Molecular Immunology 58 (2014) 98–107
Contents lists available at ScienceDirect
Molecular Immunology journal homepage: www.elsevier.com/locate/molimm
Human complement C3 deficiency: Th1 induction requires T cell-derived complement C3a and CD46 activation Arije Ghannam a , Jean-Luc Fauquert b , Caroline Thomas c , Claudia Kemper d , Christian Drouet a,∗ a
Université Joseph Fourier, GREPI/AGIM CNRS FRE3405, Grenoble, France Pédiatrie, CHU de Clermont-Ferrand, Clermont-Ferrand, France c Oncologie pédiatrique, CHU de Nantes, Nantes, France d Division of Transplantation Immunology and Mucosal Biology (DTIMB), MRC Centre for Transplantation, King’s College London, Guy’s Hospital, London, UK b
a r t i c l e
i n f o
Article history: Received 12 November 2013 Accepted 13 November 2013 Available online 8 December 2013 Keywords: Complement T cell effector response C3 deficiency C3a
a b s t r a c t Human T helper type 1 (Th1) responses are essential in defense. Although T cell receptor (TCR) and co-stimulator engagement are indispensable for T cell activation, stimulation of additional receptor pathways are also necessary for effector induction. For example, engagement of the complement regulator CD46 by its ligand C3b generated upon TCR activation is required for IFN-␥ production as CD46-deficient patients lack Th1 responses. Utilizing T cells from two C3-deficient patients we demonstrate here that normal Th1 responses also depend on signals mediated by the anaphylatoxin C3a receptor (C3aR). Importantly, and like in CD46-deficient patients, whilst Th1 induction are impaired in C3-deficient patients in vitro, their Th2 responses are unaffected. Furthermore, C3-deficient CD4+ T cells present with reduced expression of CD25 and CD122, further substantiating the growing notion that complement fragments regulate interleukin-2 receptor (IL-2R) assembly and that disturbance of complement-guided IL-2R assembly contributes to aberrant Th1 effector responses. Lastly, sustained intrinsic production of complement fragments may participate in the Th1 contraction phase as both C3a and CD46 engagement regulate IL-10 co-expression in Th1 cells. These data suggest that C3aR and CD46 activation via intrinsic generation of their respective ligands is an integral part of human Th1 (but not Th2) immunity. © 2013 Elsevier Ltd. All rights reserved.
1. Introduction Complement has initially been defined as key part of innate immunity and as the first line of defense in the fight against invading pathogens. The complement system is a multi-protein network and consists of fluid-phase and membrane-bound proteins that are activated in a cascade-like fashion upon presence of pathogens or other danger-associated signals (Volanakis, 1998). Complement activation leads to the formation of the C3 convertase, which cleaves C3 into the key effector molecules, C3a (anaphylatoxin) and C3b (opsonin) which then drive microbe removal (Volanakis, 1998). However, complement is now also firmly established as instructor of adaptive immunity and is required for germinal centre formation, antibody responses and normal B cell memory development (Pepys, 1974; Molina et al., 1996; Fearon and Locksley,
∗ Corresponding author at: Université Joseph Fourier, GREPI/AGIM CNRS FRE3405, CHU Grenoble, CS10217, F-38043 Grenoble, France. Tel.: +33 476767201; fax: +33 476766251. E-mail addresses: [email protected], [email protected] (C. Drouet). 0161-5890/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.molimm.2013.11.010
1996; Carroll, 2004). Early evidence for the participation of complement in adaptive host responses came from studies using mice deficient for the gene encoding the key complement component 3 (C3). C3-deficient animals fail to produce antibody (Pepys, 1974; Molina et al., 1996) and to mount appropriate CD4+ and CD8+ T cell responses in infection models (Kopf et al., 2002; Suresh et al., 2003). Similarly, patients with C3 gene mutations and a defect in C3 protein expression suffer from severe pyogenic infections in early childhood and some display low antibody titres (and lack or have reduced vaccination responses) (Ghannam et al., 2008). The insufficient lymphocyte activation observed in C3-deficient animals and patients has been attributed to defective dendritic cell (DC) maturation leading to suboptimal antigen presentation and costimulation because C3a receptor (C3aR)-mediated signals induce major histocompatibility complex (MHC) class II and CD86 upregulation on DCs (Ghannam et al., 2008; Peng et al., 2006; Li et al., 2008; Reis et al., 2008). Recent data indicate that direct signalling of complement receptors on the T cells themselves is required for CD4+ T cell effector lineage induction and regulation. Activation of cultured human CD4+ T cells through T cell receptor (TCR) stimulation leads to C3a and C3b generation (Cardone et al., 2010). Subsequent engagement
A. Ghannam et al. / Molecular Immunology 58 (2014) 98–107
of the complement regulator CD46 on the CD4+ T cell surface via cell-derived C3b (CD46 ligand) in the presence of endogenous IL-2 is an absolute requirement for IFN-␥ production and Th1 induction: Purified T cells from CD46-deficient patients are unable to assume a Th1 phenotype in vitro and in vivo (but have normal Th2 responses) and these patients suffer consequently from recurrent viral infections (Le Friec et al., 2012). Interestingly, rodents do not express CD46 on somatic tissue (Tsujimura et al., 1998) and a functional complement protein homologue has not yet been identified, suggesting that there are differences in these induction pathways between mice and humans (Cope et al., 2011). Nonetheless, complement receptor-mediated signalling events also participate in CD4+ T cell effector function in mice: Experiments using T cells isolated from C3aR (C3AR1) and C5aR (C5AR1) double-knock-out mice demonstrated the need for engagement of these T cell anaphylatoxin receptors for cell proliferation and Th1 cytokine production (Liu et al., 2005; Strainic et al., 2008). Furthermore, these studies are in line with observations made using human CD46-activated T cells (Cardone et al., 2010; Le Friec et al., 2012) and support a current paradigm shift in the complement field: they demonstrate that the required complement fragments C3a and C5a are not serum-derived but generated by the APC or T cell or both and that such immune cell-derived local production is fully sufficient for Th1 induction (Heeger and Kemper, 2012). The potential role of anaphylatoxin receptors expressed by human CD4+ T cells in the induction of effector functions is unexplored and individuals with deficiencies in either C3aR or C5aR expression have not been described. However, we have been able to obtain blood samples from two rare patients with C3 gene mutations. Importantly, although 30 patients from 22 families and 15 mutations have been described world-wide (OMIM ID: 120700); in reality, obtaining blood samples from C3-deficient individuals is difficult. We describe here the immunological phenotype of CD4+ T cells in regards to Th1 and Th2 induction from Patient C3-1 (who had previously been described (Ghannam et al., 2008)), and from Patient C3-2 with a novel molecular defect in C3 protein generation. Using purified CD4+ T cells from these two patients (these T cells cannot generate C3a and C3b locally), we demonstrate that normal Th1 induction requires the two T cell-generated C3 activation fragments C3a and C3b. Importantly, Th2 induction is independent of local complement activation. Additionally, both C3a and C3b may also participate in the contraction phase of human Th1 responses, which is characterized by co-secretion of IL-10 by IFN-␥-positive Th1 cells (Cardone et al., 2010), further underpinning the developing role of complement not only in the initiation of inflammatory responses but also in their negative control (Kemper and Atkinson, 2007).
2. Materials and methods 2.1. Patients and donors Blood was collected from patients and healthy donors after informed consent was obtained. Since early childhood, the C3deficient Patients C3-1 and C3-2 suffered from recurrent pyogenic infections, were appropriately vaccinated and submitted to antibiotic prophylaxis (Ghannam et al., 2008; Dejoie et al., 2009). At 10 years of age, Patient C3-1 developed anti-nuclear antibodies (Sm ENA specificity). He was then subjected to an i.v. immunoblobulin regimen and concomitantly recovered from systemic infections. The blood samples were collected in between infectious disease episodes in the cases of both patients. Each experimental condition was performed in triplicate and data shown are the mean value of three data points generated in one experiment. Control conditions using healthy donors have been performed with an n = 4 or n = 5.
99
2.2. Molecular analysis of C3-deficient patients The mutation in the C3 gene of Patient C3-1 has previously been published (Ghannam et al., 2008). To define the C3 mutation of the here newly described Patient C3-2, total RNA was prepared using TRIzol® (Invitrogen) from cultured monocytes after 24 h activation with 0.1 mg/ml E. coli LPS. C3-specific overlapping sequences were amplified using the Reverse Transcriptor® system (Roche Molecular Biochemicals) with primers chosen using the cDNA reference sequence (Gen Bank accession NM 000064) from the generated cDNA fragments. Amplicons were directly sequenced using the ABI PRISM Big Dye Terminator Cycle Sequencing v3.0 Reaction Kit, and were analyzed on an ABI 3100 DNA Analyzer (Applera). The two mutated alleles were identified as carrying c.2002delC, p.Arg668AlafsX35 (exon 16), also found in the mother (not shown) and c.2290C > T, p.Arg764Stop (exon 18) changes.
2.3. Antibodies and proteins Cell-stimulating mAbs were bought from BD Pharmingen (antihuman CD28, CD28.2), purified from a specific hybridoma line (anti-CD3; OKT-3) or generated in house (anti-CD46; TRA-2-10 (Wang et al., 2000)). Cell surface marker and protein expressions were assessed using fluorochrome-labelled antibodies to CD4 (555349), CD8 (555635), CD25 (555431), CD45RA (555488), CD45RO (559865), and CD127 (557938) obtained from BD Biosciences. The antibody to CD122 (FAB224A) was obtained from R&D Systems and the anti-human C3a receptor (561178) antibody was bought from BD Pharmingen. FACS analyses were performed on a FACSCalibur® instrument (BD Biosciences) using the CellQuest® software (BD Biosciences). Recombinant human IL-2 was obtained from Genzyme Therapeutics Ltd. The complement activation fragments C3a (A118), C3b (A114), C5a (A144) and iC3b (A115) were serum-purified proteins pooled from different donors and purchased from Complement Technologies Inc.
2.4. T cell purification and activation PBMCs were isolated from heparinized blood by FicollHypaque® density gradient centrifugation and CD4+ T cells then positively isolated using magnetic beads (CD4 Isolation Kit II® , Dynabeads® (Dynal Biotech) or the CD4+ Magnetic Bead Purification Kit® (Miltenyi Biotec)), according to the manufacturer’s protocol. Purity of isolated cell populations were consistently >95%. Because of limiting cell numbers obtained from the patient samples, purified CD4+ T cells were cultured in 96-well plates at 1 × 105 cells/wells in wells precoated with activating antibodies in the presence or absence of IL-2 or purified complement proteins. A small number of T cell activation experiments were performed in 48-well plates (2–2.5 × 105 cells/well; Figs. 1C and 2D and these condition usually led to better T cell activation with higher cytokine secretion/cell when compared to cultures activated in 96-well plates. Complement activation fragments were added at the following concentrations: C3a (80 nM), C5a (80 nM), C3b (80 g/ml), iC3b (80 g/ml). 2.5. Cytokine, C3b and C3a measurements IFN-␥, IL-2, IL-4 and IL-10 production by cells was assessed using the CBA® Human Th1/Th2 Cytokine Kit II® (BD Biosciences) or the appropriate Cytokine Secretion Assays® (Miltenyi BioTec), according to the manufacturer’s protocol. The presence of C3b or C3a in serum or cell culture media was measured utilizing the ‘Complement fragment 3b, C3b ELISA Kit’ from MyBiosource (San
100
A. Ghannam et al. / Molecular Immunology 58 (2014) 98–107
Fig. 1. Mutations in the C3-deficient patients, their local complement production and composition of their CD4+ , CD8+ , and natural Treg pool. (A) Schematic diagram of the C3 protein domains. The codon changes in the protein based on mutations in the C3 gene of Patient C3-1 and Patient C3-2 are indicated by arrows. (B) Chromatograms of exons 16 and 18 from DNAs of Patient C3-2 and an healthy control. The chromatogram of Patient C3-2 exon 16 was also seen in his mother (GenBank ID: KC493376, KC493377). (C) Measurement of serum and PBMC-derived C3a and C3b proteins. C3b and C3a content in either serum samples from 3 healthy donors and from Patient C3-2 or in cell culture supernatants from CD3 + CD46-activated T cells (12 h) was measured by ELISA or by using the Anaphylatoxin Bead Array, respectively. (D) Assessment of CD4+ , CD8+ and Treg cell numbers in Patients C3-1 and C3-2 in comparison to healthy donors. Stainings for patient cell populations were performed in triplicate with one representative FACS plot of each result shown. Cell populations derived from healthy donors were assessed from five distinct individuals (n = 5) in duplicate with one FACS plot showing the average cell number expression from one selected donor. For the assessment of CD4+ and CD8+ T cells, the gate was set on the lymphocyte portion of PBMCs (upper three panels). For the measurement of Treg cell numbers, the FACS gate was set on the CD4+ portion of the lymphocyte population (lower three panels). Treg cells were identified as CD4+ CD25+ CD127low/− . HD, healthy donor.
Diego, USA) or the ‘Human Anaphylatoxin Bead Array’ from BD Biosciences, respectively. 2.6. RNA silencing siRNA targeting the human C3aR (siRNA sc42840) and the negative control siRNAs (sc37007 and sc36869; fluorescein-labelled) were purchased from Santa Cruz Biotechnology, Inc. siRNA was delivered into primary human CD4+ T cells by electroporation
(2 × 106 cells ml/transfection buffer (Ambion); 3 g/ml siRNA; 200 V and 325 mF using the Bio-Rad Gene Pulser (Bio-Rad Laboratories). Transfection efficiency was consistently above 75% and protein knock-down peaked at 20–36 h post transfection. 2.7. Statistical analysis Statistical analyses were performed using the Student’s twotailed t test and Bonferroni correction for multiple comparisons
A. Ghannam et al. / Molecular Immunology 58 (2014) 98–107
101
Fig. 2. Human Th1 induction in CD4+ T cells requires complement activation fragments C3a and C3b. (A) IFN-␥, IL-4 and IL-10 secretion by T cells isolated from the C3deficient patients and a healthy donor (representative values among four individual donors tested) upon indicated activation conditions (36 h post activation). (B) CD4+ T cells upregulate C3aR expression upon TCR engagement as assessed by flow cytometry (6 h post activation). (C) Inhibition of C3aR signalling reduces Th1 induction. Purified T cells from healthy donors were activated under the depicted conditions in the presence or absence of an anti-C3aR antagonizing antibody and IFN-␥, IL-10 and IL-4 production measured 36 h post activation (n = 3). Error bars indicate means ± SEM; *p < 0.05; **p < 0.01 when compared to isotype control-treated cells. (D) Knock-down of C3aR protein expression via siRNA reduces Th1 induction. Purified T cells were transfected with a control siRNA or a C3aR-specific siRNA and C3aR protein expression assessed via Western Blotting 20 h post transfection (upper panel). Cells were then stimulated in 48-well plates as shown and IFN-␥ production measured 24 h post activation. (E) Addition of C3a (and C3b) significantly increases Th1 induction in normal CD4+ T cells. T cells from healthy donors were either left non-activated (NA), CD3-, CD3+CD28- or CD3+CD46activated (top panel to bottom panels) with or without addition of C3b and/or C3a, and cytokine production measured 36 h post activation (n = 4). Shown are results ± SEM; *p < 0.05; **p < 0.01 when cytokine secretion is compared to similarly activated cells but without addition of C3a or C3b. Note that IL-4 production (Th2 induction) is not affected by C3a and C3b, in line with observations when using T cells from both C3-deficient patients.
102
A. Ghannam et al. / Molecular Immunology 58 (2014) 98–107
(Excel® software, Windows® package). When possible, data were represented as mean ± SEM. Values were considered significant at p < 0.05. 2.8. Ethics statement The Institutional Review Board of Grenoble University Hospital (South-East committee V) approved these studies. Investigations were performed in accordance with the principles of the Helsinki declaration and with the French ethical policies employed when using biological sample collection and processing (Ministry of Health authorization). 3. Results 3.1. The molecular defect in C3-deficient patients C3-1 and C3-2 Patient C3-1 was previously described (Ghannam et al., 2008) and has a Ser550Pro mutation (Fig. 1A). This mutations leads to the generation of an immature, non-secreted C3 protein and the patient presents with compromised functions of several vital immune cells including DCs, and B and T cells (Ghannam et al., 2008). Patient C3-2 carries the nonsense mutation c.2290C > T (p.Arg764Stop, resulting in a shorter protein; paternal transcript) in exon 18 and the single nucleotide deletion c.2002Cdel (resulting in a frameshift at Arg668AlafsX35, maternal transcript) in exon 16 and is described here for the first time. Fig. 1B shows the pertinent portion of the C3 gene sequences harbouring the mutations of patient C3-2, compared to the control sequence. These abnormalities have not been detected in 100 alleles from healthy controls and they should hypothetically lead to the expression of non-secreted, abrogated/shorter and unstable proteins. Similar to Patient C3-1 (Ghannam et al., 2008), Patient C3-2 has an undetectable serum C3 level (
Contents lists available at ScienceDirect
Molecular Immunology journal homepage: www.elsevier.com/locate/molimm
Human complement C3 deficiency: Th1 induction requires T cell-derived complement C3a and CD46 activation Arije Ghannam a , Jean-Luc Fauquert b , Caroline Thomas c , Claudia Kemper d , Christian Drouet a,∗ a
Université Joseph Fourier, GREPI/AGIM CNRS FRE3405, Grenoble, France Pédiatrie, CHU de Clermont-Ferrand, Clermont-Ferrand, France c Oncologie pédiatrique, CHU de Nantes, Nantes, France d Division of Transplantation Immunology and Mucosal Biology (DTIMB), MRC Centre for Transplantation, King’s College London, Guy’s Hospital, London, UK b
a r t i c l e
i n f o
Article history: Received 12 November 2013 Accepted 13 November 2013 Available online 8 December 2013 Keywords: Complement T cell effector response C3 deficiency C3a
a b s t r a c t Human T helper type 1 (Th1) responses are essential in defense. Although T cell receptor (TCR) and co-stimulator engagement are indispensable for T cell activation, stimulation of additional receptor pathways are also necessary for effector induction. For example, engagement of the complement regulator CD46 by its ligand C3b generated upon TCR activation is required for IFN-␥ production as CD46-deficient patients lack Th1 responses. Utilizing T cells from two C3-deficient patients we demonstrate here that normal Th1 responses also depend on signals mediated by the anaphylatoxin C3a receptor (C3aR). Importantly, and like in CD46-deficient patients, whilst Th1 induction are impaired in C3-deficient patients in vitro, their Th2 responses are unaffected. Furthermore, C3-deficient CD4+ T cells present with reduced expression of CD25 and CD122, further substantiating the growing notion that complement fragments regulate interleukin-2 receptor (IL-2R) assembly and that disturbance of complement-guided IL-2R assembly contributes to aberrant Th1 effector responses. Lastly, sustained intrinsic production of complement fragments may participate in the Th1 contraction phase as both C3a and CD46 engagement regulate IL-10 co-expression in Th1 cells. These data suggest that C3aR and CD46 activation via intrinsic generation of their respective ligands is an integral part of human Th1 (but not Th2) immunity. © 2013 Elsevier Ltd. All rights reserved.
1. Introduction Complement has initially been defined as key part of innate immunity and as the first line of defense in the fight against invading pathogens. The complement system is a multi-protein network and consists of fluid-phase and membrane-bound proteins that are activated in a cascade-like fashion upon presence of pathogens or other danger-associated signals (Volanakis, 1998). Complement activation leads to the formation of the C3 convertase, which cleaves C3 into the key effector molecules, C3a (anaphylatoxin) and C3b (opsonin) which then drive microbe removal (Volanakis, 1998). However, complement is now also firmly established as instructor of adaptive immunity and is required for germinal centre formation, antibody responses and normal B cell memory development (Pepys, 1974; Molina et al., 1996; Fearon and Locksley,
∗ Corresponding author at: Université Joseph Fourier, GREPI/AGIM CNRS FRE3405, CHU Grenoble, CS10217, F-38043 Grenoble, France. Tel.: +33 476767201; fax: +33 476766251. E-mail addresses: [email protected], [email protected] (C. Drouet). 0161-5890/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.molimm.2013.11.010
1996; Carroll, 2004). Early evidence for the participation of complement in adaptive host responses came from studies using mice deficient for the gene encoding the key complement component 3 (C3). C3-deficient animals fail to produce antibody (Pepys, 1974; Molina et al., 1996) and to mount appropriate CD4+ and CD8+ T cell responses in infection models (Kopf et al., 2002; Suresh et al., 2003). Similarly, patients with C3 gene mutations and a defect in C3 protein expression suffer from severe pyogenic infections in early childhood and some display low antibody titres (and lack or have reduced vaccination responses) (Ghannam et al., 2008). The insufficient lymphocyte activation observed in C3-deficient animals and patients has been attributed to defective dendritic cell (DC) maturation leading to suboptimal antigen presentation and costimulation because C3a receptor (C3aR)-mediated signals induce major histocompatibility complex (MHC) class II and CD86 upregulation on DCs (Ghannam et al., 2008; Peng et al., 2006; Li et al., 2008; Reis et al., 2008). Recent data indicate that direct signalling of complement receptors on the T cells themselves is required for CD4+ T cell effector lineage induction and regulation. Activation of cultured human CD4+ T cells through T cell receptor (TCR) stimulation leads to C3a and C3b generation (Cardone et al., 2010). Subsequent engagement
A. Ghannam et al. / Molecular Immunology 58 (2014) 98–107
of the complement regulator CD46 on the CD4+ T cell surface via cell-derived C3b (CD46 ligand) in the presence of endogenous IL-2 is an absolute requirement for IFN-␥ production and Th1 induction: Purified T cells from CD46-deficient patients are unable to assume a Th1 phenotype in vitro and in vivo (but have normal Th2 responses) and these patients suffer consequently from recurrent viral infections (Le Friec et al., 2012). Interestingly, rodents do not express CD46 on somatic tissue (Tsujimura et al., 1998) and a functional complement protein homologue has not yet been identified, suggesting that there are differences in these induction pathways between mice and humans (Cope et al., 2011). Nonetheless, complement receptor-mediated signalling events also participate in CD4+ T cell effector function in mice: Experiments using T cells isolated from C3aR (C3AR1) and C5aR (C5AR1) double-knock-out mice demonstrated the need for engagement of these T cell anaphylatoxin receptors for cell proliferation and Th1 cytokine production (Liu et al., 2005; Strainic et al., 2008). Furthermore, these studies are in line with observations made using human CD46-activated T cells (Cardone et al., 2010; Le Friec et al., 2012) and support a current paradigm shift in the complement field: they demonstrate that the required complement fragments C3a and C5a are not serum-derived but generated by the APC or T cell or both and that such immune cell-derived local production is fully sufficient for Th1 induction (Heeger and Kemper, 2012). The potential role of anaphylatoxin receptors expressed by human CD4+ T cells in the induction of effector functions is unexplored and individuals with deficiencies in either C3aR or C5aR expression have not been described. However, we have been able to obtain blood samples from two rare patients with C3 gene mutations. Importantly, although 30 patients from 22 families and 15 mutations have been described world-wide (OMIM ID: 120700); in reality, obtaining blood samples from C3-deficient individuals is difficult. We describe here the immunological phenotype of CD4+ T cells in regards to Th1 and Th2 induction from Patient C3-1 (who had previously been described (Ghannam et al., 2008)), and from Patient C3-2 with a novel molecular defect in C3 protein generation. Using purified CD4+ T cells from these two patients (these T cells cannot generate C3a and C3b locally), we demonstrate that normal Th1 induction requires the two T cell-generated C3 activation fragments C3a and C3b. Importantly, Th2 induction is independent of local complement activation. Additionally, both C3a and C3b may also participate in the contraction phase of human Th1 responses, which is characterized by co-secretion of IL-10 by IFN-␥-positive Th1 cells (Cardone et al., 2010), further underpinning the developing role of complement not only in the initiation of inflammatory responses but also in their negative control (Kemper and Atkinson, 2007).
2. Materials and methods 2.1. Patients and donors Blood was collected from patients and healthy donors after informed consent was obtained. Since early childhood, the C3deficient Patients C3-1 and C3-2 suffered from recurrent pyogenic infections, were appropriately vaccinated and submitted to antibiotic prophylaxis (Ghannam et al., 2008; Dejoie et al., 2009). At 10 years of age, Patient C3-1 developed anti-nuclear antibodies (Sm ENA specificity). He was then subjected to an i.v. immunoblobulin regimen and concomitantly recovered from systemic infections. The blood samples were collected in between infectious disease episodes in the cases of both patients. Each experimental condition was performed in triplicate and data shown are the mean value of three data points generated in one experiment. Control conditions using healthy donors have been performed with an n = 4 or n = 5.
99
2.2. Molecular analysis of C3-deficient patients The mutation in the C3 gene of Patient C3-1 has previously been published (Ghannam et al., 2008). To define the C3 mutation of the here newly described Patient C3-2, total RNA was prepared using TRIzol® (Invitrogen) from cultured monocytes after 24 h activation with 0.1 mg/ml E. coli LPS. C3-specific overlapping sequences were amplified using the Reverse Transcriptor® system (Roche Molecular Biochemicals) with primers chosen using the cDNA reference sequence (Gen Bank accession NM 000064) from the generated cDNA fragments. Amplicons were directly sequenced using the ABI PRISM Big Dye Terminator Cycle Sequencing v3.0 Reaction Kit, and were analyzed on an ABI 3100 DNA Analyzer (Applera). The two mutated alleles were identified as carrying c.2002delC, p.Arg668AlafsX35 (exon 16), also found in the mother (not shown) and c.2290C > T, p.Arg764Stop (exon 18) changes.
2.3. Antibodies and proteins Cell-stimulating mAbs were bought from BD Pharmingen (antihuman CD28, CD28.2), purified from a specific hybridoma line (anti-CD3; OKT-3) or generated in house (anti-CD46; TRA-2-10 (Wang et al., 2000)). Cell surface marker and protein expressions were assessed using fluorochrome-labelled antibodies to CD4 (555349), CD8 (555635), CD25 (555431), CD45RA (555488), CD45RO (559865), and CD127 (557938) obtained from BD Biosciences. The antibody to CD122 (FAB224A) was obtained from R&D Systems and the anti-human C3a receptor (561178) antibody was bought from BD Pharmingen. FACS analyses were performed on a FACSCalibur® instrument (BD Biosciences) using the CellQuest® software (BD Biosciences). Recombinant human IL-2 was obtained from Genzyme Therapeutics Ltd. The complement activation fragments C3a (A118), C3b (A114), C5a (A144) and iC3b (A115) were serum-purified proteins pooled from different donors and purchased from Complement Technologies Inc.
2.4. T cell purification and activation PBMCs were isolated from heparinized blood by FicollHypaque® density gradient centrifugation and CD4+ T cells then positively isolated using magnetic beads (CD4 Isolation Kit II® , Dynabeads® (Dynal Biotech) or the CD4+ Magnetic Bead Purification Kit® (Miltenyi Biotec)), according to the manufacturer’s protocol. Purity of isolated cell populations were consistently >95%. Because of limiting cell numbers obtained from the patient samples, purified CD4+ T cells were cultured in 96-well plates at 1 × 105 cells/wells in wells precoated with activating antibodies in the presence or absence of IL-2 or purified complement proteins. A small number of T cell activation experiments were performed in 48-well plates (2–2.5 × 105 cells/well; Figs. 1C and 2D and these condition usually led to better T cell activation with higher cytokine secretion/cell when compared to cultures activated in 96-well plates. Complement activation fragments were added at the following concentrations: C3a (80 nM), C5a (80 nM), C3b (80 g/ml), iC3b (80 g/ml). 2.5. Cytokine, C3b and C3a measurements IFN-␥, IL-2, IL-4 and IL-10 production by cells was assessed using the CBA® Human Th1/Th2 Cytokine Kit II® (BD Biosciences) or the appropriate Cytokine Secretion Assays® (Miltenyi BioTec), according to the manufacturer’s protocol. The presence of C3b or C3a in serum or cell culture media was measured utilizing the ‘Complement fragment 3b, C3b ELISA Kit’ from MyBiosource (San
100
A. Ghannam et al. / Molecular Immunology 58 (2014) 98–107
Fig. 1. Mutations in the C3-deficient patients, their local complement production and composition of their CD4+ , CD8+ , and natural Treg pool. (A) Schematic diagram of the C3 protein domains. The codon changes in the protein based on mutations in the C3 gene of Patient C3-1 and Patient C3-2 are indicated by arrows. (B) Chromatograms of exons 16 and 18 from DNAs of Patient C3-2 and an healthy control. The chromatogram of Patient C3-2 exon 16 was also seen in his mother (GenBank ID: KC493376, KC493377). (C) Measurement of serum and PBMC-derived C3a and C3b proteins. C3b and C3a content in either serum samples from 3 healthy donors and from Patient C3-2 or in cell culture supernatants from CD3 + CD46-activated T cells (12 h) was measured by ELISA or by using the Anaphylatoxin Bead Array, respectively. (D) Assessment of CD4+ , CD8+ and Treg cell numbers in Patients C3-1 and C3-2 in comparison to healthy donors. Stainings for patient cell populations were performed in triplicate with one representative FACS plot of each result shown. Cell populations derived from healthy donors were assessed from five distinct individuals (n = 5) in duplicate with one FACS plot showing the average cell number expression from one selected donor. For the assessment of CD4+ and CD8+ T cells, the gate was set on the lymphocyte portion of PBMCs (upper three panels). For the measurement of Treg cell numbers, the FACS gate was set on the CD4+ portion of the lymphocyte population (lower three panels). Treg cells were identified as CD4+ CD25+ CD127low/− . HD, healthy donor.
Diego, USA) or the ‘Human Anaphylatoxin Bead Array’ from BD Biosciences, respectively. 2.6. RNA silencing siRNA targeting the human C3aR (siRNA sc42840) and the negative control siRNAs (sc37007 and sc36869; fluorescein-labelled) were purchased from Santa Cruz Biotechnology, Inc. siRNA was delivered into primary human CD4+ T cells by electroporation
(2 × 106 cells ml/transfection buffer (Ambion); 3 g/ml siRNA; 200 V and 325 mF using the Bio-Rad Gene Pulser (Bio-Rad Laboratories). Transfection efficiency was consistently above 75% and protein knock-down peaked at 20–36 h post transfection. 2.7. Statistical analysis Statistical analyses were performed using the Student’s twotailed t test and Bonferroni correction for multiple comparisons
A. Ghannam et al. / Molecular Immunology 58 (2014) 98–107
101
Fig. 2. Human Th1 induction in CD4+ T cells requires complement activation fragments C3a and C3b. (A) IFN-␥, IL-4 and IL-10 secretion by T cells isolated from the C3deficient patients and a healthy donor (representative values among four individual donors tested) upon indicated activation conditions (36 h post activation). (B) CD4+ T cells upregulate C3aR expression upon TCR engagement as assessed by flow cytometry (6 h post activation). (C) Inhibition of C3aR signalling reduces Th1 induction. Purified T cells from healthy donors were activated under the depicted conditions in the presence or absence of an anti-C3aR antagonizing antibody and IFN-␥, IL-10 and IL-4 production measured 36 h post activation (n = 3). Error bars indicate means ± SEM; *p < 0.05; **p < 0.01 when compared to isotype control-treated cells. (D) Knock-down of C3aR protein expression via siRNA reduces Th1 induction. Purified T cells were transfected with a control siRNA or a C3aR-specific siRNA and C3aR protein expression assessed via Western Blotting 20 h post transfection (upper panel). Cells were then stimulated in 48-well plates as shown and IFN-␥ production measured 24 h post activation. (E) Addition of C3a (and C3b) significantly increases Th1 induction in normal CD4+ T cells. T cells from healthy donors were either left non-activated (NA), CD3-, CD3+CD28- or CD3+CD46activated (top panel to bottom panels) with or without addition of C3b and/or C3a, and cytokine production measured 36 h post activation (n = 4). Shown are results ± SEM; *p < 0.05; **p < 0.01 when cytokine secretion is compared to similarly activated cells but without addition of C3a or C3b. Note that IL-4 production (Th2 induction) is not affected by C3a and C3b, in line with observations when using T cells from both C3-deficient patients.
102
A. Ghannam et al. / Molecular Immunology 58 (2014) 98–107
(Excel® software, Windows® package). When possible, data were represented as mean ± SEM. Values were considered significant at p < 0.05. 2.8. Ethics statement The Institutional Review Board of Grenoble University Hospital (South-East committee V) approved these studies. Investigations were performed in accordance with the principles of the Helsinki declaration and with the French ethical policies employed when using biological sample collection and processing (Ministry of Health authorization). 3. Results 3.1. The molecular defect in C3-deficient patients C3-1 and C3-2 Patient C3-1 was previously described (Ghannam et al., 2008) and has a Ser550Pro mutation (Fig. 1A). This mutations leads to the generation of an immature, non-secreted C3 protein and the patient presents with compromised functions of several vital immune cells including DCs, and B and T cells (Ghannam et al., 2008). Patient C3-2 carries the nonsense mutation c.2290C > T (p.Arg764Stop, resulting in a shorter protein; paternal transcript) in exon 18 and the single nucleotide deletion c.2002Cdel (resulting in a frameshift at Arg668AlafsX35, maternal transcript) in exon 16 and is described here for the first time. Fig. 1B shows the pertinent portion of the C3 gene sequences harbouring the mutations of patient C3-2, compared to the control sequence. These abnormalities have not been detected in 100 alleles from healthy controls and they should hypothetically lead to the expression of non-secreted, abrogated/shorter and unstable proteins. Similar to Patient C3-1 (Ghannam et al., 2008), Patient C3-2 has an undetectable serum C3 level (
