The genomic organization and expression pattern of the low-affinity Fc gamma receptors (FcγR) in the Göttingen minipig
- 384 Downloads
Safety and efficacy of therapeutic antibodies are often dependent on their interaction with Fc receptors for IgG (FcγRs). The Göttingen minipig represents a valuable species for biomedical research but its use in preclinical studies with therapeutic antibodies is hampered by the lack of knowledge about the porcine FcγRs. Genome analysis and sequencing now enabled the localization of the previously described FcγRIIIa in the orthologous location to human FCGR3A. In addition, we identified nearby the gene coding for the hitherto undescribed putative porcine FcγRIIa. The 1′241 bp long FCGR2A cDNA translates to a 274aa transmembrane protein containing an extracellular region with high similarity to human and cattle FcγRIIa. Like in cattle, the intracellular part does not contain an immunoreceptor tyrosine-based activation motif (ITAM) as in human FcγRIIa. Flow cytometry of the whole blood and single-cell RNA sequencing of peripheral blood mononuclear cells (PBMCs) of Göttingen minipigs revealed the expression profile of all porcine FcγRs which is compared to human and mouse. The new FcγRIIa is mainly expressed on platelets making the minipig a good model to study IgG-mediated platelet activation and aggregation. In contrast to humans, minipig blood monocytes were found to express inhibitory FcγRIIb that could lead to the underestimation of FcγR-mediated effects of monocytes observed in minipig studies with therapeutic antibodies.
KeywordsCD32 FcγRIIa FCGR locus Flow cytometry Single-cell RNA sequencing Sus scrofa
Therapeutic antibodies of the IgG (immunoglobulin G) isotype represent an important group of new medical entities and interactions of Fc gamma receptors (FcγRs) with the Fc part of IgG antibodies are crucial in the antibody-based immunotherapy. Most mammals were shown to have three functionally distinct classes of FcγRs with different affinities and properties. FcγRIa (CD64) is capable of binding to free IgG antibodies and is hence considered as a high-affinity receptor. Its expression and function are conserved in most mammalian species, including pigs (Akula et al. 2014; van der Poel et al. 2011). Low-affinity receptors efficiently bind immune complexes and are divided into inhibitory and activating FcγRs. The structure and function of FcγRIIb (CD32b), the inhibitory low-affinity receptor, is also highly conserved in humans, pigs, mice, and other mammalian species (Akula et al. 2014). FcγRIIIa (CD16a) is an activating low-affinity FcγR that requires the association with FcR γ-chain (Fc receptor common gamma chain) for signaling (Kim et al. 2003). Different affinities to IgG were observed for the human FcγRIIIa V158F polymorphism within the extracellular domain (ECD). It was shown to be associated with differential response to therapeutic antibodies and disease progression (Mellor et al. 2013). Although FcγRIIIa is the most widely analyzed Fc receptor in pigs (Halloran et al. 1994), its gene structure and genetic localization has not yet been determined. In mouse, the orthologous receptor of FcγRIIIa is known as FcγRIV (Nimmerjahn and Ravetch 2006). FcγRIIa (CD32a) is another activating low-affinity receptor present in humans, non-human primates (NHPs), cattle, and rat and named as FcγRIII in mouse (Lux and Nimmerjahn 2013). In humans, FcγRIIa is expressed on the cell surface of monocytes, neutrophils, macrophages, eosinophils, basophils, dendritic cells, and platelets. It is involved in the process of phagocytosis, antibody-dependent cellular cytotoxicity (ADCC), and cytokine release (Powell and Hogarth 2008). The FcγRIIa R131H polymorphism is associated with severity and progression of idiopathic pulmonary fibrosis and with response to rituximab therapy (Bournazos et al. 2010; Ziakas et al. 2016). Immune complexes binding to FcγRIIa on human platelets can lead to thrombus formation (Zhi et al. 2015) and ultimately to heparin-induced thrombocytopenia (Greinacher 2009). Despite its importance, the minipig FcγRIIa and its gene FCGR2A could not be identified yet.
The Göttingen minipig is increasingly used as a valuable animal model for preclinical pharmacology and drug safety studies. The high similarity to humans in terms of genetics, genomics, physiology, and anatomy makes the minipig a desired alternative to NHPs (Ganderup et al. 2012). Additionally, Göttingen minipigs have a controlled health status, are easy to handle, and need less food, space, and pharmacological products compared to domestic pigs and other non-rodent species (McAnulty et al. 2011). Minipigs mainly differ from domestic pigs in their growth range and size at sexual maturity but not in anatomical structures (Swindle et al. 2012). Regarding the immune system, no major differences between pigs and minipig have been reported so far but detailed studies are lacking (Descotes et al. 2018). The use of the minipig as an adequate species for toxicity and efficacy evaluation of therapeutic antibodies requires a detailed knowledge of the FcγR composition and their interaction with human IgGs. However, to date, the knowledge on the binding properties of porcine FcγR to human antibodies is still scarce. In addition, the number of low-affinity FcγRs existing in the minipig and the allocation of the FCGR genes in the corresponding locus of the Göttingen minipig genome was not conclusively determined. The latest version of the Göttingen minipig genome was generated by Heckel et al. by mapping of the whole genome-sequencing data on the Duroc pig genome Sus scrofa 10.2 (Heckel et al. 2015). There, FCGR2B was the only gene annotated in the low-affinity FCGR locus. Recently, the new assembly Sus scrofa 11.1 was released containing a more accurate view of the pig genome including this particular locus (Li et al. 2017).
In this paper, we describe the complete assembly of the genetic FCGR locus of the Göttingen minipig including the exact mapping of FCGR3. Additionally, we demonstrate the identification, sequence characterization, and genomic location of FCGR2A, and the expression of low- and high-affinity FcγRs in the Göttingen minipig across blood cell types.
Materials and methods
FCGR locus assembly and FCGR mapping
Genomic DNA was isolated from the frozen spleen of a Göttingen minipig using the DNeasy Blood and Tissue Kit (Qiagen). PCR on genomic DNA with primers JE24/JE26 (see Fig. 1 for primer positions and Online Resource 1 for primer sequences) allowed sequencing of the gap within an intron of FCGR3A (Fig. 1) (GenBank ID: MH574548). The two remaining gaps in the putative FCGR2A introns were amplified by nested PCR using primers JE62/JE64 followed by JE47/JE49 and JE58/JE61 followed by JE41/JE42, respectively (Fig. 1, Online Resource 1). The obtained products were cloned using the TOPO TA cloning kit and sequenced (GenBank ID: MH574549, and MH574550). All sequencing reactions were performed by Microsynth.
Identification and sequencing of putative porcine FCGR2A
Total RNA was isolated from blood cells of Göttingen minipigs and RNA integrity was determined on the Agilent 2100 Bioanalyzer System (Agilent Technologies). Then, putative FCGR2A cDNA ends were amplified in a nested PCR approach using SMARTer RACE 5′/3′ kit (Clontech). Rapid amplification of cDNA ends (RACE) PCR was performed by generation of 5′- and 3′-RACE-ready cDNA and subsequent PCR reactions using SMARTer RACE 5′/3′ kit (Clontech). More precisely, 5′- and 3′-RACE-ready cDNA was generated from total RNA serving as a template. In the first round of PCR, the supplied universal primer mix (UPM) was used together with primer JE5 or JE28, designed on predicted putative FCGR2A sequences. In a second round, nested UPM-short was used with primers JE4 or JE2 to generate 5′ or 3′ cDNA ends, respectively (Fig. 1, Online Resource 1). The products were analyzed on a 0.8% agarose gel and purified using the QIAquick gel extraction kit. Sanger sequencing was performed using several primers designed on predicted putative FCGR2A exons to identify the cDNA ends.
A final nested RT-PCR was performed on total RNA from minipig blood using first strand cDNA synthesis (New England Biolabs), the outer primers JE35/JE5, and the inner primers JE36/JE4 (Fig. 1, Online Resource 1). The product was cloned using the TOPO TA cloning kit and 30 colonies were sequenced from both sides using M13 and M13r primer. RACE PCR and RT-PCR sequences were assembled to generate the full-length transcript of the putative porcine FCGR2A.
Sequence analysis and comparison
Signal sequences were predicted by similarity to porcine FCGR2B (Qiao et al. 2006) by signalP 4.1 Server (Nielsen 2017), SMART (Letunic and Bork 2018), and Sigcleave (von Heijne 1986). SMART also predicted the extracellular structures. Transmembrane (TM) helices were predicted from similarity to human FcγRIIa (Moi et al. 2010) and by the average result from the following prediction tools: TMpred (Hofmann and Stoffel 1993), DAS (Cserzo et al. 1997), SOSUI (Hirokawa et al. 1998), PredictProtein (Yachdav et al. 2014), Phobius (Kall et al. 2004), SMART, and ALOM (a program implemented at Roche according to Klein et al. (1985)).
For the phylogenetic tree, protein sequences were first aligned with MUSCLE (Edgar 2004) then poorly aligned positions and divergent regions were filtered with GBLOCKS (Castresana 2000) so that only the conserved ECD region remained. PHYLIP software package was used to calculate a protein sequence distance matrix followed by bootstrapping with 1000 replicates (Felsenstein 2005). Data was graphically displayed with the TreeExplorer software V2.12 (Jie 2017).
Single-cell RNA sequencing
PBMCs were isolated using Ficoll-Paque Plus (GE Healthcare) and Leucosep tubes (Greiner bio-one, 12 mL) from K2 EDTA–treated whole blood of three different healthy human donors, Göttingen minipigs, or mice. Lysis buffer (BD Pharm Lyse) was used for subsequent removal of erythrocytes. Cell count and viability were determined using the Countess Automated Cell Counter (Invitrogen).
Single-cell capture was performed using the microfluidic chromium instrument (10x Genomics) capturing single cells in microdroplets. Cell suspensions containing approximately 4000 cells per sample from three different individuals were loaded together with the provided enzyme mix, beads, and oil. According to the manufacturer’s protocol, cDNA was generated, purified, and quality was checked on the Agilent 2100 Bioanalyzer System (Agilent Technologies). In a second step, a sequencing library was prepared by attaching Illumina Indices to fragmented cDNA strands. After size selection for approximately 500 bp fragments, library concentration was measured by a Qubit fluorometer (ThermoFisher). Every sample was adjusted to a final concentration of 2.5 nM, by dilution with buffer EB (Qiagen). All samples were pooled in same amounts. A PhiX solution was added, resulting in a spike-in amount of 1% in the final pool. Pooled fragments were denatured and mixed with a master mix consisting of EPX reagents 1–3 (Illumina), resulting in a final volume of 50 μL and a final concentration of 225 pM. After cluster generation, the flow cell was inserted into a HighSeq4000 instrument (Illumina). The sequencer cycle program consisted of 27 cycles for read one, 8 cycles for the index read and 99 cycles for read two.
Sequencing data were further processed using cell ranger version 2.0.0. First, fastq files were generated using the mkfastq function. Second, count files were generated using the count function. Human sequences were mapped against the genome assembly hg19, mouse sequences against the mm10, and minipig sequences against the RefSeq (reference sequence) (Pruitt et al. 2012) genome assembly Sus scrofa 11.1 containing all FCGR gene entries. Raw counts were further processed using an R (version 3.3.2) based in an in-house pipeline. First, data were imported using scater::read10XResults (version 1.6.3) function and QC parameters were calculated. The human raw cells were filtered using a minimum of 1.000 and a maximum of 50.000 umi (unique molecular identifier) counts in total. Second, cells having less than 300 genes expressed or more than 5% mitochondrial gene counts were filtered out. Mouse raw counts were filtered using a minimum of 700 and a maximum of 20.000 umi counts in total and at least 200 genes expressed. Finally, minipig raw counts were filtered using a minimum of 800 and a maximum of 20.000 umi counts in total and at least 200 genes expressed. Next, data were processed using the scater::normaliseExprs function using the 99th percentile for normalization. Confounding factors were determined based on their correlation to the first ten principle components of the normalized data. For human, we identified pct_counts_top_100_endogenous_features, log10_total_features, and donor; for mouse, we identified pct_counts_top_500_features and total_counts; and for minipig, we identified pct_counts_top_50_features, log10_total_counts, and donor as independent confounding factors. We applied a linear regression model to remove the effects of the identified confounders on the normalized data. Finally, we used the Seurat:: FindClusters function (version 188.8.131.52) and Seurat::RunTSNE function to run the t-SNE (t-distributed stochastic neighbor embedding) dimensionality reduction on selected features. Clusters were summarized according to the differential expression of various genes (Online Resource 2).
Antibodies directed against porcine FcγRIIa (AbD29332.1) and FcγRIIa/b (AbD32591.1) were generated by Bio-Rad using the HuCAL technology. Generation and specificity of the HuCAL antibodies used here will be published elsewhere. Whole blood from three different Göttingen minipigs was collected in K2 EDTA–coated vacutainer tubes (BD). Erythrocytes were removed with the lysing buffer (BD Pharm Lyse) prior to staining of dead cells with amine-reactive dye Zombie Aqua (BioLegend). Leukocytes were then incubated in separate stainings with antibodies against porcine FcγRIIa (AbD29332.1), FcγRIIa/b (AbD32591.1), FcγRIIIa (CD16-PE, clone G7, Bio-Rad), and HuCAL Fab-A-FH-negative control antibody (AbD05930). Unlabeled HuCAL antibodies were then stained with a secondary PE-conjugated goat F(ab’)2 fragment anti-human IgG, F(ab’)2 fragment specific polyclonal antibody from Jackson ImmunoResearch. Cell events were acquired on BD LSRFortessa with BD FACSDiva and analyzed using FlowJo software.
Localization of porcine FCGR3A and identification of putative FCGR2A
The low-affinity FCGR locus on chromosome 4 in the minipig genome draft based on Sus scrofa 10.2 was successfully supplemented with contigs from the Göttingen and the Wuzishan minipig and completed by PCR, cloning, and sequencing (Fig. 1, Online Resource 3). Sequences from the two minipig breeds differ in 0.31% mismatches and 1.25% indels spread over the total alignment comprising 115,000 nucleotides. The new assembly enabled the identification of exon sequences of FCGR3A in a forward orientation. Additionally, exon sequences were detected with high similarity to the porcine FCGR2B extracellular domain (ECD) and to porcine FCGR3A transmembrane/cytoplasmic (TM/C) region. These sequences belong to the putative porcine FCGR2A gene that is located in reverse orientation where the orthologue to human FCGR2A was expected (Fig. 1 and Fig. 4). Thus, the obtained sequence of the minipig low-affinity FCGR locus is completed and entirely contiguous. The newly characterized locus is highly similar to the most recent reference sequence (RefSeq) genome assembly of Sus scrofa 11.1 (Li et al. 2017).
The putative porcine FCGR2A cDNA is 1′241 bp long, contains an 822 bp open reading frame (ORF) translating to a 274 amino acids (aa) long protein (RefSeq No. XM_021089520.1). Bioinformatic analysis revealed a 45aa long signal peptide followed by an ECD region containing two immunoglobulin-like parts (Ig1, 74aa; Ig2, 78aa). Like porcine FcγRIIb, the ECD contains four potential N-glycosylation sites (Asn79, Asn89, Asn187, and Asn211) identified by the common motif (N-X-S/T) (Aebi 2013). The receptor sequence predicts a 23aa hydrophobic TM part with a negatively charged aspartic acid residue allowing interaction with the FcR γ-chain (Kim et al. 2003). In the 27aa long intracellular part, no immunoreceptor tyrosine-based activation motif (ITAM; Y-X-X-L/I) or immunoreceptor tyrosine-based inhibition motif (ITIM; S/I/V/L-X-Y-X-X-I/V/L) was found in contrast to human FcγRIIa or FcγRIIb, respectively (Isakov 1997; Ravetch and Lanier 2000) (Fig. 2).
The putative porcine FcγRIIa.1 variant revealed a 24aa deletion within the Ig2-like part of the ECD (Gly192_Gln215del) (Fig. 2). Further variants include FcγRIIa.2 lacking the whole Ig2-like part of the ECD (Asp131_Gln215del) and FcγRIIa.3 lacking the whole Ig1-like part of the ECD (Ala45_Ser130del) and bearing the 24aa deletion of FcγRIIa.1. Furthermore, four single nucleotide polymorphisms were detected, two of them affecting the coding sequence and thus representing potential polymorphisms. The A11S polymorphism is located in the signal sequence and the H205Y polymorphism in the Ig2-like part of the ECD.
A complete picture of the genomic organization of the porcine FCGR locus
The new RefSeq assembly contains genes and curated transcripts of FCGR1A (gene ID, 613130; transcript ID, NM_001033011.1.1), FCGR2B (gene ID, 613131; transcript ID, NM_001033013.2.1), and recently also FCGR3A (gene ID, 397684; transcript ID, NM_214391.1.1). The predicted transcript (transcript ID: XM_021089520.1) from the RefSeq gene LOC110260307 (gene ID, 110260307) codes for the 11A 205H polymorphism of putative porcine FcγRIIa. In contrast, the transcript identified from sequences of the Göttingen and the Wuzishan minipig (Fig. 1, Online Resources 3) codes for the 11S 205Y polymorphism of putative porcine FcγRIIa. However, both polymorphic variants were detected by sequencing of one Göttingen minipig.
Cellular distribution of FcγRs
The activating low-affinity FcγRIIIa is most strongly expressed among the FcγRs in all the species studied here. Minipig PBMCs revealed a strong and relatively homogeneous FcγRIIIa expression on all monocytes, DCs, NK cells, and cytotoxic T lymphocytes. Interestingly, T cells and B cells showed heterogeneous expression suggesting either different cell subsets or activation states. Human monocytes are often separated in classical, intermediate, and non-classical monocytes according to the CD14 and CD16 (FcγRIIIa) expression (Ziegler-Heitbrock 2015). As expected, the larger CD14high classical monocyte subset did not express FcγRIIIa, whereas the minor non-classical CD14low subset was strongly positive for FcγRIIIa. Also, in mice, it is the cluster containing the monocytes that shows expression for FcγRIIIa, while the other immune cell types, in contrast to the other species, show no expression. The inhibitory low-affinity FcγRIIb was found to be expressed mainly on monocytes, B cells, and DCs of the minipig. Human monocytes were not found to express FcγRIIb, while mouse FcγRII was weaker expressed in the monocyte and DC cluster as compared to the minipig. Expression of FcγRIIb in human and mouse PBMCs was mainly found in B cells. FcγRIIa, the activating low-affinity receptor we identified with our mapping strategy, is expressed at lowest levels in minipigs and humans. In the minipig, FcγRIIa mRNA was only detected in very few cells of the monocyte cluster. More monocytes were positive in the human and expression levels are slightly higher. Mouse FcγRIII, the orthologue of FcγRIIa, is expressed on most cells of the monocyte/DC cluster at highest levels compared to the other species. Similar expression levels and patterns were observed for FcγRIa. In the minipig, the expression is at low levels and restricted to monocytes. In humans, CD14high CD16- classical monocytes express FcγRIa, in contrast to CD14low CD16+ non-classical monocytes. Mice show a similar FcγRI expression pattern on a subset of the monocyte/DC cluster.
The three different classes of FcγRs form a finely tuned system required for efficient immune reactions in mammals. Minipigs represent a valuable alternative to NHP in preclinical studies. Thus, it is of particular importance to know all FcγR components in a preclinical animal model intended for testing of therapeutic antibodies. The characterization of the low-affinity FcγR proteins and genes in minipigs should provide a basis for preclinical studies with therapeutic antibodies.
While the inhibitory receptor is widely described as FcγRIIb (in the mouse known as FcγRII), the nomenclature of the low-affinity-activating FcγRs has evolved in a far more divergent manner. The low-affinity FcγRIIa is well-known in humans and has been described in the NHP, cattle, and other mammals, such as rabbits and sheep (Akula et al. 2014). The orthologue in the mouse, however, was named FcγRIII at its discovery (Nimmerjahn and Ravetch 2006). This receptor was initially not known in pigs due to an incomplete genome characterization and therefore was not described by Akula et al. (2014). In the present study, we were able to identify the putative porcine FcγRIIa located on chromosome 4 of the Göttingen minipig. The orthologue to human FcγRIIIa is known in NHPs, cattle, and other mammals, including the mouse, where it was designated as FcγRIV (Nimmerjahn et al. 2005). The orthologous FcγRIIIa cDNA and protein were also described in the pig but the corresponding gene and its genomic localization was unknown (Akula et al. 2014; Halloran et al. 1994). Here, we describe the localization of the gene FCGR3A encoding the minipig FcγRIIIa between FCGR2B and the putative FCGR2A on chromosome 4 of the Göttingen minipig on the forward strand. The identification of the putative FCGR2A and the localization of FCGR3A in pigs allow the comparison of the low-affinity FCGR locus to other species. We found that this locus of the minipig is organized similarly as in NHP, cattle, rat, and mouse with the position of the putative porcine FCGR2A gene coinciding with the other species. Nevertheless, significant differences to the human FCGR locus were observed. Thus, the complete characterization of the low-affinity FCGR locus of the minipig presented here confirms the absence of genes coding for homologs of the human FcγRIIIb and FcγRIIc, as is the case for all other animal species studied so far.
Sequence similarity displayed in the phylogenetic tree in Fig. 5 shows that FcγRIIa and FcγRIIb of the same species usually cluster together, probably originating from a duplication event early in speciation (Akula et al. 2014). The high similarity of the ECD region of porcine FcγRIIb to the newly identified porcine FcγRIIa fits in the pattern observed with the corresponding receptors from other species. Therefore, we suggest naming the transcript FcγRIIa. However, as detailed in Fig. 3, exons coding for the signal region and the TM/C region of the FcγRs appear to be shuffled during gene duplications and rearrangements leading to a mosaic structure that is characteristic for primates, rodents, and artiodactyls, respectively. Predictions suggest intracellular signaling by porcine FcγRIIa via interaction with the FcR γ-chain as it is described for cattle FcγRIIa and mouse FcγRIII (Lux and Nimmerjahn 2013). This similarity strengthens the hypothesis of the orthology among these receptors. On the other hand, FcγRIIa in primates is known to signal via integrated intracellular ITAM. It should be considered that differences in ITAMs potentially lead to functional differences between Fc receptors (Herik et al. 1995).
Two potential polymorphisms, A11S and H205Y, were identified in the main FcγRIIa transcript. The first located in the signal region and the latter was identified in the Ig2-like part of the ECD involved in the interaction with IgG antibodies (Fig. 3). Due to its location, the H205Y polymorphism could potentially influence binding affinities to certain IgG subclasses. Apart from that, we found three potential isoforms of porcine FcγRIIa with unknown functions and significance, probably generated by alternative splicing. Similar splice variants were already described for porcine FcγRIIb (Xia et al. 2012; Xia et al. 2011) and FcγRIIIa (Jie et al. 2009). In particular, humans were shown to have splice variants and polymorphisms with significant functional consequences. Altered binding affinities are associated with the outcome of therapeutic antibody treatments and with disease progression (Bournazos et al. 2010; Ziakas et al. 2016). Studies with more minipigs are required in order to assess the potential incidence of polymorphisms, splice variants, and sub-isoforms. Additionally, their biological relevance remains to be assessed.
Biological responses triggered by FcγRs do not only depend on the affinity of IgG interaction but also on their cellular distribution (Albanesi and Daeron 2012). Knowing the expression of FcγRs on immune cells facilitates the estimation of effects triggered by IgG interaction. We performed single-cell RNA sequencing on minipig, human, and mouse PBMCs to study the FcγR expression profile on various cell types.
In the Göttingen minipig, FcγRIa transcripts were only identified in monocytes at similar levels as observed in human and mouse. Like in humans, no FcγRIa expression was detected in minipig blood DCs although FcγRIa expression was often reported in human DCs (Nimmerjahn et al. 2015; Tamoutounour et al. 2012). FcγRIa expression, however, was usually analyzed in tissue resident or induced DCs and not found in blood DCs (Langlet et al. 2012). Devriendt et al. (2013) showed that the FcγRIa expression profile on porcine DCs depends on the activation stimulus, and similar findings were observed for human DCs. Therefore, FcγRIa expression can neither be excluded from minipig blood DCs nor from tissue-resident subsets. Varying expression levels of FcγRIa between minipig and human DCs could, however, result in varying capacity for antigen presentation by immune complexes and cytokine production (Cohen-Solal et al. 2004; van der Poel et al. 2011).
Only a few monocytes of the minipig showed weak staining for FcγRIIa. Generally, the FcγRIIa expression in PBMCs seems to be lower in the minipig as compared to humans and mice. This low expression was also observed in porcine gene expression data from NCBI (Li et al. 2017). Low expression of FcγRIIa in monocytes could theoretically be upregulated upon inflammatory stimuli similar to other activating Fc receptors (Nimmerjahn et al. 2005; Pricop et al. 2001). Like humans, minipigs express FcγRIIa on platelets as detected by flow cytometry (Rosenfeld et al. 1985). Platelets are mediators of immune responses upon binding of IgG immune complexes via FcγRIIa. This interaction can lead to platelet activation, phagocytosis, and ultimately to thrombus formation with pathological consequences (Worth et al. 2006; Zhi et al. 2015). The minipig might thus be a good model to study platelet-mediated functions and side effects of therapeutic antibodies, such as bevacizumab-induced retinal vein thrombosis, in contrast to mice that do not express FcγRIIa on platelets (Meyer et al. 2009). Gene expression data from NCBI Gene show that FcγRIIa is mainly expressed in the liver and the lung of pigs. Generally, the porcine FcγR expression is mainly detected in the liver, lung, and spleen tissue. This expression profile suggests that FcγRIIa mediates important immune functions in tissue-resident cells other than platelets in the blood.
Single-cell RNA sequencing of minipig PBMCs shows FcγRIIb expression on B cells, DCs, and monocytes. FcγRIIb expression on monocytes correlated with flow cytometry data using FcγRIIa specific and FcγRIIa/b cross-reactive HuCAL antibodies. Presently, the exact cellular distribution of FcγRIIb cannot be evaluated due to the lack of specific antibodies. A previous study postulates cross-reactivity of anti-human CD32 antibody (AT10) without showing data (Balmelli et al. 2005), a finding that could not be confirmed in our hands (not shown). The expression of FcγRIIb on minipig B cells and DCs reflects the situation in humans. On the other hand, minipig and mouse blood monocytes were found to express FcγRIIb as well, whereas human blood monocytes do not (Nimmerjahn et al. 2015). Low levels of FcγRIIa together with high levels of FcγRIIb on minipig monocytes could result in enhanced inhibitory signaling compared to humans. Hence, this could lead to an underestimation of effects or toxicity observed in minipig studies with therapeutic antibodies with FcγR-mediated effector functions.
Porcine FcγRIIIa was so far the best studied Fc receptor due to its high expression and the availability of specific antibodies. Its expression pattern was closely reflected in our single-cell RNA sequencing and flow cytometry analysis (Piriou-Guzylack and Salmon 2008). Minipig and human FcγRIIIa was found to be the highest expressed FcγR in PBMCs. In both species, T cells and B cells were found to express FcγRIIIa mRNA. Whereas FcγRIIIa expression on human T cells is controversially discussed in the literature (Nimmerjahn and Ravetch 2008), it can be excluded on B cells. Therefore, the FcγRIIIa expression in T cells and B cells of both species is considered as unspecific or represents different subsets or activation states. The difference between minipig and human is that FcγRIIIa is only expressed on monocyte subpopulations in humans, whereas it is expressed in all monocytes in the pig (Rubic-Schneider et al. 2016). The ubiquitous expression of activating FcγRIIIa on minipig monocytes could possibly counteract the inhibitory effects of FcγRIIb and the low levels of FcγRIIa. In therapeutic antibody research, a careful evaluation of the interaction to the various FcγRs would be needed to estimate the activation or inhibition potential of the antibody on minipig monocytes. Altogether, the human expression pattern of these FcγRs is more concordant with porcine than with murine monocytes (Fairbairn et al. 2013). The expression pattern of FcγRs is known to vary not only between species but also between individuals. As mentioned before, it can also be influenced by different stimuli, the immune status, or upon treatment. Therefore, further studies with more minipigs under different conditions are required to make a precise statement about the FcγR distribution in health and disease.
Our work allowed the localization of FcγRIIIa and the identification of the hitherto undescribed FcγRIIa on chromosome 4 of the Göttingen minipig. The newly identified FcγRIIa described here is considered as an orthologue to human, NHP, and cattle FcγRIIa as well as to mouse FcγRIII due to the highly conserved extracellular structures. The identification of FcγRIIa completes the picture of FcγRs in the pig and provides the genetic foundation for further studies. Our expression studies are the first to describe the expression of FcγRIa in monocytes and FcγRIIa on platelets of the Göttingen minipig. Additionally, FcγRIIb was found in monocytes, DCs, and B cells. The higher expression of FcγRIIIa and FcγRIIb and the lacking expression of FcγRIIa on monocytes are different to humans. Therefore, effects on monocytes should be carefully evaluated before using the minipig in preclinical studies with therapeutic antibodies. Nevertheless, FcγRIIa expression on platelets makes the minipig a valuable model to study platelet-mediated effects of therapeutic antibodies which are hard to evaluate in mice.
The authors acknowledge Roland Jenni and David Waiz for minipig handling and blood sampling, Tobias Heckel for RNA preparation, and Laetitia Petersen for assistance with flow cytometry. This work was made possible by the managerial support of Thomas Singer and Olivia Spleiss.
- Felsenstein J (2005) PHYLIP (phylogeny inference package) version 3.6Google Scholar
- Halloran PJ, Sweeney SE, Strohmeier CM, Kim YB (1994) Molecular-cloning and identification of the porcine cytolytic trigger molecule G7 as a fc-gamma-Riii-alpha (Cd16) homolog. J Immunol 153:2631–2641Google Scholar
- Heckel T, Schmucki R, Berrera M, Ringshandl S, Badi L, Steiner G, Ravon M, Kung E, Kuhn B, Kratochwil NA, Schmitt G, Kiialainen A, Nowaczyk C, Daff H, Khan AP, Lekolool I, Pelle R, Okoth E, Bishop R, Daubenberger C, Ebeling M, Certa U (2015) Functional analysis and transcriptional output of the Gottingen minipig genome. BMC Genomics 16:932CrossRefGoogle Scholar
- Herik V, Oudijk IE, TBM W, Tempelman MJ, Capel PJ, Van de Winkel JG (1995) Functional differences between two Fc receptor ITAM signaling motifs. BloodGoogle Scholar
- Hofmann K, Stoffel W (1993) TMbase - a database of membrane spanning proteins segments. Biol Chem Hoppe Seyler 3874Google Scholar
- Jie T (2017) TreeExplorer:2.12 http://en.bio-soft.net/
- Li M, Chen L, Tian S, Lin Y, Tang Q, Zhou X, Li D, Yeung CKL, Che T, Jin L, Fu Y, Ma J, Wang X, Jiang A, Lan J, Pan Q, Liu Y, Luo Z, Guo Z, Liu H, Zhu L, Shuai S, Tang G, Zhao J, Jiang Y, Bai L, Zhang S, Mai M, Li C, Wang D, Gu Y, Wang G, Lu H, Li Y, Zhu H, Li Z, Li M, Gladyshev VN, Jiang Z, Zhao S, Wang J, Li R, Li X (2017) Comprehensive variation discovery and recovery of missing sequence in the pig genome using multiple de novo assemblies. Genome Res 27:865–874CrossRefGoogle Scholar
- Nimmerjahn F, Ravetch JV (2008) Fcgamma receptors as regulators of immune responses. Nat Rev, ImmunolGoogle Scholar
- Tamoutounour S, Henri S, Lelouard H, de Bovis B, de Haar C, van der Woude CJ, Woltman AM, Reyal Y, Bonnet D, Sichien D, Bain CC, Mowat AM, Reis e Sousa C, Poulin LF, Malissen B, Guilliams M (2012) CD64 distinguishes macrophages from dendritic cells in the gut and reveals the Th1-inducing role of mesenteric lymph node macrophages during colitis. Eur J Immunol 42:3150–3166CrossRefGoogle Scholar
- Villani AC, Satija R, Reynolds G, Sarkizova S, Shekhar K, Fletcher J, Griesbeck M, Butler A, Zheng S, Lazo S, Jardine L, Dixon D, Stephenson E, Nilsson E, Grundberg I, McDonald D, Filby A, Li W, De Jager PL, Rozenblatt-Rosen O, Lane AA, Haniffa M, Regev A, Hacohen N (2017) Single-cell RNA-seq reveals new types of human blood dendritic cells, monocytes, and progenitors. Science 356:eaah4573CrossRefGoogle Scholar
- Warmerdam PAM, Nabben NMJM, Vandegraaf SAR, Vandewinkel JGJ, Capel PJA (1993) The human low affinity immunoglobulin-G Fc receptor-Iic gene is a result of an unequal crossover event. J Biol Chem 268:7346–7349Google Scholar
- Yachdav G, Kloppmann E, Kajan L, Hecht M, Goldberg T, Hamp T, Honigschmid P, Schafferhans A, Roos M, Bernhofer M, Richter L, Ashkenazy H, Punta M, Schlessinger A, Bromberg Y, Schneider R, Vriend G, Sander C, Ben-Tal N, Rost B (2014) PredictProtein--an open resource for online prediction of protein structural and functional features. Nucleic Acids Res 42:W337–W343CrossRefGoogle Scholar
- Ziakas PD, Ls P, Zintzaras E (2016) FcgammaRIIa-H131R variant is associated with inferior response in diffuse large B cell lymphoma: A meta-analysis of genetic risk. J, BuonGoogle Scholar
OpenAccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.