List of Supplementary Materials
Supplementary Figures
Supplementary Figure 1. Isolation of RGC from rat retina.
Supplemementary Figure 2. Double immunohistochemistry to localize cleaved caspase-3 (C-CASP3) in rat retina 7 days after ONC.
Supplementary Figure 3. Western blot analysis of caspase-2 expression and cleavage in rat retinas at 7 days following ONC.
Supplementary Figure 4. Assessment of target knockdown activity of siCASP2 in rat PC12 cells and serum stability of siCASP2.
Supplementary Figure 5. Modeling of human and rat caspase-2 mRNA folding in the vicinity of siCASP2 cleavage site.
Supplementary Figure 6. Assesment of seed- and passenger strand-mediated off-target activities of siCASP2.
Supplementary Figure 7. Lack of activation of innate immune response by siCASP2 in human peripheral blood mononuclear cells.
Supplementary Figure 8. Cy3-labelled siRNA uptake by RGC.
Supplementary Figure 9. In vivo knockdown of caspase-2 mRNA using siCASP2 assessed in enrinched RGC.
Supplementary Figure 10. Sustained neuroprotection of RGC by siCASP2 in ONC model.
Supplementary Materials and Methods
Antibodies
PCR primers
Testing siCASP2 activity in vitro
Caspase-2 immunohistochemical detection in frozen retinal sections
Activated caspase-2 immunohistochemical detection in paraffin retinal sections
Protein extraction and western blotting
Assessment of siCASP2 off-target activity
Ago2 immunoprecipitation from siCASP2-transfected PC3 cells and identification of co-immunoprecipitated siRNA strands
Testing of siRNA-induced innate immune response in peripheral blood mononuclear cells (PBMC)
Supplementary Figures
Supplementary Figure 1. Isolation of RGC from rat retina. (a). Analysis of the purity of retinal cell populations following separation on Thy1.1(CD90.1)-conjugated magnetic beads. Bound and unbound cells were stained with anti-Thy1.1 antibodies and analyzed by FACS. The histogram of non-stained cells is shaded with grey. The data indicate that the unbound cell fraction is contaminated with ~11% of CD90.1-positive cells, whereas the purity of the bound cell fraction is ~73%. (b). qPCR analysis of RGC marker mRNAs for NEFL (neurofilament light polypeptide; upper panel) and Thy1.1 (lower panel) in bound and unbound cell fractions.
Supplementary Figure 2. Western blot analysis of caspase-2 expression and cleavage in rat retinas at 7 days after ONC. (a) Western blotting and (b) subsequent densitometry (± SD) confirms that ONC elevates the levels of procaspase-2 and enhances cleavage of capase-2 in retinal lysates, detected using the p12 (12kDa) cleaved caspase-2 antibody. For example, there was a 2-fold increase in the levels of the 12kDa cleaved caspase-2 band after ONC.
Supplementary Figure 3. Double immunohistochemistry to localize cleaved caspase-3 (C-CASP3) in rat retina 7 days after ONC. Cleaved caspase-3 (C-CASP3) immunolocalisation was absent from (a) βIII-tubulin-positive RGC and (b) GFAP-positive astrocytes in the rat retina after ONC but was present in cells of the inner nuclear layer (INL). Scale bars in (a) and (b) = 50µm. GCL = ganglion cell layer; INL = inner nuclear layer; ONL = outer nuclear layer.
Supplementary Figure 4. Assessment of target knockdown activity of siCASP2 in rat PC12 cells and serum stability of siCASP2. (a). siCASP2 was transfected into PC12 cells using Lipofectamine. Greater than 60% caspase-2 mRNA knockdown was achieved in PC12 cells with an IC50 of 0.291 nM. Small horizontal bars represent individual data points. (b). siCASP2 was incubated (at a final concentration of 0.08 mg/mL) in complete human serum (final serum concentration ~95%) at 37oC over the 24 hour period. Aliquots containing 0.4 mg siRNA eachwere taken at the indicated time points and transferred to 15 µL of 1.5xTBE-loading buffer, frozen (until use) and then analyzed on non-denaturing 20% polyacrylamide gel with ethidium bromide post-staining. A control non-exposed siRNA sample was suspended in PBS at the same final concentration and loaded on the same gel as a size control.
Supplementary Figure 5. Modeling of human and rat caspase 2 mRNA folding. Folding of rat and human caspase-2 mRNAs was modelled using RNAfold web server (http://rna.tbi.univie.ac.at/cgi-bin/RNAfold.cgi). The structure was predicted using a loop-based minimal energy model and the dynamic programming algorithm introduced by Zuker et al. (1981) and (2003) (1). The best predictions are shown.
The local energy of the most stable conformation of target mRNA in the region that aligns with siRNA was calculated asthe sum of the hydrogen bonds DG’s between the nucleotides comprising this region. A color scale reflects the probability (purple being the lowest and red being the highest) of existence of particular local structures based on their presence in 50 predicted minimal energy mRNA folding models.(a)and (b). Low and high resolution images, respectively, of human caspase-2 mRNA folding. (c) and (d). Low and high resolution images, respectively, of rat caspase-2 mRNA folding. Positions complementary to the siCASP2 used in this study are shown with a red line in all four images.
Supplementary Figure 6. Assesment of seed- and passenger strand-mediated off-target activities of siCASP2. (a-c). Testing seed- and passenger strand-mediated off-target activities of siCASP2 in siCHECK system: dose response siRNA activity curves (for technical details, see Supplementary Materials and Methods). Small horizontal lines represent individual data points. Error bars represent standard deviations. (a). siCASP2 guide strand (guide strand - complete match, GS-CM) specific inhibition of Renilla Luciferase activity with IC50 0.055 nM. (b). Lack of the siCASP2 guide strand seed region-mediated (guide strand – seed match, GS-SM) inhibition of Renilla Luciferase activity. (c). Lack of siCASP2 passenger strand-mediated (passenger strand – complete match, PS-CM) inhibition of Renilla Luciferase activity. Small horizontal bars represent individual data points. Error bars represent standard deviations. (d, e). Hybridization-based detection of sense (passenger) and antisense (guide) strands of siCASP2 in complex with immunoprecipitated AGO2 (argonaut-2; part of the RNA induced silencing complex (RISC)) protein. (d) Presence of AGO2 protein was assessed by immunoblotting with anti-Ago2 antibodies in the samples of the whole cytosolic extract (load), the immunoprecipitated (IP) and the unbound (UB) fractions. Immunoprecipitation with non-specific IgG was used as negative control. (e) RNA was extracted from aliquots of immunopreciptated protein fractions, separated on 15% denaturing PAAG, blotted to nylon membranes and examined for the presence of individual strands of siCASP2 by hybridization to 33P-end labelled strand-specific probes: an antisense strand (AS) specific probe for detection of sense strand and a sense strand specific probe for detection of the AS strand. Amounts of 1, 0.1 and 0.01 ng of siCASP2 siRNA were dissolved in 5 µL PBS and loaded on the same gel as the migration/quantity reference.
Supplementary Figure 7. Lack of activation of innate immune response by siCASP2 in human peripheral blood mononuclear cells (PBMC). (a). Fold mRNA induction of IFN-responsive genes in human PBMC treated ex vivo with different concentrations of siCASP2 or the TLR ligands, Poly(I:C) (Polyinosinic-polycytidylic acid; a synthetic analog of double-stranded RNA), or CL075 (TLR8 ligand), as assessed by real-time RT-PCR. (b).Fold of TNFa and IL-6 induction in human PBMC treated ex vivo with different concentrations of siCASP2 or the TLR8 ligand CL075, as assessed in cytokine-specific ELISA assays.
Supplementary Figure 8. Cy3-labelled siRNA uptake by RGC. Demonstration that Cy3-labelled siRNA was taken up efficiently by all RGCs in retinal wholemounts after only 5 hours (a) and persisted for 24 hours (b) post-intravitreal injection. Scale bars in a and b = 100 µm.
Supplementary Figure 9. In vivo knockdown of caspase-2 mRNA in the enriched RGC fraction isolated from retinas of intravitreally injected eyes. Groups of rats (n=6 for experiment shown in b and n=16 for experiment shown in a) received intravitreal injections of 20 μg of siCASP2, control siRNA (siCNL or siGFP), or same volume of PBS. Retinas were dissected out and RGC seperated using Thy1.1 coated beads, as described in RGC isolation section of Supplementary Methods. Only samples with at least 60% enrichment of RGC were further processed. For RNA extraction, 4 or 2 samples were pooled for experiments shown in (a) and (b), respectively, and caspase-2 mRNA levels were determined by real time RT-PCR analysis. An example of qPCR analysis of caspase-2 mRNA expression in the RGC of rats 72 hours (a) or 96 hours (b) after intravitreal injection with siCASP2 or control siRNA (siCNL or siGFP) or PBS. Each data point originates from a pool of RGC isolated from 2-4 independently injected eyes. The horizontal bars show median expression values. The values are expressed as arbitrary units. NS = not significant (p>0.05).
Supplementary Figure 10. Sustained neuroprotection of RGC by siCASP2 (day 30 after the ONC). (a) 12-fold more FG-labelled RGC (± SD) were present in retinal wholemounts after ONC and treatment with siCASP2 than with PBS, equating to (b) nearly 10% of the total RGC population found in intact retinas. (c) Representative FG-labelled images of retinal wholemounts taken 30d after ONC to demonstrate the sustained enhanced neuroprotection afforded by siCASP2 treatment. Scale bar in (c) = 50µm.
Supplementary Materials and Methods
Antibodies. Caspase-2 was recognised using rabbit polyclonal antibodies (Santa Cruz Biotechnology, San Diego, USA) at a concentration of 0.5 µg/ml; cleaved caspase-2 was recognised using rabbit polyclonal antibodies (Abcam, Cambridge, UK) at a concentration of 100 µg/ml and goat polyclonal antibodies (p12 fragment, Santa Cruz) at a concentration of 1 µg/ml. RGC were immunostained with a monoclonal antibody to βIII-tubulin (Sigma, Poole, UK) at a dilution of 1:400 (ascites). Specific primary antibody staining was detected using Alexa 488 anti-mouse IgG, Alexa 594 anti-rabbit IgG, Alexa 594 anti-goat IgG (all from Invitrogen, Poole, UK), HRP-conjugated anti-rabbit IgG (EnVision® System-HRP; DAKO, Carpinteria, CA, USA), HRP-labelled anti-goat IgG (Vector Laboratories, Peterborough, UK), all used at a concentration of 5 µg/ml. The purity of CD90.1-positive and negative RGC isolated cells was assessed by FACS analysis (Beckton Dickinson, Franklin Lakes, NJ, USA) after staining with anti-CD90.1 antibodies (eBioscience, San Diego, CA, USA). Cleaved caspase-3 was detected using rabbit anti-human cleaved caspase-3 IgG (Cell Sinalling Tech, Boston, MA, USA) while astrocytes were stained with a monoclonal anti-GFAP antibody (Sigma).
PCR primers
Official gene symbols are shown in parentheses.
Human Caspase-2 (CASP2) mRNA:
Forward: 5’-TGGTCCACCTTCCAGCACAA-3’
Reverse: 5’-GCTAGCACCACTCGGTTCTT-3’
Rat Caspase-2 (CASP2) mRNA:
Forward: 5’-CTGACAGGAGGAGCAGGATT-3’
Reverse 5’-CTAACAGTTCGCTCAGCAGC-3’
Human Cyclophilin A (PPIA) mRNA:
Forward 5’-GCACTGCCAAGACTGAG-3’
Reverse 5’-CCTGCAATCCAGCTAGG-3’
Rat Cyclophilin A (PPIA) mRNA:
Forward: 5’-CGACTGTGGACAGCTCTAAT-3’
Reverse: 5’-CCTGAGCTACAGAAGGAATG-3’
Rat beta-Actin (ACTB) mRNA:
Forward: 5’- AGAGCTATGAGCTGCCTGAC-3’
Reverse: 5’- AATTGAATGTAGTTTCATGGATG-3’
Human interferon-induced protein with tetratricopeptide repeats 1 (IFIT1) mRNA:
Forward: 5’-CCTGAAAGGCCAGAATGAGG-3’
Reverse: 5’- TCCACCTTGTCCAGGTAAGT -3’
Rat (interferon-induced protein with tetratricopeptide repeats 1)IFIT1 mRNA:
Forward: 5’ – GCCTACGTGAGACACCTGAA-3’
Reverse: 5’ – AGGTCACCAGGCTTCTCTTG-3’
Human interferon-induced 15 kDa protein (ISG15) mRNA:
Forward: 5’- GCAGATCACCCAGAAGATTG -3’ (Forward)
Reverse: 5’- CATTTGTCCACCACCAGCAG -3’ (Reverse)
Human 2'-5' oligoadenylate synthetase 1B (OAS1B) mRNA:
Forward: 5’- ACGCTGACCTGGTTGTCTTC -3’
Reverse: 5’- CCTGGACCTCAAACTTCACG -3’
Rat 2'-5' oligoadenylate synthetase 1B (OAS1B) mRNA:
Forward: 5’-TGATGTGCTTCCAGCCTATG-3’
Reverse: 5’-TGCGCTCACTGATGAGATTG-3’
Human interferon-inducible protein p78 (MX1) mRNA:
Forward: 5’- GCAGCCAGTATGAGGAGAAG -3’
Reverse: 5’- AAGGGCAACTCCTGACAGTG -3’
Rat neurofilament, light polypeptide (NEFL) mRNA:
Forward: 5' – TCTGCGTACTCCAGCTACTC-3’
Reverse: 5' – AGCTGTGCCTTCTCCTGTGT-3’
Rat thymus cell antigen 1, theta (THY1.1) mRNA:
Forward: 5' – TGCCTGGTGAACCAGAACCT-3’
Reverse: 5' – AGGCTGAACTCATGCTGGAT-3’
Testing CASP2 siRNA activity in vitro. Human cervical carcinoma HeLa cells (ATCC) were cultured in DMEM medium (Gibco), supplemented with 10% fetal bovine serum (FBS) and 100 U/ml penicillin and 100 ug/ml Streptomycin (Biological Industries, Israel). Rat adrenal pheochromocytoma PC12 cells (ATCC) were cultured in DMEM medium (Gibco), supplemented with 2 mM L-glutamine, 8% FBS, 8% horse serum, 100 U/ml penicillin and 100 µg/ml Streptomycin (Biological Industries, Israel). Cells were seeded to 50% confluency before transfection with different concentrations of siRNA (0.035-100 nM) using Lipofectamine 2000 (Invitrogen) as described by the manufacturer for adherent cell lines. Then 48 hours after transfection, total RNA was isolated using EZ-RNA kit (Biological Industries), reverse transcribed with Reverse Transcriptase Superscript II (Invitrogen) and random primers (Invitrogen) prior to real-time quantitative PCR (qPCR) amplification of CASP2mRNAusing gene-specific primers. Caspase-2 mRNA levels were normalized to those of the CYNA mRNA for each sample. siRNA target mRNA knockdown activity at each concentration was defined as the ratio between Caspase-2 mRNA quantity in siRNA-treated samples and mock-transfected control samples or samples transfected with siCNL. IC50 was determined by sigmoid curve fitting using the 3-point curve fit method (SAS 9.2 Software) (SAS, Cary, NC, USA).
Caspase-2 immunohistochemical detection in frozen retinal sections. Rats were euthanized by CO2 narcosis and intracardially perfused with 4% formaldehyde (TAAB Laboratories). Following excision, retinas were fixed for a further 2 hours in 4% paraformaldehyde, washed X3 changes in PBS for 30 mins each and cryoprotected in 10%, 20% and 30% sucrose. Samples were then embedded in OCT mounting medium (Miles Inc, PA, USA). Unlesioned control retinas (day 0) were also processed similarly. Sections, 10 µm thick were cut on a cryostat (Bright Instrument Co. Ltd., Huntingdon, Cambridgeshire, UK) and collected onto X-traTM adhesive slides and processed for immunohistochemistry. For double immunofluorescent staining, sections were post-fixed in 100% ethanol for 1 min, washed X3 in phosphate buffered saline (PBS), permeabilized in 0.1% Triton X-100, washed, blocked and incubated with the appropriate primary antibody diluted in PBS containing 0.5% bovine serum albumin (BSA; Sigma) and 0.05% Tween 20 (PBST-BSA) (Sigma), overnight at 4oC. Sections were then washed in PBS and incubated with the appropriate fluorescent labelled secondary antibody (either Alexa-488 or Texas Red; Molecular Probes, Oregon, USA) diluted 1:400 in PBST-BSA for 1hr at room temperature and after further washes in PBS, were mounted in Vectashield containing DAPI (Vector Labs, Peterborough, UK). Antibody specificity was verified by pre-incubating the relevant primary antibodies with excess of peptides used for their generation (Santa Cruz and Abcam) at a peptide:antibody ratio of 20:1 for 2 hrs at room temperature prior to incubation overnight and completing the staining procedure as described above. Sections were viewed under a Zeiss epifluorescent microscope (Zeiss, Hertfordshire, UK) and images captured using an AxioCam® HRc controlled by Axiovision® Software (Zeiss, Version 4,2). All images were compiled in Adobe Photoshop CS3 (Adobe Systems, San Jose, CA, USA).