Diagnostic Immunophenotyping and Serum Complement Activity Assays

Serum immunoglobulin levels and lymphocyte subpopulations were assessed as previously described [23]. Functional assessment of specific complement pathways in serum was performed using a well validated ELISA-based format [24].

Genomic DNA Isolation and cDNA Synthesis from RNA

Genomic DNA was isolated from cultured PHA T cell blasts using the QIAmp DNA blood Mini Kit. RNA was isolated from peripheral blood mononuclear cells (PBMCs) using Trizol (Thermo Fisher Scientific) for cell lysis. Chloroform (Sigma-Aldrich) was added and lysates were centrifuged for 15 min at 4°C 14’000 g according to the manufacturer’s protocol. RNA was purified with the QIAamp RNA Blood Mini Kit (Qiagen). RNA concentration was determined using the NanoDrop 2000c (Thermo Fisher Scientific). DNA was digested with DNase I (Promega). 100 ng/μl–1 μg/μl of RNA were used for cDNA synthesis. Random primers (Promega) were annealed at 70°C for 5 min. cDNA synthesis was performed according to the Qiagen’s GoScript Reverse Transcription System protocol in a TProfessional TRIO PCR Thermocycler (Core Life Sciences, Laguna Niquel, CA, USA).

End-Point PCR

Endpoint PCR was performed in a TProfessional TRIO PCR Thermocycler using the GoTaq G2 Polymerase (Promega) with a primer concentration of 0.5 μM and using DNA input template of 50–100 ng / 50 μl PCR reaction. PCR products were separated on a 1.5% agarose gel for the genomic DNA PCR or on a 2.8% agarose gel for the cDNA PCR. PCR bands were cut and DNA was isolated using the Quiaquick Gel Extraction Kit (Qiagen). Sanger sequencing was performed by Microsynth (St.Gallen, Switzerland). In CD46 mutation carriers, there was a 21-nucleotide deletion (GTAAGCCCCCAATATGTGAAA) detected skipping part of exon four.

Real-Time PCR

Real time PCR was performed using the syber green-based GoTag qPCR Master Mix (promega) on an Applied Biosystem ViiA 7 real time PCR machine with 0.5 μM primer concentration. Gene expression was normalized to ACTB/GAPDH using the calculation method designed by Michael W. Pfaffl [25].

PCR Primers

All primers were designed using the NCBI primer blast web tool (https://www.ncbi.nlm.nih.gov/tools/primer-blast/) and ordered as DNA oligos from Microsynth. The forward primer of the wild type (WT) and mutant CD46 real-time primer pair was set into the deleted part of exon 4.

Genomic DNA PCR detecting CD46 mutation:

cDNA PCR detecting CD46 deletion:

Primers for real-time PCR:

Wild type CD46:

Mutant CD46:

ACTB:

GAPDH:

Antibodies and Reagents

Antibodies to CD3 (OKT-3) and CD46 (TRA-2-10) for CD4⁺ T cell in vitro stimulation were kindly provided by Claudia Kemper (NHLBI/NIH). Agonistic anti-human CD28 (CD28.2) for co-stimulation was purchased from BD (Franklin Lakes, NJ, USA). Anti-human antibodies for flow cytometry are listed in the corresponding section. Mitotracker red/green/deep red dyes were purchased from Thermo Fisher (Waltham, MA, USA). Compounds for mitochondrial perturbation (oligomycin, FCCP and rotenone) were obtained from Sigma Aldrich (St. Louis, MO, USA).

CD4⁺ T Cell Purification and In Vitro Activation

PBMCs were isolated from freshly-drawn blood by centrifugation on after layering onto Lymphoprep separation medium (Corning, Vienna, VA). CD4⁺ T cells were then enriched by MACS separation using the MACS human CD4⁺ Positive T cell Isolation Kit (Miltenyi Biotech, Bergisch Gladbach, Germany) according to the manufacturer’s instructions. Purity of isolated CD4⁺ T lymphocytes was typically > 97%. Purified CD4⁺ T cells were either analyzed directly after isolation (ex vivo) or activated for 36 h in 48-well culture plates (Greiner, Monroe, NC) at a density of 2.5–5.0 × 10⁵ cells/well in media containing 50 U/ml recombinant human IL-2 at 37^oC and 5% CO2. 2 μg/ml stimulating antibodies (anti-CD3, anti-CD28 or anti-CD46 in the depicted combinations) were immobilized overnight at 4^oC on culture plates before CD4⁺ T cell plating.

Flow Cytometry

Freshly purified or stimulated CD4⁺ T cells were washed twice with FACS buffer, stained with a PE-conjugated anti-CD46 antibody (8E2, 1:50) (eBioscience/Thermo Fisher, San Diego, CA, USA) for 15 min at room temperature and washed twice with FACS buffer. For binding specificity mapping, PBMCs were stained similarly with anti-CD46-PE [TRA-2-10 (Biolegend, San Diego, CA, USA), E4.3 (BD), 8E2 (eBioscience), 1:100], anti-CD46-FITC (MEM-258, Biolegend, 1:100) or via a two-step stain of anti-CD46 GB24 mAb (10 μg/ml, kindly provided by Claudia Kemper) and rat anti-mouse IgG1-FITC (RMG1-1, 1:100, Biolegend). Mouse IgG1 isotype-PE (3.6.2.8.1, eBioscience; MOPC-21, Biolegend) and mouse IgG2a-PE (clone G155-178, BD) were used for isotype control staining. For ex vivo activation marker analysis, cells were laid in culture for 36 h and surface-stained with anti-CD3-Brilliant Violet 510 (UCHT1), anti-CD4-APC (SK3), anti-CD8-Brilliant Violet 421 (SK1), anti-4-1BB-PE-Cy7 (4B4-1) (from Biolegend), Fixable Viability Dye eFluor 780 and anti-OX40-PE (ACT-235) (from eBioscience).

Intracellular FoxP3 staining was performed on PBMCs with eBioscience FoxP3/transcription factor staining kit (eBioscience/Thermo Fisher) with anti-human FoxP3-PE (PCH-101) and rat IgG2a isotype control-PE (eBR2a) (from eBioscience), with prior surface co-staining of anti-CD127-APC (A019D5), anti-CD25-FITC (BC96), anti-CD8-Brilliant Violet 421 (SK1), anti-CD4-PE-Cy7 (SK3) and anti-CD3-Brilliant Violet 510 (UCHT1) (from Biolegend).

For mitotracker staining, freshly purified or stimulated CD4⁺ T cells were resuspended in 100 μl pre-warmed FACS buffer and 100 μl warm staining solution was added (FACS buffer containing mitotracker red or green and mitotracker deep red stain (100nM each) [23]. Cells were stained for 20 min at 37^oC, then washed 2 × with complete medium and 1 × with FACS buffer.

Samples were subsequently resuspended in FACS buffer and analyzed on the Accuri C6 cytometer (BD) or Cytoflex LX (Beckman Coulter, Brea, CA, USA). Data analysis was conducted using the FlowJo 10.0.8 software (FlowJo).

CD46 Protein Structure Prediction

A brouser-adapted Alphafold2 (ColabFold v1.5.2) [26] -based structure prediction of del21bp CD46-translated product was performed on 30 March 2023 with NP_758861.1 membrane cofactor protein isoform 3 precursor [Homo sapiens] utilized as the WT CD46 query.

Cytokine Measurements

Accumulated cytokine levels in the supernatant of purified CD4⁺ T cells stimulated for 36 h were assessed using the LEGENDplex human Th1 panel (Biolegend) according to the manufacturer’s instructions and analyzed on the Accuri C6 cytometer (BD). Data analysis was conducted using the FlowJo 10.0.8 software (FlowJo).

CD46 Reconstitution Assay

A transient CD46 overexpression of 21 base pair-deleted (del21bp) vs. WT CD46 GFP reporter/CMV-promotor pRP[Exp]-EGFP/Puro-CAG>hCD46[NM_172359.2] vectors (VectorBuilder, Chicago, IL, USA) was performed on a CD46-knocked out human myeloid HAP-1 cell line (Horizon Discovery, Cambridge, UK). Cells were maintained in Iscove's Modified Dulbecco's Medium (IMDM) (Gibco/Thermo Fisher) supplemented with 10% fetal bovine serum (FBS) (Cytiva, Marlborough, MA, USA) and 1% penicillin/streptomycin (Gibco/Thermo Fisher). To obtain the del21bp CD46 vector, the WT plasmid was subjected to KOD-plus-mediated site-directed mutagenesis (Toyobo, Osaka, Japan) and isolated for single colonies with the following primers:

The designed mutant plasmid was purified with Endofree Maxi plasmid purification kit (Qiagen, Hilden, Germany) and Sanger-sequenced for the full CD46 region to confirm correct mutagenesis. 1.0 × 10⁵ of CD46-knocked out HAP-1 cells were transfected with WT and/or del21bp CD46 vectors using Lipofectamine 3000 (Thermo Fisher) in 24-well flat-bottomed plates (Falcon/Corning) for 24 hours, and subsequently stained with anti-human CD46-PE (clone E4.3) for surface and intracellular expression analysis. Intracellular staining was performed with Cytofix/Cytoperm kit (BD). Samples were analyzed on FACS Lyric cytometer (BD) and data were analyzed with FlowJo 10.0.8 software (BD/FlowJo).

Measurement of Oxygen Consumption Rate (OCR) and Extracellular Acidification Rate (ECAR)

Stimulated CD4⁺ T cells were resuspended in serum-free unbuffered RPMI-1640 medium (Sigma Aldrich) and were plated onto Cell-Tak-coated (Corning, Reinach, Switzerland) XF96 Seahorse plates (Seahorse Bioscience, North Billerica, MA, USA) in triplicates at a density of 2.5x10⁵ viable cells per well. The Seahorse cartridge injection ports were loaded with oligomycin (1 μM), Carbonyl cyanide-4-(trifluoromethoxy) phenylhydrazone (FCCP, 2 μM) or rotenone (1 μM—all from Sigma Aldrich), respectively, and metabolic profiles were measured on a Seahorse XF96 Extracellular Flux analyzer (Seahorse Bioscience).

Metabolic parameters were calculated as follows:

1. basal OCR (total OCR, mitochondrial and non-mitochondrial) = OCR before mitochondrial perturbation

2. basal respiration = basal/total OCR – OCR after rotenone injection

3. Non-mitochondrial respiration = OCR after rotenone injection

4. ATP coupled respiration = basal/total OCR – OCR after oligomycine injection

5. Leak respiration = OCR after oligomycine injection – OCR after rotenone injection

6. maximal respiratory capacity = OCR peak after FCCP injection – OCR after rotenone injection

7. basal ECAR = initial ECAR before mitochondrial perturbation

8. maximal ECAR = ECAR peak after mitochondrial perturbation

Whole Exome Sequencing

Whole exome sequencing has been performed as recently described [23]. Rare immunodeficiency associated mutations were screened using the list of primary immunodeficiencies that has been recently reported [27]. All reported variants have been confirmed by genome viewer. Immunodeficiency-related gene variants were listed when the CADD Score was >15 and the minor allele frequency listed in gnomad (https://gnomad.broadinstitute.org) was <0.01. All listed variants had a sequencing coverage >25 with the exception of the TNFRSF4 and STXBP2 variants, where the coverage was <20 but the heterozygous variants were confirmed by integrated genome viewer (IGV).

NIHR BioResource Rare Diseases Project Database Analysis

The NIHR Bioresource Rare Disease project has provided whole genome sequencing data for more than 10’000 individuals with rare diseases [28]. We selected the rare exonic variants in the whole genome sequencing (WGS) data set of 13’037 individuals in a pre-defined set of PID/IEI genes [29], which were present in participants carrying rare CD46 variants, defining rare as absent from gnomAD at frequencies above 1/1,000 in all major populations, and selecting only variants having a CADD Phred score of at least 15. The listed CD46 variants have been linked to CD46 dependent human disease [15].

CD46 del21bp Mutation Is Loss-of Surface Expression and Lacks Dominant Negativity Regarding Surface Expression of WT CD46

We next addressed variant CD46 protein encoded by the del21bp CD46 mutation. The del21bp CD46 mutation results in a deletion of 6+1 amino acid residues at positions 152-157 and 159 (Gly-Lys-Pro-Pro-Ile-Cys- and Lys), flanking glutamic acid (Glu) residue 158 (Fig. ​(Fig.3a,3a, top). This results in a loss of two conserved prolines at positions 154-155 (PP motif) [31] and, importantly, a β sheet including one of the four universally CCP-conserved cysteines at position 157 building a disulfide bond with cysteine 127 [32–34] (Fig. ​(Fig.3a,3a, bottom). Flow-cytometric analysis showed that both diseased and healthy mutation carriers expressed lower CD4⁺ T cell-intrinsic cell surface CD46 levels compared to healthy controls ex vivo (Fig. ​(Fig.3b+c).3b+c). A comprehensive CD46-specific antibody surface binding mapping on peripheral CD4⁺ T cells with a set of epitope-mapped [35], commercially available and non-available (GB24) [36] anti-CD46 mAbs in the mutation-carrying SLE patient revealed approximately 50% reduced surface expression (Fig. ​(Fig.3d).3d). This was independent of N-terminal vs. C-terminal mAb binding specificity, arguing against intact surface expression of the mutant CD46 protein.

Open in a separate window

Fig. 3

del21bp CD46 is a loss-of-expression mutation and lacks dominant negativity. a Alignment (top) of exon skipping-mediated CD46 protein 6+1-residue deletion flanking glutamic acid (Glu) 158 and its orientation on a reported WT CD46 structure (middle) appended with a magnified view (bottom). Purple boxes show orientation of a CCP domain-conserved disulfide bond between deleted cysteine residue 157 and cysteine residue 127. Structure adopted/downloaded from Alphafold2 (https://alphafold.ebi.ac.uk/entry/P15529). b+c) Flow-cytometric analysis of CD46 surface expression on freshly purified CD4⁺ T cells ex vivo. Representative histogram plots are shown in b and MFIs are depicted in c. d Schematic of CD46-specific monoclonal antibody (mAb) binding sites and CD46 surface binding signal intensity in mutation carrier 5 (SLE) in comparison with unrelated healthy donor controls. Dotted line for 8E2 mAb indicates incomplete epitope identification in literature. Fluorescence deviation values compared with intra-patient isotype control staining are normalized against mean of healthy controls for each. CCP1-specific binding is shown in gray and CCP2/3/4 binding are shown in white for mutation carrier 5. e Representative flow-cytometric plot of transient CD46 overexpression via WT versus del21bp CD46 vectors by transfection of a CD46-knocked-out human myeloid HAP1 cell line. f Comparison of WT versus del21bp CD46 protein surface and intracellular expression positivity. Analyzed by unpaired t tests. g ColabFold v1.5.2-based structure prediction of del21bp CD46-variant protein (right) in comparison with queried/predetermined WT CD46 structure (left). Black wedges show a CCP3/CCP4-junction bending contrasting CCP1/CCP2/CCP3 alignment (highlighted in dotted line). Green wedge shows a de novo impairment of CCP2/CCP3. h CD46 expression upon WT and del21bp mutant CD46 co-transfection (1:1 ratio). P values represent Tukey’s post hoc multiple comparison tests of one-way ANOVA. Representative of two experiments performed in quadruplicate (f+h)

We then performed transient protein overexpression of the del21bp CD46 variant compared to WT CD46 in a CD46-knocked out human myeloid HAP1 cell line. WT CD46 reconstitution resulted in CD46 surface expression which was absent in the parental cell line (Fig. ​(Fig.3e,3e, columns 1–3 top). In contrast, reconstitution of del21bp CD46 resulted in a near-complete loss of surface CD46 expression (Fig. ​(Fig.3e,3e, column 4 top), defining the original intronic mutation as a loss-of-expression (Fig. ​(Fig.3f,3f, left). However, intracellular levels of the del21bp CD46 variant were comparable with WT CD46 (Fig. ​(Fig.3e+f).3e+f). Thus, the del21bp CD46 variant is expressed as a protein but does not reach the cell surface. This protein trajectory was analogous to a CCP cysteine-to-arginine/glycine substitution described in a case of human complement factor H deficiency [37, 38]. A browser-adapted AlphaFold2-based structure prediction of del21bp CD46-translated product suggested an inter-CCP bending, existing only between CCP3/CCP4 in WT [34] (Fig. ​(Fig.3g,3g, left, black wedge), to newly arise between CCP2/CCP3 (Fig. ​(Fig.3g,3g, right, green wedge), losing their near-linear alignment with CCP1 (Fig. ​(Fig.3g,3g, left, dotted box). Co-transfection of WT and del21bp CD46 at a 1:1 ratio resulted in 50% surface expression compared with WT-only, whereas intracellular expression level was comparable. Thus, the del21bp variant-encoded CD46 protein does not confer dominant negativity regarding surface expression of WT CD46 (Fig. ​(Fig.33h).

Immune Phenotyping of Diseased vs. Healthy CD46 Mutation Carriers

Next, we performed a detailed phenotypic analysis of lymphocytes, T and B cell subpopulations in all family members using a clinically validated protocol allowing us to link the results to normal in-house reference ranges [23]. Lymphocyte subpopulations were within normal reference ranges for all tested individuals with the exception of the patient with aHUS who showed low-normal CD4⁺ T cells in peripheral blood. Absolute B cells were normal (Supplemental Figure 3). B cell subpopulations were unremarkable with the exception of the patient with SLE who demonstrated high memory and low naïve B cells (Supplemental Figure 3). Four out of five mutation carriers had low naïve CD4⁺ T cells and low recent thymic emigrant (CD45RA⁺CD31^hi) CD4⁺ T cells (Fig. ​(Fig.4a).4a). The only mutation carrier with normal naïve CD4⁺ T cell frequencies was the patient with aHUS, which plausibly could demonstrate higher naïve CD4⁺ T cells due to the pediatric age at analysis [39] (Fig. ​(Fig.4a).4a). Within the CD8⁺ T cells, five out of five mutation carriers had elevated central memory (CD45RO⁺CD27⁺) subpopulations (Fig. ​(Fig.4a).4a). Intriguingly, while healthy mutation carriers had regulatory CD4⁺CD25^hiCD127^low CD4⁺ T cells (Treg) frequencies within normal reference ranges, both diseased mutation carriers had low Treg (Fig. ​(Fig.4a).4a). CD127^lowCD25^hi (surface-defined Tregs) vs. CD127^hiCD25^low CD4⁺ T cell populations in mutation carrier 5 (SLE) segregated with intracellular FoxP3 positivity, non-discrepant with two unrelated healthy donors (Fig. ​(Fig.4b),4b), in keeping with literature on Treg surface phenotyping [40, 41]. As tested in the SLE patient, all T-cell subpopulation frequencies were recapitulated when reassessed several years after initial testing, arguing for stable traits (Supplemental Figure 3c). With the exception of slightly elevated IgM in the SLE patient, none of the mutation carriers had dysregulated serum IgG, IgM, or IgA levels (Supplemental Figure 3d). The two main complement pathways were functionally assessed in sera of all tested family members (Supplemental Figure 3e) [24]. The classical complement pathway was functionally unaffected in all family members. Sera from four out of five mutation carriers had activities of the alternative pathway above the reference range, a finding requiring further scrutiny.

Open in a separate window

Fig. 4

Immune cell subpopulations in heterozygous CD46 c.475+1G>A mutation carriers. a CD4⁺ and CD8⁺ T cell subpopulations as indicated of all tested family members are depicted. The blue and green dots represent the aHUS and SLE patient, respectively. Closed symbols represent mutation carriers while the open symbol represents the family member lacking the mutated allele. The lines mark normal reference values. b Left: FoxP3 transcription factor positivity in mutation carrier 5 (SLE) CD4⁺ T cell subpopulations. Right: subpopulation FoxP3 positivity aligned with two unrelated healthy controls. Values are subtracted for positivity in isotype control staining

Cytokine Secretion and Immune-Metabolic Profiling of CD4⁺ T Cells of Diseased vs. Healthy CD46 Mutation Carriers

Normal CD46 abundance and activity is required for Th1 and CTL effector activity induction [8, 42]. We thus next measured cytokine production by purified CD4⁺ T cells in all family members plus 4-8 non-related healthy blood donors following T cell receptor ligation alone or in combination with agonistic anti-CD28 or anti-CD46 antibody co-stimulation. T cells of mutation carriers showed a trend to produce equal or more pro-inflammatory cytokines (IL-6, IFN-γ, TNF-α) following T cell receptor ligation plus CD28 co-stimulation compared to controls (Supplemental Figure 4), although these data did not reach statistical significance. This difference was normalized or reversed following CD46 co-stimulation, in keeping with impaired functional CD46 co-stimulation in mutation carriers. IL-10 secretion in CD46 co-stimulated mutation carrier-derived cells was approximately 50%-decreased (p = 0.0657) (Supplemental Figure 4), implicating certain involvement in the described immune-dysregulation linked with CD46 haploinsufficiency [5].

A major activity of CD46 during T cell activation is the induction of increased nutrient influx and T cell metabolic adaptation [9], including the support of glycolysis and oxidative phosphorylation (OXPHOS). To assess this, we first approximated mitochondrial mass and membrane potential in resting and stimulated conditions in T cells of all family members. As observed ex vivo, CD46 surface expression was lower in mutation carriers following in vitro CD3-only or CD3+CD28 CD4⁺ T cell stimulation (Fig. ​(Fig.5a).5a). No differences were found in mitochondrial parameters directly ex vivo in mutation carriers vs. controls (data not shown). However, 36 h following CD4⁺ T cell activation in vitro, the mean fluorescence intensity (MFI) of mitotracker deep red (indicating mitochondrial potential) was significantly lower upon CD46 co-stimulation in mutation carriers, likely reflecting decreased surface CD46 expression (Fig. ​(Fig.5b).5b). Next, the extracellular acidification rate (ECAR), as a surrogate of cellular lactate production (glycolysis), and the oxygen consumption rate (OCR) (OXPHOS) were measured in real-time by metabolic flux analysis in T cells from all family members and controls in the absence or following in-machine T cell activation for 36 h. There was no difference in mutation carriers vs. controls in basal or maximal ECAR induced by mitochondrial perturbation (data not shown). In contrast, basal and ATP-coupled respiration (Fig. ​(Fig.5c)5c) were reduced specifically following CD46 co-stimulation in activated T cells of CD46-mutated family members (Fig. ​(Fig.5d+e).5d+e). Maximal respiratory capacity and spare respiratory capacity were not affected by CD46 mutation status, not even following CD46 co-stimulation (data not shown). In conclusion, the identified CD46 loss-of-expression mutation was directly linked with altered T cell-intrinsic metabolic adaptation following CD46 co-stimulation.

Open in a separate window

Fig. 5

Impaired CD46-dependent immunometabolic adaptation in heterozygous CD46 c.475+1G>A mutation carriers. a CD46 expression (MFI) on CD4⁺ T cells following in vitro stimulation. Purified CD4⁺ T cells were left non-activated (NA) or stimulated with immobilized agonistic antibodies against CD3 or CD3 + CD28 for 36 hours before flow-cytometric analysis. b Purified CD4⁺ T cells were either analyzed ex vivo or left non-activated (NA) or stimulated with immobilized agonistic antibodies against CD3, CD3 + CD28 or CD3 + CD46 for 36 h. Flow cytometric analysis of the MFI of mitotracker deep red T cell fluroescence as a marker of mitochondrial membrane potential and the MFI of mitotracker green staining as a marker for mitochondrial mass of CD4⁺ T cells was performed. The ratio of the two MFI’s represents mitochondrial function per mitochondrial mass. c An explanatory graph showing cellular oxygen consumption rate and the respiratory indices after perturbation with oligomycin, FCCP and rotenone in the Seahorse flux analyzer. d + e) Quantification of T cell intrinsic oxygen consumption. Purified CD4⁺ T cells were left non-activated (NA) or stimulated with immobilized agonistic antibodies against CD3, CD3 + CD28 or CD3 + CD46. 36 h post stimulation, basal c and ATP-coupled d oxidative phosphorylation (OXPHOS, OCR) in CD4⁺ T cells was measured on the Seahorse metabolic extracellular flux analyzer. All bars indicate means ± SD. Compared by unpaired t-tests as indicated

We thank the patients and their relatives for participating in this study.

We thank NIHR BioResource volunteers for their participation, and gratefully acknowledge NIHR BioResource centres, NHS Trusts and staff for their contribution. The NIHR BioResource – Rare Disease Study is a multi-center whole-genome sequencing (WGS) study of approximately 10’000 patients [28]. Members of the NIHR BioResource consortium provided clinical and genetic data and assisted with interpretation of results. We thank the National Institute for Health Research, NHS Blood and Transplant, and Health Data Research UK as part of the Digital Innovation Hub Program. The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health and Social Care.