Age-related macular degeneration associated polymorphism rs10490924 in ARMS2 results in deficiency of a complement activator
- 1.5k Downloads
Age-related macular degeneration (AMD) is the leading cause of blindness in developed countries. The polymorphism rs10490924 in the ARMS2 gene is highly associated with AMD and linked to an indel mutation (del443ins54), the latter inducing mRNA instability. At present, the function of the ARMS2 protein, the exact cellular sources in the retina and the biological consequences of the rs10490924 polymorphism are unclear.
Recombinant ARMS2 was expressed in Pichia pastoris, and protein functions were studied regarding cell surface binding and complement activation in human serum using fluoresence-activated cell sorting (FACS) as well as laser scanning microscopy (LSM). Biolayer interferometry defined protein interactions. Furthermore, endogenous ARMS2 gene expression was studied in human blood derived monocytes and in human induced pluripotent stem cell-derived microglia (iPSdM) by PCR and LSM. The ARMS2 protein was localized in human genotyped retinal sections and in purified monocytes derived from AMD patients without the ARMS2 risk variant by LSM. ARMS2 expression in monocytes under oxidative stress was determined by Western blot analysis.
Here, we demonstrate for the first time that ARMS2 functions as surface complement regulator. Recombinant ARMS2 binds to human apoptotic and necrotic cells and initiates complement activation by recruiting the complement activator properdin. ARMS2-properdin complexes augment C3b surface opsonization for phagocytosis. We also demonstrate for the first time expression of ARMS2 in human monocytes especially under oxidative stress and in microglia cells of the human retina. The ARMS2 protein is absent in monocytes and also in microglia cells, derived from patients homozygous for the ARMS2 AMD risk variant (rs10490924).
ARMS2 is likely involved in complement-mediated clearance of cellular debris. As AMD patients present with accumulated proteins and lipids on Bruch’s membrane, ARMS2 protein deficiency due to the genetic risk variant might be involved in drusen formation.
KeywordsMicroglia Cell Polymorphism Rs10490924 Normal Human Serum Indel Mutation ARMS2 Gene
Age-related macular degeneration
Age-related maculopathy susceptibility 2
Fluorescence-activated cell sorting
Induced pluripotent stem cell-derived microglia
Laser scanning microscopy
Polymerase chain reaction
Age-related macular degeneration (AMD) is a multi-factorial disease and a prevalent cause of visual impairment in developed countries . Genome-wide association studies revealed that variations in or near the complement genes CFH [2, 3, 4, 5], CFI , CFB , and C3  are significantly associated with AMD. Thus, inappropriate complement activation and innate immunity are linked to the pathogenesis of AMD . The complement system is a major part of innate immunity and plays an essential role in cellular homeostasis, tissue remodeling, as well as in host defense and inflammation [9, 10]. Deregulated complement function or uncontrolled activation due to defective regulation has been implicated in AMD and other diseases including C3-glomerulopathy, paroxysmal nocturnal hemoglobinuria, and systemic lupus erythematosus [9, 10, 11].
In addition to mutations in complement genes, a polymorphism (rs10490924) in ARMS2 shows the highest association with AMD with an estimated relative risk of 8.1 for homozygous carriers [12, 13, 14]. The ARMS2 gene is present only in higher primates , and cellular expression and function of ARMS2 are widely unknown. Here, we focused on the role of ARMS2 in AMD pathophysiology and aimed at defining the functional consequences of the AMD associated polymorphism (rs10490924) in ARMS2. Our findings demonstrate ARMS2 expression in human monocytes and microglia cells and suggest that ARMS2 functions are directly linked to the complement system, thereby mediating the opsonization of apoptotic and necrotic cells.
The polymorphism in the ARMS2 gene, rs10490924 was described as highly associated with both forms of AMD leading to geographic atrophy (dry form) or neo-vascularization (wet form) [12, 13]. Patients diagnosed with the wet form of AMD according to the modified version of AMD study grading system (AREDS) as described previously by Spencer et al.  were genotyped for the polymorphisms in the ARMS2 gene rs2736911, rs10490924, and del443ins54 as described [2, 12]. Genomic DNA was extracted from 10 ml whole blood cells of each patient using the PAX gene blood DNA kit (QiaGen). ARMS2 was amplified with primers (forward 5′TGTCACCACATTATGTCCC3′ or 5′TGTCACTGCATTCCCTCCTGTCAT3′ and reverse 5′GGCACCACTCCAGAATTT3′ or 5′AAGCTTCTTACCCTGACTTCCAGC3′), and the PCR products were separated by agarose gel electrophoresis, visualized under UV light and subsequently validated by bi-directional sequencing on an automated DNA sequencer (ABI/1130x, Applied Biosystems). According to the presence of the polymorphisms rs2736911, rs10490924, and del443ins54 in the ARMS2 gene, three groups of genotypes were created (homozygous without these polymorphisms (type I/I), heterozygous for rs10490924 and del443ins54 (type I/II), homozygous for rs10490924 and del443ins54 (type II/II), and homozygous for rs2736911 (type III/III).
Human donor eyes
Retinal samples of controls and AMD patients were obtained from the Center of Ophthalmology Eye Bank, University of Cologne. Retina 1: type I/I, craniocerebral injury, unknown hour postmortem, age 22. Retina 2: type I/I, intracranial bleeding, 27 h postmortem, age 82. Retina 3: type I/I, hypoxia brain damage, 4.5 h postmortem, age 53. Retina 4: type II/II, exenteratio orbitae, 8 h postmortem, age 78.
CHO-K1 Chinese ovary hamster cells (ATCC-CCL-61), pgsD-677 heparan sulfate deficient CHO cells (ATCC CRL-2244), pgsA-745 xylosyltransferase 1 deficient CHO cells (ATCC CRL-2242), THP-1 human monocytes (ATCC TIB-202), RAW264.7 Mouse leukemic macrophages (ATCC TIB-71), and native RPE cells (InnoProt) were all cultivated according to the costumer’s advise. Human T cells, monocytes, and human erythrocytes were obtained from human blood samples of healthy volunteers. Human T cells, peripheral blood mononuclear cells (PBMCs), and erythrocytes were isolated with micro beads from Miltenyi Biotech, according to the manufacturer’s protocol. Apoptosis of cells was induced by incubation of the cells with 0.4 μg/ml staurosporine for 24 h and necrosis by 1 h at 65 °C. Apoptosis was confirmed by PI and annexin V-pacific blue positive staining (Life Technologies) using flow cytometry. Human microglia cell lines (iPSdM) were generated from induced pluripotent stem (iPS) cell lines obtained by reprogramming skin fibroblasts as previously described . The cells proliferate without addition of growth factors, and they were passaged 1:3 twice a week. The microglia phenotype was confirmed by flow cytometry by expression of CD11b, CD11c, CD14, CD16, CD32, CD36, CD45, CD206, CX3CR1, and TREM2.
ARMS2 gene expression (PCR)
Cells (about 1 × 106 from each cell type) with or without stimulation by 400 ng PMA LPS for 24 h or of 10 ml whole blood were harvested and homogenized; total RNA was isolated using the PAX gene blood RNA kit (QiaGen). A 20 ng of isolated RNA was transcribed into cDNA using QuantiTect® Reverse Transcription Kit (QiaGen). cDNA was amplified using Phusion PCR Kit (New England Biolabs), and primers situated in exon 1 (5′TCGGTGGTTCCTGTGTCCTTCATT3′) and exon 2 (5′TCACCTTGCTGCAGTGTGGATGAT3′) of ARMS2 or for amplification of actin (forward 5′ACCAACTGGGACGACAT3′; reverse 5′CTAGAAGCATTTGCGGTG3′). Synthesis was performed by denaturing for 60 s at 96 °C followed by annealing for 30 s at 69 °C (ARMS2) or 60 °C (actin) and synthesis for 60 s (ARMS2) or 120 s (actin) at 72 °C for 35 cycles. Extension was performed for 480 s at 72 °C. Amplified PCR products were separated in agarose and visualized under UV light. Bands were excised, purified, and sequenced on an automated DNA sequencer (ABI/1130x, Applied Biosystems). Sequences were aligned to ARMS2 genomic sequences (NM_001099667).
Expression and purification of recombinant ARMS2
ARMS2 was recombinant expressed using the Pichia pastoris expression system. The expression vector pPICZB contained codon usage optimized full length cDNA of the ARMS2 gene coupled to a myc and 6 × histidine coding tag for purification (Life Technologies) (Additional file 1: Figure S1). P. pastoris cells (strain X33) were transformed with the recombinant expression vector, and ARMS2 expressing clones were selected via zeocin containing selection medium according to the standard protocols. Expressed His-tagged proteins were purified by Ni2+-chelate affinity chromatography from the cells [18, 19]. Cell lyses were performed in binding buffer (10 mM Na2HPO4, 10 mM NaH2PO4, 10 mM imidazol, 1 M NaCl, 2% (v/v) Nonidet P-40, 20% (v/v) glycerol; pH 7.0) together with glass beads (Roth) by FastPrep®-24 (MP Biomedicals) at 4 m/s for 60 s. Protease activity was inhibited by addition of complete EDTA-free (Roche). Protein was eluted from HisTrap™ HP column (GE Healthcare) using elution buffer (10 mM Na2HPO4, 10 mM NaH2PO4, 500 mM imidazol, 500 mM NaCl; pH 7.0). Purified ARMS2 was concentrated in storage buffer (10 mM Na2HPO4, 10 mM NaH2PO4, 500 mM NaCl; pH 7.0). In addition to the recombinant protein, three ARMS2 peptides were generated. Peptide 1 included amino acids 1 to 40, peptide 2 amino acids 41 to 70, and peptide 3 amino acids 71 to 107 (Jerini Peptide Biotechnologies). Polyclonal ARMS2 antiserum was generated by immunization of rabbits with the here described recombinant ARMS2 protein expressed in P. pastoris (Davids Biotechnologies). Generated ARMS2 antiserum (ARMS2Jena) was purified with HiTrap Protein A HP 1 ml column (GE Healthcare) and the major epitope of these antibodies determined by pepspot analysis. Peptides of ARMS2 (33 peptides of 13 amino acids with three amino acids overlap) were spotted to a membrane (Jerini Peptide Technologies), incubated with the antiserum and secondary anti-rabbit antibodies for detection. Polyclonal rabbit antiserum generated to the N-terminus of ARMS2 (ARMS2com) was purchased from CovaLab. In addition, a mouse monoclonal antibody was generated to the C-terminal peptide 3 of ARMS2 (aa 71–107 = IHTELCLPAFFSPAGTQRRFQQPQHHLTLSIIHTAAR) (ARMS2mAb) by standard methods.
Recombinant ARMS2 was separated by SDS-PAGE under reducing conditions (50 mM TRIS-HCL, 1.6% (w/v) SDS, 7% (v/v) glycerol, 8 M UREA, 4% (v/v) β-mercaptoethanol, 0.016% (w/v) bromophenol blue; pH 6.8) and visualized either by silver staining or immune-blotting using ARMS2com, ARMS2Jena antiserum, or mouse α-Penta-His monoclonal antibodies (QiaGen). Mass of whole deglycosylated recombinant ARMS2 was determined by mass spectrometry. To remove the carbohydrate side chains from recombinant ARMS2 (10 μg), the protein was incubated with 0.5 units of PNGase (Roche) for 3 h at 37 °C or the purification tag was cleaved off by enterokinase (New England Biolabs) overnight at RT and repurified with a HisTrap™ HP column.
For immunoprecipitation, monoclonal α-ARMS2 antibody (20 μg) was loaded onto Protein G Magnetic beads (New England BioLabs) for 1 h at 4 °C in binding buffer (20 mM sodium phosphate buffer, pH 7.0). THP-1 cells (1 × 107) were lysed in 1 ml lysis buffer (150 mM NaCl, 1% NP-40, 25 nM Tris-HCl, 1 mM EDTA, 5% glycerol) containing 1 mM PMSF, centrifuged at 16,000 g for 10 min at 4 °C, added to the antibody-loaded beads and incubated on a rotating shaker overnight at 4 °C. After removal of the supernatant, the beads were washed and proteins were eluted from the beads in 30 μl of elution buffer (0.1 M glycine, pH 2.7). The eluted proteins were evaluated for the presence of ARMS2 by SDS-PAGE and mass spectrometry. To show endogenous ARMS2 expression, monocytes were isolated from fresh blood and about 1 × 106 cells were stressed by addition of H2O2 (0.001–1 mM) for 1 h. Supernatants were replaced by cell culture medium, and cells were incubated for 20 h. Cells were lysed in buffer (50 mM TRIS-HCL, 1.6% (w/v) SDS, 7% (v/v) glycerol, 8 M UREA, 4% (v/v) β-mercaptoethanol, 0.016% (w/v) bromophenol blue; pH 6.8), centrifuged and supernatants were separated by SDS PAGE, blotted to a membrane and ARMS2 detected with αARMS2Jena.
Recombinant ARMS2 or immunoprecipitated ARMS2 from THP-1 cells were separated by SDS-PAGE and stained with coomassie blue. Protein bands were excised from the gel and washed repeatedly in water and 50 mM NH4HCO3/acetonitrile 1 + 1 (v/v) for 15 min. Gel pieces in acetonitrile were dried by vacuum centrifugation. Gel pieces were soaked in 50 mM NH4HCO3 with 10 mM DTT for 45 min at 56 °C followed by alkylation (50 mM iodoacetamide in 50 mM NH4HCO3 for 30 min in the dark). Upon washing in 50 mM NH4HCO3/acetonitrile, gel pieces were incubated in trypsin digestion buffer (20 ng/μl trypsin in 25 mM NH4HCO3) overnight at 37 °C. The peptides were extracted from the gel plugs by incubation in acetonitrile/trifluoroacetic acid 1 + 1 (v/v) for 30 min at RT. Extracted peptides (1 μl) were premixed with the same volume of the matrix α-CHCA solution and spotted on a matrix-assisted laser desorption ionization (MALDI) plate. Mass was analyzed using MALDI TOF-mass spectrometer (UltrafleXtreme, Bruker Daltonics, Germany) in the reflector mode with appropriate m/z range and laser intensity. For analysis of the whole ARMS2 protein, the recombinant protein was deglycosylated and untagged as described above. Analysis of the protein, the instrument was utilized in linear mode. Data analysis was performed with current NCBInr database using Mascot search with a peptide mass tolerance of 100 ppm. The analysis was repeated three times.
Detection of endogenous ARMS2 by LSM microscopy
For detection of endogenous ARMS2 protein, monocytes of seven ARMS2 genotyped individuals (two (I/I), one (I/II), two (II/II) one (III/III)) were isolated from PBMCs derived from 20 to 40 ml human blood samples. Blood samples were diluted 1:1 in PBS and overlaid with 15 ml Ficoll. After centrifugation at 1600 rpm for 20 min, PBMCs were collected and washed in 10 ml PBS. Monocytes were isolated from PBMCs using a Percoll gradient (3 ml of PBMCs were overlaid by 6 ml of 54% Percoll) centrifuged as above, harvested, and washed in ice-cold PBS. Cells (1 × 106) were transferred into a 4-well slide chamber (Lab-Tek®) and stained for laser scanning microscopy as previously described . Briefly, isolated monocytes or THP-1 cells in chamber slides were fixed with paraformaldehyde (3% at RT for 15 min) permeabilized in Triton X-100 (0.3%) (Roth) 5 min on ice, blocked with FcR blocking reagent (Miltenyi) 10 min at 4 °C and stained with polyclonal rabbit ARMS2 antiserum (αARMS2Jena) (1:200), monoclonal ARMS2 antibodies (αARMS2mAb) (1:200) followed by incubation with Alexa-647 labeled anti-rabbit or anti-mouse (1:400) for 1 h at RT. DNA was stained with DAPI (10 μg/ml) for 15 min at RT. After washing, the samples were examined by LSM (LSM 510, Carl Zeiss, Jena).
CHO-K1 cells (~1 × 105) were incubated with 10 μg ARMS2 and properdin, either alone or together. After 1 h, cells were washed and placed on a chamber slide (Nunc). After 1 h, adherent cells were fixed with 3% paraformaldehyde for 15 min. CHO surface-bound ARMS2 was visualized using rabbit ARMS2 antiserum (1:100) and Alexa-488 conjugated secondary antibody (1:200; Life Technologies) in assay buffer (1% BSA in RPMI media). Properdin was detected by mouse monoclonal properdin antibody (1:100; QuiDel) and secondary Alexa-647 conjugated antibody (1:200; Life Technologies) in assay buffer. DAPI (Sigma) was used to stain DNA. Images were taken by LSM 710 Meta (Zeiss) and co-localization was analyzed with Zen 2009 software (Zeiss). All incubation steps were done in RPMI media at RT.
Eyes were soaked in phosphate buffered saline (PBS) containing 4% formaldehyde followed by punching of a tissue strip ranging from the optic nerve to the fovea/parafovea which was then processed for paraffin embedding. Sections of 12 μm thickness derived from 2 genotyped individuals (genotype (I/I) and (II/II)) were obtained and subjected to immunohistochemical staining. After de-paraffinization by a series of graded ethanol slides were placed in a 97 °C hot water bath in antigen retrieval buffer (0.05 M Tris buffered saline, 0.05% Tween-20, pH 9.0) for 20 min. Slides were then washed with 1% goat serum in 1xPBS-T (PBS with 0.4% Triton-X) for 10 min for two times. To prevent unspecific binding, sections were covered with blocking buffer (5% goat serum in 1xPBS-T) in a humidified chamber and incubated for 30 min at room temperature. Primary antibodies included anti-ARMS2 (1:200, rabbit polyclonal), anti-CD68 (1:200, monoclonal rat anti-mouse, AbDSerotec), and anti-CD68 (1:200, monoclonal mouse anti-human, Dako). Primary antibodies were incubated overnight at 4 °C. After the two washing steps with 1% serum PBS-T, sections were incubated with labeled secondary antibodies coupled to Alexa-488 (green) or Alexa-594 (red) (Jackson ImmunoResearch, West Grove, PA, USA). Retinal nuclei were counter-stained with DAPI and mounted in DAKO fluorescent mounting medium (Dako Deutschland GmbH, Hamburg, Germany) and analyzed on an Axioskop 2 MOT plus Apotome microscope (Zeiss, Jena, Germany).
ARMS2 binding to complement proteins
C3, C3b, C3c, C3d, iC3b, factor B, properdin (all Complement Technologies) and human serum albumin (HSA) (Nutritional Biochemicals) (each 500 ng in PBS) were coated onto MediSorp microtiter plates (Nunc) over night at 4 °C. Wells were blocked with buffer I (AppliChem) for 2 h at RT and incubated with 500 ng ARMS2 in assay buffer (2% (w/v) BSA, 0.05% (v/v) Tween-20 in PBS) for 1 h at RT. ARMS2 was detected with ARMS2 antiserum diluted 1:1000 and a secondary HRP conjugated antibody 1:2000 in CrossDown buffer (AppliChem) at 492 nm.
Properdin binding to ARMS2
Recombinant ARMS2 or HSA (each 500 ng in PBS) was coated onto MaxiSorp microtiter plates (Nunc) over night at 4 °C. Wells were then washed with PBS + 0.05% (v/v) Tween-20 and blocking was performed with blocking buffer (5% (w/v) BSA, 0.05% (v/v) Tween-20 in PBS) for 2 h at RT. After another washing step, wells were incubated with increasing amounts of properdin (12.5–100 nM) in assay buffer (2% (w/v) BSA, 0.05% (v/v) Tween-20 in PBS). After incubation for 1 h at RT, wells were washed and bound properdin was detected with a goat properdin antiserum diluted 1:2000 and a secondary horse radish peroxidate (HRP)–conjugated antibody diluted 1:2000 in assay buffer. Multiscan Ascent ELISA Reader measured absorption at 492 nm. To determine the interaction domain in ARMS2 responsible for properdin binding to ARMS2, peptide 1, peptide 2, and peptide 3 (each 500 nM) were coated to the microtiter plate, incubated with 2 μM properdin (TECOmedical GmbH) in PBS and binding of properdin was detected as described before. Primary monoclonal properdin antibody (1:1000)  and secondary anti mouse 1:2000 (Dako, Denmark) were used for detection.
The binding affinity of properdin to ARMS2 was measured using biolayer interferometry on a BLITZ system (Forte Bio). His-tagged ARMS2 was loaded onto Ni2+ NTA biosensors (Forte Bio) in assay buffer (PBS (Lonza) with 0.1% (m/v) gelatin). Loaded biosensors were blocked using assay buffer with 10 μg/ml biocytin (Sigma-Aldrich) for 60 s. After washing, the biosensors were dipped in 4-μl properdin solution with concentrations varying from 70 to 1800 nM properdin. The k D was determined by fitting data to a 1:1 model algorithm with the BLITZ software.
Recruitment of C3b by ARMS2 bound properdin
Properdin (100 nM) was bound to immobilized recombinant ARMS2 as described before. Wells were washed with assay buffer (2% (w/v) BSA, 0.05% (v/v) Tween-20 in PBS) and incubated with increasing amounts of C3b (4–12 nM) in assay buffer for 1 h at RT. Unbound C3b was removed by additional wash steps, and binding was detected at 492 nm by mouse C3d monoclonal antibody (1:2000, Quidel) and secondary HRP-conjugated antibody (1:2000) in assay buffer.
ARMS2 (500 nM) was incubated with 20% NHS in Mg2+EGTA buffer (1% (w/v) BSA, 20 mM HEPES, 144 mM NaCl, 7 mM MgCl2, 10 mM EGTA, pH 7.4) for 15 min at 37 °C. The generated Ba product was analyzed at 540 nm using a Ba MicroVUE ELISA Kit (QuiDel).
All fluoresence-activated cell sorting (FACS) analyses were performed with a LSR II flow cytometer (BD Science); 10,000 cells were evaluated and data were calculated using FlowJo Software (Tree Star).
CHO-K1, pgsD-677, pgsA-745 (each 1 × 106), native RPE (1 × 105), human erythrocytes (each 1 × 107) as well as isolated naive T cells (1 × 106) were incubated with increasing amounts of ARMS2 (10–500 nM) in assay buffer (1% (w/v) BSA in PBS) for 1 h on ice. ARMS2 binding was detected with rabbit ARMS2 antiserum (1:200) and an Alexa-647 conjugated antibody (Life Technologies) (1:400). Similarly, factor H (10–500 nM, Comptech) was bound to CHO-K1 cells.
Heparan sulfate competition assay
Recombinant ARMS2 (500 nM) was incubated with 1 × 107 heparin coated beads (GE Healthcare) and increasing amounts of heparan sulfate (0–12 mM, Sigma) in assay buffer (1% (w/v) BSA in PBS) for 1 h on ice. ARMS2 binding was detected with rabbit ARMS2 antiserum 1:200 and Alexa-647 conjugated antibodies (Life Technologies) 1:400. Factor H (500 nM, Comptech) was incubated with the beads and detected with factor H antiserum (Comptech).
Complement activation assays were performed as previously described . CHO-K1 and pgsA-745 (each 1 × 106) were incubated in 20% NHS, or heat-inactivated 20% NHS (hiNHS) (30 min at 56 °C) forms healthy human donors in Mg2+EGTA buffer (1% (w/v) BSA, 20 mM HEPES, 144 mM NaCl, 7 mM MgCl2, 10 mM EGTA, pH 7.4) for 45 min at 37 °C. C3b deposition was determined by flow cytometry using mouse α-C3b monoclonal antibody (FitzGerald) diluted 1:100 and secondary Alexa-647 conjugated antibodies (Life Technologies) diluted 1:200 in Mg2+EGTA buffer for 1 h on ice. Deposition was normalized and NHS alone defined as 100%. Influence of ARMS2, factor H (CompTech), or HSA (Nutritional Biochemicals) on C3b deposition was evaluated, after preincubation of either cells (1 h on ice) or serum (15 min at 37 °C) with increasing amounts of each protein (100 to 750 nM).
CHO-K1 cells (1 × 106) were preincubated with 500 nM ARMS2 for 1 h on ice. Washed cells were incubated either in 20% NHS diluted with Mg2+EGTA buffer (1% (w/v) BSA, 20 mM HEPES, 144 mM NaCl, 7 mM MgCl2, 10 mM EGTA, pH 7.4) or EDTA buffer (1% (w/v) BSA, 20 mM HEPES, 144 mM NaCl, 10 mM EDTA, pH 7.4) for 45 min at 37 °C or with 100 nM purified properdin (CompTech) in PBS containing 1% BSA for 1 h at RT. Properdin deposition was analyzed by flow cytometry using a properdin monoclonal antibody (QuiDel) (1:200) followed by an Alexa-647 conjugated antibody (Life Technologies) (1:400).
Significant differences between two groups were analyzed using the Student’s two-tailed t test. Values of *p < 0.05, **p < 0.01, ***p < 0.001 were considered statistically significant.
Human monocytes and iPS-derived microglia cells express ARMS2
Expression of recombinant ARMS2 and generation of antibodies
To detect the endogenous ARMS2 protein in monocytes, a new polyclonal rabbit ARMS2 antiserum was raised against a recombinant histidine-tagged ARMS2 protein expressed in P. pastoris. The polyclonal antiserum was characterized by the identification of the main epitope in ARMS2 bound by the antibodies. Therefore, 33 ARMS2 peptides each 13 amino acid long and with an overlap of three amino acids were spotted to a membrane and subsequently stained with the polyclonal ARMS2 antiserum. The antibodies bound predominantly to the C-terminal end of ARMS2 (Fig. 1b). When separated by SDS-PAGE under reducing conditions, recombinant ARMS2 appeared with mobilities of 17 and 15 kDa by immune staining using the new antiserum and also with a commercial ARMS2 antiserum or a His-tag mAb (Fig. 1c). The 17 kDa protein band was cut out, deglycosylated, digested by trypsin, and then analyzed by mass spectrometry. All peptides derived from the 17 kDa band covered regions (95%) of ARMS2. MS of the untagged, deglycosylated protein revealed a dominant protein peak with a mass of 11.349 Da.
Monocytes derived from AMD patients with the ARMS2 risk haplotype lack the ARMS2 protein
The ARMS2 risk variant is defined by the polymorphism rs10490924, which is linked to an indel mutation (del443ins54) in the 3′ untranslated region of the ARMS2 gene . The ARMS2 indel mutation is considered to lead to ARMS2 mRNA instability, as this mutation deletes the poly A signal and integrates two AUUUA motifs that mediate rapid mRNA turnover. To determine whether this ARMS2 risk variant (rs10490924 with del443ins54) affects ARMS2 expression, we determined ARMS2 expression on the protein level in peripheral blood monocytes isolated from genotyped AMD patients. Monocytes of three different genotypes were compared as follows: type I/I harbors two non-risk alleles of ARMS2. type I/II has one ARMS2 risk allele (rs10490924 with del443ins54) and a non-risk allele, and type III has two alleles of the ARMS2 risk variants (rs10490924 with del443ins54). Blood monocytes of the three genotypes were isolated from AMD patients, fixed in chamber slides, permeabilized, and stained with ARMS2 antiserum and with a monoclonal ARMS2 antibody. The ARMS2 protein was identified in the cytoplasm of monocytes harboring one or two alleles of the ARMS2 non-risk variant (i.e., type I/I or I/II) (Fig. 1d). In contrast, monocytes derived from the three patients with two risk alleles (II/II) showed no ARMS2 protein staining (Fig. 1c). Also, monocytes derived from a single patient homozygous with an ARMS2 stop mutation (rs2736911, genotype III/III) lacked the ARMS2 signal. To verify ARMS2 presence in the monocytes, ARMS2 was also stained in monocytic cells THP-1 using polyclonal antiserum (αARMS2Jena). Again, single spots were detected in the cytoplasm of the cells (Fig. 1d). In contrast, macrophages of murine origin lacked an ARMS2 signal. As the signal of ARMS2 in monocytes was very weak, we tested whether ARMS2 expression is enhanced under stress signals. Therefore, human blood monocytes were challenged for 1 h with H2O2 and grown over night and lysates investigated for ARMS2 expression by Western blot analysis. Interestingly, ARMS2 expression increased in monocytes upon oxidative stress (Fig. 1e). Having shown that ARMS2 is expressed in THP-1 cells, immunoprecipitation was performed combined with mass spectrometry to detect the endogenous ARMS2 protein. A new generated monoclonal ARMS2 antibody was bound to a protein G column, incubated with THP-1 whole cell lysate, washed and antibody bound proteins were eluted from the column. After separation via SDS PAGE and coomassie blue staining, single bands were extracted from the gel (Fig. 1f). Mass spectrometry revealed ARMS2 peptides in the cell lysate of THP-1 cells (Fig. 1f) and in probes derived from precipitated recombinant ARMS2 (not shown). Altogether, the results demonstrate for the first time ARMS2 protein expression in monocytes and that the risk variants of ARMS2 lead to ARMS2 deficiency in these cells.
ARMS2 binds to cell surfaces
ARMS2 binds to heparan sulfate
To identify cellular ARMS2 ligands, we studied whether ARMS2 cell surface binding is mediated by surface exposed glycosaminoglycans (GAGs). To this end, we analyzed ARMS2 binding to CHO-K1 (Chinese Hamster Ovary) cells, which expose sulfated GAGs and also to two mutant lines, which lack surface heparan sulfates. ARMS2 bound to wild type CHO-K1 cells but did not bind to the mutant CHOpgsA or to CHOpgsD cells (Fig. 3d). GAGs represent complex polysaccharides and are exposed at the surface of mammalian cells; GAGs sulfation provides a recognition signal for proteins . To confirm ARMS2 interaction with heparan sulfate in the absence of other cell surface ligands, we studied ARMS2 binding to heparin-coated beads. ARMS2 bound to the coated beads, and binding was blocked when ARMS2 was incubated with heparan sulfate prior to binding (Fig. 3e). Thus, ARMS2 binds to cell surfaces via GAGs.
Cell surface bound ARMS2 enhances complement activation
ARMS2 binds the human complement activator properdin
To clarify whether ARMS2-properdin complexes recruit C3b, binding of C3b to preformed ARMS2-properdin complexes was followed. ARMS2 was immobilized to a microtiter plate, properdin was attached, and subsequently, C3b was added. Purified C3b bound to the immobilized ARMS2-properdin complex as determined by ELISA, and binding was dose-dependent (Fig. 5c). Thus, surface-bound ARMS2 binds properdin and subsequently, ARMS2-properdin complexes recruit C3b. These results are in agreement with enhanced C3b opsonization when ARMS2 was coated onto CHO-K1 cells prior to challenge with complement active NHS (Fig. 4a). The ARMS2-properdin interaction was confirmed by biolayer interferometry. Recombinant ARMS2 was immobilized to a Ni2+ NTA biosensor, and purified properdin was added as analyte. In this set up, properdin bound to immobilized ARMS2 with a k D = 0.5 ± 0.07 × 10−6 M (Fig. 5d). In order to localize the domain in ARMS2, that interacts with properdin, three ARMS2 peptides encompassing all amino acids of ARMS2 were coated to a microtiter plate and incubated with properdin. Peptide 3 of ARMS2, but not peptides 1 or 2, showed binding to properdin (Fig. 5e). These results confirm interaction of ARMS2 with properdin and localize the interaction domain in the C-terminal part of ARMS2.
ARMS2 mRNA is expressed in human monocytes and also in stem cell induced human microglia cells, but not in murine macrophages. Expression in human but not in murine cells is in agreement with the relatively late evolutionary appearance of the ARMS2 gene in humans and higher primates . Using a new antiserum, which was generated by immunizing rabbits with recombinant ARMS2 proteins, endogenous ARMS2 protein expression was detected in monocytes derived from individuals with the wild type ARMS2 genotype and also in the human monocytic cell line THP-1, which is heterozygous for the risk ARMS2 haplotype. Similarly, ARMS2 was immunoprecipitated from THP-1 cells with a novel monoclonal ARMS2 antibody. However, the ARMS2 concentration was very low indicating that a trigger is necessary for ARMS2 expression in the cells. Indeed, after blood monocytes were stressed with H2O2, ARMS2 expression was induced and detected also by Western blotting. Thus, ARMS2 is expressed upon oxidative stress in monocytes and likely also microglia cells, which is in agreement with oxidation specific epitopes on damaged cells in the retina which induce complement activation (3). The recombinant ARMS2 protein, when separated under reducing conditions, appeared as a band with a molecular mass of about 17 kDa, which was identified as ARMS2 by mass spectrometry. The whole deglycosylated, untagged recombinant protein showed a mass of 11.349 kDa, which is smaller than the calculated mass of 11.4 kDa. The difference is explained by pre-treatment of the protein (deglycosylation and removal of the histidine tag). However, deglycosylation of recombinant ARMS2 did not alter binding of ARMS2 to cells or interaction with proteins (data not shown).
ARMS2 gene variations represented by two different polymorphisms, affect ARMS2 protein expression in monocytes. The risk variant represented by polymorphisms rs10490924 is linked with the indel mutation, which affects mRNA stability . The second polymorphism, i.e., rs2736911, which is very rare in homozygous setting in Caucasians but more frequent in individuals of Chinese origin , generates a premature stop codon at position 38. Apparently, both polymorphisms rs10490924 and rs2736911 result in ARMS2 protein deficiency as shown in monocytes and retinal section stainings. These results confirm the recent reports demonstrating that the indel mutation in the 3′ untranslated region of the ARMS2 gene leads to RNA instability [12, 25] and ARMS2 protein deficiency in placenta tissue, derived from a person with the ARMS2 risk genotype . These data may help to resolve the contradictory views that exist over ARMS2 as a key effector protein in AMD and on the functional consequences of the polymorphism rs10490924 in ARMS2 . The polymorphism rs10490924 is in strong linkage disequilibrium with SNP rs1100638 in the HTRA1 (high temperature requirement factor A1) gene [12, 34, 35] which makes effects by these polymorphisms indistinguishable in statistical analyses.
The innate immune system constantly monitors a spectrum of surfaces, including self, altered self and non-self. Potential targets are recognized on the basis of their characteristic surface molecular patterns. We here demonstrate that the recombinant ARMS2 protein like the complement regulator factor H binds preferentially to apoptotic and necrotic surfaces and identifies heparan sulfate as an ARMS2 ligand. Interestingly, surface-bound recombinant ARMS2, but not in fluid phase, serves as an anchor for the human complement activator properdin. ARMS2-properdin complexes formed at cellular surfaces enhanced C3 convertase formation and C3 convertases mediated C3b opsonization. Such marking of surfaces with C3 cleavage products accelerates recognition, uptake and clearance of debris by human phagocytes [21, 22], and binding of properdin to apoptotic cells enhances this process . Notably, in addition to surface complement activation, complement regulators like factor H and C4bp are recruited to the surfaces to restrict the complement cascade and to inhibit inflammation and autoimmune reactions [3, 21, 22].
The data presented suggest that ARMS2 assists in the clearance of cellular debris in the human retina. Recombinant ARMS2 distinguishes certain modified cells via surface glycosaminoglycan structures and initiates complement. Therefore, in case of ARMS2 deficiency complement-mediated opsonization, uptake and phagocytosis of dying cells are expected to be impaired. Over a long time, i.e., over years, such a reduced clearance may support accumulation of cellular debris in the retinal space and along Bruch’s membrane enhancing formation of drusen, the hallmark of AMD.
The authors thank all German AMD patients for their participation. We also thank Maria Pötsch (Leibniz Institute for Natural Product Reseach and Infection Biology, Jena) for MS analyses.
This research was supported by the German Council “Deutsche Forschungs-Gemeinschaft” SK46, Zi432, LA1206, the “Pro Retina” foundation and the Thuringian Ministry of Science and Education, Germany. HN is a member of the DFG-funded excellence cluster ImmunoSensation (EXC 1023). YL is a doctoral researcher at the International Leibniz Research School (ILRS), part of the Jena school of Microbial Communication (JSMC).
Availability of data and materials
Materials are available at firstname.lastname@example.org.
CS conceptualized this study, CS and PFZ wrote the manuscript and interpreted all results. SM, YL, H-MD, MA, and MvH performed protein expression, functional assays, microscopy, sequencing, surface plasmon resonance, and discussed data. PD, SL, and LS analyzed gene expression in monocytes. BHW and FG provided ARMS2 expression vectors. HN, TL, and MK generated, analyzed, provided iPSdM cells, and performed immunohistochemistry of human retinas. DP generated the ARMS2 mAb. AMJ and SJ recruited and evaluated subjects and provided blood samples. SF provided and analyzed retinal sections from human donor eyes. All authors approved and commented on it.
The authors declare that they have no competing interests.
Consent for publication
Ethics approval and consent to participate
All research conducted followed the tenets of the Declaration of Helsinki. The local ethics committee of the University of Cologne has approved the study under No. 14-247 and the Institutional Review Board of the Charité Berlin, Germany under study protocol No. EA2/004/2014.
- 2.Wang G, Spencer KL, Scott WK, Whitehead P, Court BL, Ayala-Haedo J, Mayo P, Schwartz SG, Kovach JL, Gallins P, Polk M, Agarwal A, Postel EA, Haines JL, Pericak-Vance MA. Analysis of the indel at the ARMS2 3′UTR in age-related macular degeneration. Hum Genet. 2010;127:595–602.CrossRefPubMedPubMedCentralGoogle Scholar
- 3.Weismann D, Hartvigsen K, Lauer N, Bennett KL, Scholl HP, Charbel Issa P, Cano M, Brandstätter H, Tsimikas S, Skerka C, Superti-Furga G, Handa JT, Zipfel PF, Witztum JL, Binder CJ. Complement factor H binds malondialdehyde epitopes and protects from oxidative stress. Nature. 2010;478:76–81.CrossRefGoogle Scholar
- 4.Lauer N, Mihlan M, Hartmann A, Schlötzer-Schrehardt U, Keilhauer C, Scholl HP, Charbel Issa P, Holz F, Weber BH, Skerka C, Zipfel PF. Complement regulation at necrotic cell lesions is impaired by the age-related macular degeneration-associated factor-H His402 risk variant. J Immunol. 2011;187:4374–83.CrossRefPubMedGoogle Scholar
- 5.Raychaudhuri S, Iartchouk O, Chin K, Tan PL, Tai AK, Ripke S, Gowrisankar S, Vemuri S, Montgomery K, Yu Y, Reynolds R, Zack DJ, Campochiaro B, Campochiaro P, Katsanis N, Daly MJ, Seddon JM. A rare penetrant mutation in CFH confers high risk of age-related macular degeneration. Nat Genet. 2011;43:1232–6.CrossRefPubMedPubMedCentralGoogle Scholar
- 6.Van de Ven JP, Nilsson SC, Tan PL, Buitendijk GH, Ristau T, Mohlin FC, Nabuurs SB, Schoenmaker-Koller FE, Smailhodzic D, Campochiaro PA, Zack DJ, Duvvari MR, Bakker B, Paun CC, Boon CJ, Uitterlinden AG, Liakopoulos S, Klevering BJ, Fauser S, Daha MR, Katsanis N, Klaver CC, Blom AM, Hoyng CB, den Hollander AI. A functional variant in the CFI gene confers a high risk of age-related macular degeneration. Nat Genet. 2013;45:813–7.CrossRefPubMedGoogle Scholar
- 8.Zhan X, Larson DE, Wang C, Koboldt DC, Sergeev YV, Fulton RS, Fulton LL, Fronick CC, Branham KE, Bragg-Gresham J, Jun G, Hu Y, Kang HM, Liu D, Othman M, Brooks M, Ratnapriya R, Boleda A, Grassmann F, von Strachwitz C, Olson LM, Buitendijk GH, Hofman A, van Duijn CM, Cipriani V, Moore AT, Shahid H, Jiang Y, Conley YP, Morgan DJ, Kim IK, Johnson MP, Cantsilieris S, Richardson AJ, Guymer RH, Luo H, Ouyang H, Licht C, Pluthero FG, Zhang MM, Zhang K, Baird PN, Blangero J, Klein ML, Farrer LA, DeAngelis MM, Weeks DE, Gorin MB, Yates JR, Klaver CC, Pericak-Vance MA, Haines JL, Weber BH, Wilson R, Heckenlively JR, Chew EY, Stambolian D, Mardis ER, Swaroop A, Abecasis GR. Identification of a rare coding variant in complement 3 associated with age-related macular degeneration. Nat Genet. 2013;45:1375–9.CrossRefPubMedPubMedCentralGoogle Scholar
- 14.Kanda A, Chen W, Othman M, Branham KE, Brooks M, Khanna R, He S, Lyons R, Abecasis GR, Swaroop A. A variant of mitochondrial protein LOC387715/ARMS2, not HTRA1, is strongly associated with age-related macular degeneration. Proc Natl Acad Sci U S A. 2007;104:16227–32.CrossRefPubMedPubMedCentralGoogle Scholar
- 18.Heinen S, Hartmann A, Lauer N, Wiehl U, Dahse HM, Schirmer S, Gropp K, Enghardt T, Wallich R, Hälbich S, Mihlan M, Schlötzer-Schrehardt U, Zipfel PF, Skerka C. Factor H-related protein 1 (CFHR-1) inhibits complement C5 convertase activity and terminal complex formation. Blood. 2009;114:2439–47.CrossRefPubMedGoogle Scholar
- 20.Pauly D, Nagel BM, Reinders J, Killian T, Wulf M, Ackermann S, Ehrenstein B, Zipfel PF, Skerka C, Weber BH. A novel antibody against human properdin inhibits the alternative complement system and specifically detects properdin from blood samples. PLoS One. 2014;9:e96371. doi: 10.1371/journal.pone.0096371.CrossRefPubMedPubMedCentralGoogle Scholar
- 25.Friedrich U, Myers CA, Fritsche LG, Milenkovich A, Wolf A, Corbo JC, Weber BH. Risk- and non-risk-associated variants at the 10q26 AMD locus influence ARMS2 mRNA expression but exclude pathogenic effects due to protein deficiency. Hum Mol Genet. 2011;20:1387–99.CrossRefPubMedPubMedCentralGoogle Scholar
- 26.Kortvely E, Hauck SM, Duetsch G, Gloeckner CJ, Kremmer E, Alge-Priglinger CS, Deeg CA, Ueffing M. ARMS2 is a constituent of the extracellular matrix providing a link between familial and sporadic age-related macular degenerations. Invest Ophthalmol Vis Sci. 2010;51:79–88.CrossRefPubMedGoogle Scholar
- 31.Cortes C, Ohtola JA, Saggu G, Ferreira VP. Local release of properdin in the cellular microenvironment: role in pattern recognition and amplification of the alternative pathway of complement. Front Immunol. 2013;3:412. doi: 10.3389/fimmu.2012.00412. eCollection 2012.CrossRefPubMedPubMedCentralGoogle Scholar
- 34.Fritsche LG, Igl W, Bailey JN, Grassmann F, Sengupta S, Bragg-Gresham JL, Burdon KP, Hebbring SJ, Wen C, Gorski M, Kim IK, Cho D, Zack D, Souied E, Scholl HP, Bala E, Lee KE, Hunter DJ, Sardell RJ, Mitchell P, Merriam JE, Cipriani V, Hoffman JD, Schick T, Lechanteur YT, Guymer RH, Johnson MP, Jiang Y, Stanton CM, Buitendijk GH, Zhan X, Kwong AM, Boleda A, Brooks M, Gieser L, Ratnapriya R, Branham KE, Foerster JR, Heckenlively JR, Othman MI, Vote BJ, Liang HH, Souzeau E, McAllister IL, Isaacs T, Hall J, Lake S, Mackey DA, Constable IJ, Craig JE, Kitchner TE, Yang Z, Su Z, Luo H, Chen D, Ouyang H, Flagg K, Lin D, Mao G, Ferreyra H, Stark K, von Strachwitz CN, Wolf A, Brandl C, Rudolph G, Olden M, Morrison MA, Morgan DJ, Schu M, Ahn J, Silvestri G, Tsironi EE, Park KH, Farrer LA, Orlin A, Brucker A, Li M, Curcio CA, Mohand-Saïd S, Sahel JA, Audo I, Benchaboune M, Cree AJ, Rennie CA, Goverdhan SV, Grunin M, Hagbi-Levi S, Campochiaro P, Katsanis N, Holz FG, Blond F, Blanché H, Deleuze JF, Igo Jr RP, Truitt B, Peachey NS, Meuer SM, Myers CE, Moore EL, Klein R, Hauser MA, Postel EA, Courtenay MD, Schwartz SG, Kovach JL, Scott WK, Liew G, Tan AG, Gopinath B, Merriam JC, Smith RT, Khan JC, Shahid H, Moore AT, McGrath JA, Laux R, Brantley Jr MA, Agarwal A, Ersoy L, Caramoy A, Langmann T, Saksens NT, de Jong EK, Hoyng CB, Cain MS, Richardson AJ, Martin TM, Blangero J, Weeks DE, Dhillon B, van Duijn CM, Doheny KF, Romm J, Klaver CC, Hayward C, Gorin MB, Klein ML, Baird PN, den Hollander AI, Fauser S, Yates JR, Allikmets R, Wang JJ, Schaumberg DA, Klein BE, Hagstrom SA, Chowers I, Lotery AJ, Léveillard T, Zhang K, Brilliant MH, Hewitt AW, Swaroop A, Chew EY, Pericak-Vance MA, DeAngelis M, Stambolian D, Haines JL, Iyengar SK, Weber BH, Abecasis GR, Heid IM. A large genome-wide association study of age-related macular degeneration highlights contributions of rare and common variants. Nat Genet. 2016;48:134–43.CrossRefPubMedGoogle Scholar
Open AccessThis 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. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.