Differential Macrophage Polarization Promotes Tissue Remodeling and Repair in a Model of Ischemic Retinopathy
Valentina Marchetti1┼, Oscar Yanes2┼, Edith Aguilar1, Matthew Wang1, David Friedlander1, Stacey Moreno1, Kathleen Storm5, Min Zhan5,Samia Naccache3,Glen Nemerow3, Gary Siuzdak2,4, Martin Friedlander1*
Departments of Cell Biology1, Molecular Biology2and Immunology3 and the Scripps Center for Mass Spectrometry4, The Scripps Research Institute, La Jolla, CA, 92037; Source MDX5, Boulder, CO, USA
┼ These authors equally contributed to this work.
* To whom correspondence may be addressed.
Supplementary Material
Supplementary information contains: Supplementary Figures and Legends, Supplementary Material and Methods. Supplementary References.
Supplementary Figures and legends
Supplementary Fig1
A) Mouse model of OIR.C57BL/6J 7-day-old pups (post-natal day 7; P7) and their mothers are placed in 75% oxygen chambers for 5 days followed by return to ambient room air at P12. The retinas are analyzed at P17 when maximum pathology is observed in C57BL/6J mice. Exposure to hyperoxia from P7 to P12 leads to attenuation of normal retinal vascular growth and obliteration of the central retinal vessels in mice as early as P9. Red dots show examples of obliteration areas close to the optic nerve. Upon return to normoxia, the vaso-obliterated central retina becomes hypoxic tissue leading to abnormal, pathological NV in C57BL/6J mouse retinas by P17 (green arrows).
Supplementary Fig2
CD14+ cells express higher levels of anti-oxidative stress genes when injected in OIR retinas compared to freshly isolated CD14+ cells. Relative mRNA expression of four human genes in CD14+ injected retinas compared to the CD14- treated eyes at day 0 and P17. The glucoronidase B (GUSB) and the transforming growth factor 1 (TGFB1) show similar levels of expression in the CD14+ cells compared to the negative fraction when freshly isolated from the blood (day0, white columns) (n=6, n=number of hCB-CD14+/CD14- sample). The catalase (CAT) and the superoxide dismutase 1 (SOD1) are down regulated in the positive fraction compared to the control population. After 11 days in the mouse retina, the CD14+ cells up regulate the expression of these four genes (black columns) shifting significantly from the expression level observed initially at day 0 (n=6, n=number of retinas extracted and analyzed).
Supplementary Fig 3
Human cord blood-derived CD14+ cells reduce the levels of oxysterols and normalize the level of 11-cis-retinalin OIR retinas.
Boxplots showing the differences in theretina levels of oxysterols(A: 5,6-epoxy-cholesterol, B: 7-hydroxycholesterol, and C: 7-ketocholesterol)and 11-cis- retinal (D) attributed to the OIR modelin C57BL/6J or BALB/cByJ mice, and the treatment with human cord bloodCD14+ cells. The n shows the number of independently prepared retina replicates analyzed.
Supplementary Fig4
A) Toxic effect of oxycholesterols on astrocytes is observed in a dose-dependent response.Cultured astrocytes were exposed tothree different concentrations of epoxy-cholesterol (green), 7hydroxycholesterol (blue) and 7-ketocholesterol (red) overnight. The number of dead cells was calculated by trypan blue exclusion.B) And C) Overnight treatment of astrocytes C8D1Aand HUVEC cells with three different metabolites induces cell death in vitro. (B) C8-D1A astrocytes are treated overnight with 40 g/ml ofepoxy-cholesterol, 7-hydroxycholesterol and 7-ketocholesterol. 4.8% and 0% of the astrocytes survived after treatment with epoxy-cholesterol and 7-ketocholesterol, respectively (n=3, n=number of experiments; Bonferroni corrected t-test *: P<0.001). The epoxy-cholesterol and 7-ketocholesterol show an activity comparable to the positive control H2O2, while the same concentration of 7--hydroxycholesterol does not show this toxic effect. (C) The endothelial cell line HUVEC at confluence is treated overnight with 40 g/ml of epoxy-cholesterol and 1mM of H2O2as a positive control. The percentage of dead cells is counted by trypan blue exclusion. 4.6% of the cells treated with the epoxy-cholesterol survived after overnight treatment (n=3, Bonferroni corrected t-test *: P<0.001).
Supplementary Fig5
Oxygen treatment induces caspase 3 activation in P15 and P18 retinas. Cleaved Caspase 3 is increased in OIR retinas at P15 and P18 compared to normal air condition retinas (NOX).
Supplementary Fig 6
Pro and anti-inflammatory cytokine secretion by M1 and M2 CD14+ derived cells. The level of pro inflammatory (IL8, IL1, IL6) and anti inflammatory cytokines (IL10 and IL12p70), secreted by human CB CD14+ cells differentiated in M1 and M2 populations, is measured by a flow cytometry bead array. M1 up regulates the level of proinflammatory molecules compare to M2 that instead induces expression of the antinflammatory cytokine IL10.
Supplementary Material and Methods
Staining and quantification of retinal vasculature.
Quantification of obliteration and neovascularization area was done at P12 and P18.
The retinas were fixed in 4% PFA for 1 hour on ice followed by blocking in 50% FBS/20% normal goat serum for 1 hour at room temperature. An overnight incubation was performed in Griffonia Simplicifolia (GS) lectin conjugated to a fluorochrome to identify vessels. Retinas were laid flat with radial relaxing incisions to obtain whole mount preparations. To distinguish between the human cells and endogenous mouse macrophages, immunohistochemistry was performed using anti mouse F4/80 (CALTAG laboratories) oran antibody to the mouse Mannose Receptor (CellsScience). Images of vasculature in superficial, intermediate and deep retinal plexus were obtained using a Radiance MP2100 confocal microscope (BioRad; Zeiss). Quantification of vaso-obliteration and neovascularization (at P12 and P18) was done as described previously1. One-way Anova or Student’s t test was used to statistically compare the different experimental groups as indicated in the Figures. Volocity software was used to quantify the number of recruited lectin positive cells at P12.
Quantitative PCR (QPCR) Gene Expression Analysis
Isolation of mouse retinal tissue and human cord blood derived CD14+ RNA was performed using the RNeasy Mini Kit (QIAGEN) following the manufacturer’s recommended procedure. RNA quality was assessed using the Agilent Bioanalyzer 2100.
First strand cDNA was synthesized from random hexamer-primed RNA templates using TaqMan® Reverse Transcription reagents (Applied Biosystems) for stem cell and retinal samples and the High Capacity cDNA Reverse Transcription Kit (ABI) for the CB injected retinal samples. High capacity cDNA templates were pre-amplified with all target gene primer probe sets prior to QPCR analysis.
Target gene amplification was performed in a multiplexed QPCR reaction using TaqMan® Universal PCR Master Mix (ABI) and Source MDx proprietary primer-probe sets (Precision Profiles™) designed to amplify human-specific sequences, on the 7900HT Fast Real-Time PCR System (ABI).
Data files were reviewed and delta Ct values (Ct) calculated for all target genes (normalization was against an endogenous housekeeping gene). The Ct method was then used to determine relative expression values for all genes (i.e. a Ct value of 1 equals a two-fold difference in expression).
Cytokine assay
After 4 days the level of pro and anti-inflammatory cytokines was tested by bead array (BD Pharmingen). Samples were thawed on ice and analyzed for the presence of human IL-8, IL-1beta, IL-6, IL-10, TNF and IL-12p70 using the Cytometric Bead Array Human Inflammatory Cytokines Kit (BD Pharmingen). Samples were run undiluted as well as serial dilutions of 1:10 and 1:100 to account for variation in concentration of the different cytokines. Data was collected on a BD FACScanto flow cytometer and analyzed using FCAP Array Software (BD Biosciences).
Metabolite extraction and mass spectrometry analysis
Retinas were frozen at -80 ºC until use, at which point metabolites were extracted and proteins precipitated by adding 600µL of cold acetone. Samples were submerged 1 min in liquid nitrogen, thawed (2-3 minutes), and sonicated 5 min. This process was repeated three times and samples were stored at -20 ºC for 1h. The pellet was removed by centrifuging at 16 000g for 10 min and 400 µL of cold methanol/water/formic acid (86.5/12.5/1.0) was added to the pellet. Samples were sonicated 10 min and stored at -20 ºC for 1h. The pellet was again removed by centrifuging at 16 000g for 10 min, and supernatant was pooled with the first extraction in the clean tube.Reverse phase chromatography was performed using a 150 x 0.5mm Zorbax C18 column with 5µm particles, at a flow rate of 20 µL/min.Data was collected in continuum mode from m/z 100 to 1000 using an Agilent ESI-TOF. The capillary voltage was 3500 V, with a nebulizer gas flow of 15 L/min. We used the XCMS program to align and analyze the LC-MS data 2. To confirm the identification of significant ions, we used a quadrupole-TOF (Agilent 6520 Accurate-Mass Q-TOF).
Liquid chromatography-MS (LC-MS)
Four µL processed retinas were injected for each run. Reverse phase chromatography was performed using a 150 by 0.5mm (diameter) Zorbax C18 column with 5µm particles, at a flow rate of 20 µL/min. The LC system was an Agilent 1100 with a capillary pump. Buffer A was water wih (+ ion mode) or without (- ion mode) 0.1% formic acid, and buffer B was acetonitrile with (+ ion mode) or without (- ion mode) 0.1% formic acid. For the endcapped C18 column, the column was equilibrated in 90% A, and the gradient was 10%-98% over 60 min.
Metabolite Database searching.
Metabolites were identified by exact mass (m/z) using METLIN ( Human Metabolome Project ( Lipid Maps ( Biological Magnetic Resonance Data Bank ( and KEGG ligand (
Supplementary References
1. Banin E, Dorrell MI, Aguilar E, Ritter MR, Aderman CM, Smith AC, Friedlander J, Friedlander M. T2-TrpRS inhibits preretinal neovascularization and enhances physiological vascular regrowth in OIR as assessed by a new method of quantification. Invest Ophthalmol Vis Sci. 2006;47:2125-2134
2. Smith CA, Want EJ, O'Maille G, Abagyan R, Siuzdak G. XCMS: processing mass spectrometry data for metabolite profiling using nonlinear peak alignment, matching, and identification. Anal Chem. 2006;78:779-787
1