset exists either as CD4⁺CD8αα⁺ or CD4⁺CD8αβ⁺ with a dimmer CD8 signal than CD4⁻ CD8⁺ cells (127); hence, using bright fluorochromes for CD8 detection is crucial to obtain good separation between CD8⁻, CD8^dim, and CD8^Hi subpopulations (Figure 6).

In mammals, major Th subsets can be differentiated by intracellular staining for the transcription factors T-bet (Th1), GATA-3 (Th2), RORγt (Th17), and Foxp3 (Treg). To the best of our knowledge, no reagents are available for staining important chickens Th transcription factors, despite the obvious value of developing such reagents for flow cytometry. The success rate of identifying cross-reacting mammalian reagents is generally poor for chicken surface markers (139, 140), but for highly conserved intracellular proteins such as transcription factors, the chances may be higher. However, testing of two widely used anti-murine Foxp3 clones, FJK-16 s and MF-14, proved they were unsuitable for Foxp3 detection in chicken cells (141).

The basic Th1 response system appears to be conserved in chickens (142), and CD4⁺ cells producing IFN-γ are often interpreted as Th1 cells. Indeed, the Th1/Th2 paradigm was early on made probable through gene expression analysis of tissue from Newcastle disease virus and Ascaridia galli, infected chickens, respectively (143). However, flow cytometry studies addressing production of multiple cytokines and linking Th1 or Th2 profiles to, e.g., TCR αVβ1 or TCR αVβ2 expression are still missing. The limited quality of the chicken cytokine antibodies is often a problem that has inspired the use of alternative methods such as identification of intracellular cytokine RNA by the PrimeFlow^™ system (130).

Putative chicken regulatory T cells that can suppress T-cell proliferation in vitro were identified by co-expression of CD4 and CD25 (144). However, the CD4⁺ CD25⁺ population includes other subsets than just Tregs. More recently, FOXP3 was identified in the chicken genome, and Foxp3 mRNA was shown to be abundant in CD4⁺ CD25⁺ in contrast to CD4⁺ CD25⁻ subsets in the spleen and caecal tonsils (141). The gene expression studies identified two CD4⁺ CD25⁺ subsets where the cells expressing high levels of IL-10 and Foxp3 were suggested as mature Tregs, whereas cells expressing low levels of IL-10 in combination with IL-2 were rather activated Th cells (141). Staining with CD4 and CD25 will hence provide a mixed population, and the production of chicken Foxp3 antibodies is expected to give better opportunities for studying Treg subsets in the future. Several chicken CD25 antibodies exist where AV142 and the bivalent human recombinant Fab AbD13504 are widely used. In addition, clones like 6C9, chCD25-32, and chCD25-54 exist (145), as well as 28–4, which was originally described as an NK cell marker but later identified as recognising CD25 (46).

Various in vitro and in vivo models have shown chicken IL-17 mRNA gene expression and suggested the presence of Th17 cells (146–149) but only recently have monoclonal antibodies useful for ICS been developed, namely the 1E7 clone recognising IL-17F (with slight cross-reactivity to IL-17A) and the two IL-17A-specific clones 9F11 and 10D5 (150). The IL-17 antibodies were all able to stain a small population of CD4⁺ splenocytes upon PMA activation. The 10D5 clone was furthermore used to show that IL-17A was primarily expressed by CD3⁺ CD4⁺ T cells in the spleen and PBMC, but staining of smaller subsets of γδ T cells was also evident (151). Interestingly, IEL staining patterns were slightly different with IL-17A⁺ cells largely CD4⁻, CD8⁻, and TCR1⁻ but for the most part expressing CD3 and CD25 (150). In addition, the same antibody was used to show weak signals of IL-17A in lung T-cell populations (131).

The commercially available antibody TCR1 recognises the TCRγδ variant, and hence all TCR1⁺ cells are actually γδ T cells (111). However, whether all γδ TCRs indeed express the TCR1 epitope is still not proven but might be solved in the near future with the availability of new TCR sequencing protocols (152). As shown by frequency within a lymphocyte gate, γδ T cells are abundant in peripheral blood, immune organs, and bone marrow, and a CD8⁺ subset is often seen in the lung and spleen (137), as well as in the intestinal mucosa (153). In the intestine, γδ T cells are present both in the intraepithelial and the lamina propria areas, and interestingly, in addition to the CD4⁻ CD8⁻, CD4⁻ CD8αα⁺, and CD4⁻ CD8αβ⁺ populations, the presence of an additional small CD4⁺ CD8⁻ population has been suggested. The CD8⁺ population has received much attention and has been reported as being slightly larger in size and more prone to activation by mitogens (154). Interestingly, chB6-positive intraepithelial lymphocytes in the small intestine contain a population of TCR1⁺ cells, as shown in Figure 3.

Some molecules are differentially expressed between γδ T cells and αβ T cells but cannot be used as unique lineage identifiers. For example, CD5 was shown to be expressed on virtually all CD4⁺ αβ T cells and on the majority of γδ T cells, but the mean fluorescence intensities were low/intermediate on the γδ T cells isolated from spleen and peripheral blood (92). The CD5 antibody clone 2–191 unfortunately appears to be discontinued. CD28 is another molecule expressed on virtually all CD4⁺ αβ1 T cells but is absent from the majority of γδ T cells; however, by using the clone MoAb 2–4, a small subset was found to be CD28⁺ (mostly CD8α⁺ but also a minor CD8⁻ population) (155). The clone used by Koskela et al. may no longer be accessed, but clone AV7 recognises CD28 and is commercially available (156).

Data from scRNA-seq suggest the presence of multiple γδ T-cell subsets that may represent either phenotypic subsets or differentiation and activation states (18). A range of molecules may be differentially expressed by various γδ T subsets, but a comprehensive multiparameter immunophenotyping study has not yet been published. As the percentage of TCR1+, TCR2+, and TCR3⁺ cells does not completely add up to 100% of the CD3 population, there may be a small yet unidentified T-cell subpopulation. Hence, interesting surface marker antibodies for a future multiparameter staining panel may include TIM4 (ROS-JH9 (60)), SLAMF4 (8C7 (35)), CD25 (AbD13504 (157)). Moreover, chicken γδ T cells have the ability to secrete a number of cytokines such as IFN-γ, IL-17A, IL-6, IL-10, and IL-13 (151, 158, 159), and including cytokine staining in multiparameter staining is of value to characterize various γδ T-cell subsets.

Extensive knowledge about T-cell activation and memory cell markers is still lacking in the avian research field. In mammals, activated proliferating T cells express several molecules that are expressed to a lesser extent or even absent on resting cells, including various chemokine receptors, adhesion molecules, co-stimulatory molecules, and MHC antigens (160). The same appears to be the case for chicken lymphocytes, but most of the published data include observations made using in vitro polyclonal/mitogen-stimulated cells rather than in vivo activated cells from infected animals. For example, in vitro ConA-induced T-cell proliferation of PBMC confirmed CD25 and MHC class II as T-cell activation markers for both CD4⁺ and CD8α⁺ cells and CD28 only for CD8α⁺ cells when looking at activation marker-positive frequencies of cells (6). Interestingly, most of the tested putative activation markers (e.g., CD44, CD45, CD25, and CD28) showed increased surface expression (mean fluorescent intensity, MFI) over time, whereas the MFI of MHC class II was upregulated only 24-h post-stimulation, followed by MFI downregulation, especially in CD8α⁺ T cells, where the MFI stayed below baseline from 48- to 96-h post-stimulation.

In mammals, constitutive expression of MHC class II is confined to professional antigen-presenting cells, including DCs, B cells, monocytes, and macrophages, and upon activation, MHC class II expression (all isotypes) is also seen on the surface of T cells in various species except in mice (161). The Naghizadeh chicken PBMC study mentioned above (160) showed virtually no MHC class II expression pre-stimulation, but 24-h post-stimulation, it was readily induced in 20–25% of the CD4⁺ and CD8α⁺ cells. Although the CD8α⁺ antibody would have picked up a small subset of γδ T cells, the majority of γδ T cells (CD8⁻) were excluded from the mentioned study. Interestingly, others have reported that the majority of γδ T cells in peripheral blood express MHC class II even in a resting state, and increased MFI was shown in an in vivo experiment where chickens were provided high doses of Ulvan in their drinking water (162). CD25 is constitutively expressed on a subset of γδ T cells but is also described as an activation marker because frequencies of CD25⁺ cells within the γδ T population are increased upon activation with, e.g., Salmonella and Eimeria (124, 157). Furthermore, chickens immunized with the model antigen mycobacterial purified protein derivative or Mycobacterium tuberculosis sonicate showed increased frequencies of CD28⁺ γδ T-cell frequencies combined with increased surface expression of CD28, CD5, CD25, and MHC class II (155).

CD57 (clone HNK-1) has been identified as a B-cell activation marker (44, 97, 163, 164). We have shown higher degranulation of CD57⁺ CTLs than of CD57⁻ CTLs, which supports CD57 as chicken T-cell activation marker (165). Furthermore, frequencies of CD57⁺ cells increase within both αβ and γδ T-cell populations in PBMC stimulated with mitogens or anti-CD3 (T. Dalgaard, manuscript in preparation). Several additional potential activation markers are available, but their expression on activated chicken T cells is poorly characterized. Examples include CD276 (AV95/EH7), CD30 (AV37 (166)), CRTAM (8A10 (167)), and GITR (9C5 (168)). In addition to activated T cells, chicken memory cells are also poorly defined in terms of phenotypes. Despite the fact that different CD45 isoforms exist (29), the equivalents of CD45RA and CD45RO have not been identified in chickens. CD127 is the α chain of the IL-7 receptor, and in mammals, it is differentially expressed depending on T-cell differentiation state (naïve, effector, memory) (169). A monoclonal antibody against chicken CD127 exists (clone 8F10E11) and was used to show that the majority of CD4⁺ cells in the peripheral blood and spleen of healthy animals expressed CD127. In contrast, only 10–60% of the CD8α⁺ cells expressed CD127, and the frequency declines with age (170). To establish the value of this marker for the discrimination of T-cell differentiation stages, CD127 staining must be further investigated, e.g., on tetramer-positive cells and in conjunction with other activation or memory cell markers. In other species, CD44 expression is higher on effector and memory T cells as compared to naive. This is not yet established for chicken cells, although some studies suggest increased CD44 expression in the memory stages of in vivo vaccine or challenge experiments (5, 171).

To evaluate host–pathogen interaction and vaccine responses, it is important to understand the role of antigen-specific T cells. Few avian MHC multimer flow cytometry reports exist (172, 173) and the reagents are not commercially available in various MHC–peptide combinations as they are for human and mouse models. Hence, most analyses of avian T-cell biology have relied on proliferation, degranulation, or cytokine detection in PBMC or splenocytes after antigen re-stimulation in vitro/ex vivo for quantitative and qualitative studies of antigen-specific T-cell responses. Interestingly, a subset of activation markers in mammals are exclusively expressed by T cells activated via MHC–TCR interaction, like the commonly used marker CD154 (CD40L) for antigen-activated T cells in humans and mice (174, 175). Transient CD154 surface expression or intracellular expression stabilized with the secretion inhibitor Brefeldin can be exploited for enrichment of antigen-specific T cells before further analysis (176, 177). An avian orthologue of mammalian CD154 exists, and the gene was identified and a set of monoclonal antibodies named AV71-76 were generated (UK Immunological Toolbox). Initial analysis of splenocytes activated by PMA/IO showed binding of the antibodies and suggested CD154 expression on activated T cells. In this early study, it was evident that the IgG1 antibodies AV71 and AV74 bound weakly to CD154, whereas the IgG2a antibodies AV72, AV73, and AV75 showed slightly higher affinity; however, further in-depth analysis is warranted (36).