Department of Regenerative Medicine and Stem Cell Biology, Graduate School of Comprehensive Human Science
Correspondence: Osamu Ohneda, M.D., Ph.D., Department of Regenerative Medicine and Stem Cell Biology, Graduate School of Comprehensive Human Sciences, University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki 305–8575, Japan. Telephone: +81-29-853-2938; Fax: +81-29-853-2938; e-mail: firstname.lastname@example.org
Author contributions: S.F.: conception and design, financial support, collection and/or assembly of data, data analysis and interpretation, manuscript writing, and final approval of manuscript; M.N.: conception and design, administrative support, provision of study material or patients, collection and/or assembly of data, data analysis and interpretation, manuscript writing, and final approval of manuscript; T.Y.: conception and design, provision of study material or patients, collection and/or assembly of data, data analysis and interpretation, and final approval of manuscript; K.K. and I.T.: collection and/or assembly of data, data analysis and interpretation, and final approval of manuscript; G.S.: collection and/or assembly of data, data analysis and interpretation, manuscript writing, and final approval of manuscript; S.U.: provision of study material or patients, collection and/or assembly of data, data analysis and interpretation, and final approval of manuscript; M.K.: provision of study material or patients, collection and/or assembly of data, data analysis and interpretation, and final approval of manuscript; T.K.: data analysis and interpretation, manuscript writing, and final approval of manuscript; T.O.: conception and design, financial support, administrative support, and final approval of manuscript; O.O.: conception and design, financial support, administrative support, provision of study material or patients, data analysis and interpretation, manuscript writing, and final approval of manuscript.
Retinitis pigmentosa is a group of inherited eye disorders that result in profound vision loss with characteristic retinal neuronal degeneration and vasculature attenuation. In a mouse model of retinitis pigmentosa, endothelial progenitor cells (EPC) from bone marrow rescued the vasculature and photoreceptors. However, the mechanisms and cell types underlying these protective effects were uncertain. We divided EPC, which contribute to angiogenesis, into two subpopulations based on their aldehyde dehydrogenase (ALDH) activity and observed that EPC with low ALDH activity (Alde-Low) had greater neuroprotection and vasoprotection capabilities after injection into the eyes of an rd1 mouse model of retinitis pigmentosa compared with EPC with high ALDH activity (Alde-High). Of note, Alde-Low EPC selectively recruited F4/80+/Ly6c+ monocyte-derived macrophages from bone marrow into retina through CCL2 secretion. In addition, the mRNA levels of CCR2, the neurotrophic factors TGF-β1 and IGF-1, and the anti-inflammatory mediator interleukin-10 were higher in migrated F4/80+/Ly6c+ monocyte-derived macrophages as compared with F4/80+/Ly6c− resident retinal microglial cells. These results suggest a novel therapeutic approach using EPC to recruit neuroprotective macrophages that delay the progression of neural degenerative disease. Stem Cells2013;31:2149–2161
Retinitis pigmentosa with an incidence of 1 in 3,500 is a group of inherited eye disorders that result in night blindness, visual field loss, and arteriolar attenuation, and often progress to complete blindness . More than 100 genes responsible for retinitis pigmentosa have been identified and most of these encode for photoreceptor-specific proteins . Gene therapy [3, 4] and injecting neurotrophic factors [5, 6] have been explored as therapeutic options. In recent years, several studies have tested a wide range of cell populations for intravitreal or subretinal transplantation experiments, including retinal and neural progenitor cells (postmitotic rod precursors and embryonic stem cell-derived photoreceptor precursors), mesenchymal stromal cells, and hematopoietic stem cells [7-11]. However, despite these efforts to better understand and treat retinitis pigmentosa, there are still no effective treatments to slow or reverse the progression of this disease [2, 3].
Endothelial progenitor cell (EPC) have been used clinically to treat cardiovascular disease and severe limb ischemia [12-18]. However, the clinical applications of EPC transfusions are limited due to the small numbers of available cells . In a recent study, aldehyde dehydrogenase 1 (ALDH) activity was reported to be an effective marker that specifically identified “functional EPC” . Functional EPCs are those that have high proliferative and migratory capabilities and exhibit a significant capacity to regenerate ischemic tissue [19, 20]. It was shown that EPC with low levels of ALDH expression (Alde-Low EPC) migrated to an ischemic region and decreased necrotic area in a mouse model of skin flap ischemia after making an incision on the dorsal surface . In a rat model of acute cerebral infarction, Alde-Low EPCs accumulated and migrated into the infarct site and subsequently decreased the infarct volume . ALDH activity is an effective marker for functional EPC because Alde-Low EPC are more functional than those with high ALDH activity (Alde-High EPC) .
Inherited retinal degeneration is characterized by retinal neuronal apoptosis and retinal vasculature attenuation. Recently, the therapeutic effects of EPC on central nervous system and peripheral nerves under hypoxic conditions such as in stroke and diabetic neuropathy were investigated [21-23]. However, few reports have assessed the neuroprotective capability of EPC in nonhypoxic environments.
Lineage-negative hematopoietic stem cells that included EPC could rescue retinal blood vessels and photoreceptors in nonischemic neurodegenerative diseases, such as retinitis pigmentosa . However, it was unclear whether the increase in blood flow by revascularization improved neural damage or whether EPC had any neurotrophic effects. Hematopoietic stem cells are multipotent stem cells that give rise to all blood cell types. The lineage commitment of cells arising from hematopoietic stem cells contributed was not confirmed. In this study, purified functional EPC isolated based on their low ALDH activity showed neuroprotective and vasoprotective effects.
In this study, we report for the first time that Alde-Low EPC can induce the selective recruit of a distinct subset of F4/80+/Ly6c+ macrophages with high mRNA expression levels of chemokine C-C motif receptor 2 (CCR2), neurotrophic factors, and anti-inflammatory mediators. These results suggested that intraocular injection of Alde-Low EPC effectively recruited inflammation-resolving and neuroprotective monocyte-derived macrophages that were different from F4/80+/Ly6c− resident microglial cells.
Materials and Methods
C57BL/6J and C3H/HeN (rd1) mice were purchased from Japan SLC, Inc. (Shizuoka, Japan). Mice were treated in accordance with the NIH Guide for the Care and Use of Laboratory Animals and the ARVO Statement for Use of Animals in Ophthalmic and Vision Research. All experimental procedures were approved by the University of Tsukuba.
EPC was isolated from umbilical cord blood and cultured as previously described . In brief, mononuclear cells from human umbilical cord blood were separated by density gradient centrifugation after depleting hematopoietic cells. Then, CD45−/CD31+ cells were sorted and cultured in Iscove's Modified Dulbecco's Medium (IMDM) with 10% fetal bovine serum (FBS) and basic fibroblast growth factor (b-FGF). More than approximately 7 days, adherent cells began to grow and rapidly evolved to form colonies with tightly compact morphologies. Subsequently, DiI-Ac-LDL-positive/CD31+ cells were sorted (DiI-Ac-LDL; Molecular Probes, Eugene, OR, http://probes.invitrogen.com). Culture of sorted cells was continued in IMDM/10% FBS supplemented with b-FGF. These cells displayed endothelial cell (EC)-like morphology.
ALDH activity was analyzed with an Aldefluor reagent (StemCell Technologies, Vancouver, BC, Canada, http://www.stemcell.com). Then, EPCs were divided into two fractions based on their level of ALDH expression: Alde-High EPCs and Alde-Low EPCs . EPC were infected with lenti-GFP (Sigma-Aldrich, St Louis, MO, http://www.sigmaaldrich.com) and green fluorescent protein (GFP)-positive EPC was purified using MoFLo (MoFlo XDP; Dako, Carpinteria, CA). As a control, human umbilical vein endothelial cells (HUVEC) was purchased from Cambrex (Walkersville, MD) and maintained according to the manufacturer's instructions.
Intravitreal and Subretinal Injection of Cells
Cells were injected using a 33-gauge needle (Ito Co. Fuji, Japan, http://www.ito-ex.co.jp). To evaluate vasoprotective and neuroprotective effects, cells were injected at P6. The density of injected cells varied according to the technique used. For intravitreal injections, approximately 1 × 105 cells per 0.5 µL of EPC and HUVEC was delivered. For subretinal injections, 0.5 × 105 cells per 0.2–0.3 µL of EPC and HUVEC, approximately 1 × 104 cells per 0.2–0.3 µL of F4/80+/Ly6c+ cells were used. Immunosuppression was induced with daily intraperitoneal injection of cyclosporin-A (Wako, Osaka, Japan, http://www.wako-chem.co.jp/english) at 20 mg/kg body weight 2 days before the assay, as in previous reports [19, 20].
Flow Cytometry Analysis of Retina and Peripheral Blood Cells
One week after injecting PBS or cells, the total retinal cell population was isolated for flow cytometry analysis (MoFlo XDP; Dako). Blood was collected from the retro-orbital sinus of 6- to 8-week-old wild type (WT) mice 2 days after intravitreal injection of PBS or cells. Mononuclear cells in peripheral blood were isolated using Histopaque density gradient as described previously . The numbers of F4/80+/Ly6c+ cells were evaluated by flow cytometry. F4/80+/Ly6c+ cells isolated from the peripheral blood of WT mice were injected into the subretinal spaces of rd1 mice at P6. The following monoclonal antibodies (mAbs) were used according to the manufacturers' protocols: phycoerythrin (PE) conjugated F4/80 antibodies (1:20, A3-1, AbD Serotec, Raleigh, NC); biotin-conjugated anti-Ly6c antibody (1:50, AL-21, BD Biosciences, San Diego, CA, http://www.bdbiosciences.com), and allophycocyanin (APC) conjugated streptavidin (1:50, BD).
Terminal Deoxynucleotidyl Transferase-Mediated Biotinylated UTP Nick End Labeling Assay
A terminal deoxynucleotidyl transferase-mediated biotinylated UTP nick end labeling (TUNEL) assay kit with anti-fluorescein isothiocyanate (FITC) conjugated horseradish peroxidase (HRP) (TaKaRa, Ohtsu, Japan, http://www.takara.co.jp) was used for cryostat retinal sections (7 µm) of P13 of an rd1 mouse . Three different areas (400 µm in width; unbiased sampling) in a single section that contained the optic nerve head nuclei were counted.
Evaluation of Vasoprotective Ability Effects Based on Measured Lengths of the Retinal Vasculature
Two hundred microliters of tetramethylrhodamine isothiocyanate-conjugated Bandeiraea simplicifolia lectin (0.1 mg/mL; Sigma-Aldrich) was injected through a tail vein, after which a retinal flat mount were prepared. Four randomly chosen independent fields (600 µm × 600 µm) were used to quantify of the vasculature as previously described . The total vasculature length was measured.
Evaluation of Neuroprotective Ability by Measuring the Thickness of Outer Nuclear Layer
Harvested eyes were immersion-fixed in 4% paraformaldehyde (PFA) and embedded in OCT compound (Tissue-Tek; Sakura FineTech Inc.). Cryostat sections (7 µm) were stained with hematoxylin and eosin (H&E). The thickness of outer nuclear layer (ONL) was measured at four points that were decided upon by dividing the retina (from the nerve head to the peripheral retina) evenly into four parts (Fig. 2A).
Retinal Function Recording by Electroretinography
Alde-Low EPCs or PBS were randomly chosen and injected into the intravitreal spaces of six rd1 mice at P6, and electroretinography (ERG) was recorded at P17. ERG was recorded as described previously . Luminance was 1.0 log cd-s/m2 on a rod-desensitizing white adapting background of 1.3 log cd/m2. A total of 150 ERG recordings were averaged using a computer-assisted signal averaging system (Power Lab; AD Instruments, Castle Hill, Australia). The amplitude was measured based on the highest positive wave by a researcher blinded to the experimental conditions used.
Whole eyes were harvested and fixed with 4% PFA. Whole-mount retina or cryostat sections (7 µm) were immunohistochemically stained. To stain microglial cells, anti-F4/80 (1:20, A3-1, AbD Serotec), anti-CD11b (1:50, Abcam, Cambridge, MA, http://www.abcam.com), anti-CX3CR1 (1:50, Abcam), and anti-Ly6C (1:50, ER-MP20, Abcam) antibodies were used, followed by Alexa 488 or 594 conjugated secondary antibodies (1:1,000, Invitrogen, Carlsbad, CA, http://www.invitrogen.com) and isolectin-IB-4 conjugated to Alexa Fluor 594 (1:200, Invitrogen).
Numbers of Migrated Macrophages
Migrated macrophages in the retinas of whole-mount preparations were stained with an anti-F4/80 antibody and counted using a confocal microscope (TCS SP5, Leica, Wetzlar, Germany). Four randomly chosen independent fields (800 µm × 800 µm) were used for quantification.
Systemic Depletion of Circulating Macrophages
Clodronate-loaded liposomes were provided by N. van Rooijen (Vrije Universiteit, Amsterdam, The Netherlands) and were intraperitoneally injected (at an appropriate dilution for each experiment) at a volume of 0.1 mL per day [26, 27].
Enzyme-Linked Immunosorbent Assay for CCL2
To measure paracrine activity, Alde-Low EPC, CCL2 overexpressed Alde-Low EPC (Alde-Low CCL2 EPC), CCL2 downregulated Alde-Low EPC (Alde-Low CCL2kd EPC), Alde-High EPC, CCL2 overexpressed Alde-High EPC (Alde-High CCL2 EPC), and HUVEC were placed in basal medium with 2% serum and avoiding potentially contaminating supplemental growth factors. Samples were collected after 24 hours. The number of adherent cells was also quantified. Results were expressed as nanograms secreted per 105 cells for each paracrine factor over a 24-hour period. The concentration of human CCL2 was determined using human CCL2 enzyme-linked immunosorbent assay (ELISA) kits (R&D Systems, Minneapolis, MN, http://www.rndsystems.com).
Total RNA was extracted and purified using an RNeasy Mini Kit (Qiagen, Valencia, CA, http://www1.qiagen.com). RNA was reverse-transcribed to cDNA using ReverTra-Plus-TM (Toyobo, Osaka, Japan) and then used as template for reverse transcriptase polymerase chain reaction (RT-PCR) analysis. The sequences of the PCR primers (sense and antisense, respectively) were 5′-GAACTTTCTGCTGTCTTGGGTGCATTG-3′ and 5′-CTGCATGGTGATGTTGGACTCCTCAGT-3′ for human vascular endothelial growth factor (VEGF), 5′-CTGACTCACATAGGGTGCAGCAATCAG-3′ and 5′-AGGCTGGTTCCTATCTCCAGCATGGTA-3′ for human Ang-1, 5′-AACTCAGCTAAGGACCCCACTGTTGCT-3′ and 5′-TGAGTAAGCCTCATTCCCTTCCCAGTC-3′ for human Ang-2, 5′-TTGACGGGGTCCGGGAGAAGAGCGACC-3′ and 5′-TTGGAAGAAAAAGTATAGCTTTCTGCCCAGG-3′ for human bFGF, 5′-CTTGATCTAAGGAGGCTGGAGATGT-3′ and 5′-CTTGTTCCTGCACTCCCTCTACTTG-3′ for human insulin-like growth factor type 1 (IGF-1), 5′-GATCCGCTCCTTTGATGATCTCCAACGC-3′ and 5′-ACTGCACGTTGCGGTTGTTGCAGCAGCC-3′ for human platelet-derived growth factor β (PDGF-β), 5′-CTTGTCATGCTGCTCCTCCTG-3′ and 5′-TGCGACTCCTCACATCTCTGC-3′ for human epidermal growth factor (EGF), 5′-CAGCAGAGCACACAAGCTTCTAGGACA-3′ and 5′-GTCCAGACAGAGCTCTCTTCCTCCATCAGAA-3′ for human interleukin (IL)−8, 5′-CCATATTCCTCGGACACCAC-3′ and 5′-TGTACTCCCGAACCCATTTC-3′ for human CCL5, 5′-GGCTGAGACTAACCCAGAAAGATCCAA-3′ and 5′-TGGGTTGTGGAGTGAGTGTTCAAGTCT-3′ for human CCL2, 5′-TCAAGTCTTCGGAGTTTGGGTTTGCTT-3′ and 5′-GTACTTGCGCTCAGGAGGAGCAATGAT-3′ for human β-actin, 5′-GCTCTTACTGACTGGCATGAG-3′ and 5′-CGCAGCTCTAGGAGCATGTG-3′ for mouse IL-10, 5′-AAAATTCGAGTGACAAGCCTGTAG-3′ and 5′-CCCTTGAAGAGAACCTGGGAGTAG-3′ for mouse TNF-α, 5′-TACCATGCCAACTTCTGTCTGGGA-3′ and 5′-TGTGTTGGTTGTAGAGGGCAAGGA-3′ for mouse TGF-β1, 5′-CTACAAAAGCAGCCCGCTCT-3′ 5′-CTTCTGAGTCTTGGGCATGTCA-3′ for mouse IGF-1, 5′-TTGGAACCATCTTCCTGTCC-3′ and 5′-ACGCCCAGACTAATGGTGAC-3′ for mouse CX3CR1, 5′-ATCCACGGCATACTATCAACATC-3′ and 5′-CAAGGCTCACCATCATCGTAG-3′ for mouse CCR2, 5′-GTCGTACCACAGGCATTGTGATGGACT-3′ and 5′-CACCAGACAGCACTGTGTTGGCATAGA-3′ for mouse β-actin. Reactions for β-actin mRNA were performed concurrently in the same plate as those for the test of mRNAs, and results were normalized by the corresponding amount of β-actin mRNA.
Small hairpin RNA (shRNA)
To downregulate CCL2 expression, we used a shRNA MISSION lentiviral transduction system (NM_002982; clone: TRCN0000006279; sequence; 5′-CCGGGATGTGAAACATTATGCCTTACTCGAGTAAGGCATAATGTTTCACATCTTTTT-3′; Sigma-Aldrich) performed as specified in the manufacturer's protocol. cDNA levels were determined by quantitative RT-PCR, and protein levels were determined by ELISA.
CCL2 Overexpression in Alde-Low EPC, Alde-High EPC, and HUVEC
Human CCL2 was overexpressed in Alde-Low EPC, Alde-High EPC, and HUVEC using murine stem cell virus, Phoenix-Eco packaging cells (Clontech Laboratories, Palo Alto, CA, http://www.clontech.com), a FuGENE transfection kit (Roche, Basel, Switzerland, http://www.roche-applied-science.com), and PT67 cells (Clontech Laboratories) as previously described . The expressions of human CCL2 mRNA and protein were confirmed by quantitative RT-PCR and ELISA, respectively.
Results are given as means ± SDs. Results for two groups were compared using Student's t tests. Paired t tests were used to compare the results for the same group at different time points. One-way ANOVA with Bonferroni multiple comparisons was used to compare more than two groups. In the graphs, y-axis error bars indicate SDs. All analyses had been carried out in a masked fashion.
Alde-Low EPC Rescue Retinal Vascular Attenuation
Retinal degeneration in rd1 mice is caused by a loss-of-function mutation in the β-subunit of the rod photoreceptor cGMP-phosphodiesterase gene [29, 30]. Rod photoreceptor death and disappearance of the deep retinal vasculature is complete by P21. We investigated whether transplanted EPC could prevent this vascular attenuation. Alde-Low EPC were intravitreally injected into one eye, and PBS was injected into the contralateral eye at P6. Alde-Low EPC promoted the development of the deep vascular layer, and prevented vascular attenuation, as compared with PBS injection (Fig. 1A, 1B).
Next, Alde-Low EPC, Alde-High EPC, HUVEC, or PBS were intravitreally  and subretinally [7, 31] injected into rd1 mouse eyes at P6. Injecting Alde-Low EPC resulted in a pronounced rescue of the deep vascular layer as compared with the control (Fig. 1C–1F). At 1 week after intravitreal or subretinal injection, a large number of Alde-Low EPC with GFP survived on the retinal surface, and in the intravitreal space or subretinal space (Fig. 1E). However, Alde-Low EPC were not incorporated in any vessels of the three retinal layers (data not shown). After longer time observation, the number of Alde-Low EPC gradually decreased and had completely disappeared by 6 weeks after injection.
Neuroprotective Effect of Alde-Low EPC
To determine whether Alde-Low EPC could protect the retina from neural degeneration, Alde-Low EPC were intravitreally injected into rd1 mice at P6. After the P13 stage, this injection had retarded the degenerative process as compared with PBS injection (Fig. 2A, 2B).
Alde-Low EPC showed some functional rescue of the retina at P17 based on ERG, although the signal amplitude was considerably lower than normal, besides they also presented an abnormal pattern (Fig. 2C). These signals needed to be averaged using a strong flash for more than 100 times to detect these waveforms because their amplitudes were too small to be recorded. Because some ERG signals had an a-wave unlike the others, only b-waves were measured. Further analyses found that a greater number of photoreceptors remained in the ONL at P17 when only Alde-Low EPC were injected (Fig. 2D). In addition, using a TUNEL assay [24, 30, 32] there were fewer apoptotic cells in the ONLs of mice that were injected with Alde-Low EPC than in other mice (Fig. 2E, 2F). This indicated that Alde-Low EPC had a neuroprotective effect in the retina.
To eliminate the influence of the blood-retina barrier, Alde-Low EPC were directly injected into the subretinal space, a site closer to the degenerated retina. When Alde-Low EPC were injected into this subretinal site, their neuroprotective effect was localized around the injection site and no direct integration into the ONL was apparent (Fig. 2G, 2H). There was a tendency for a greater ONL thickness after injecting Alde-Low EPC subretinally as compared with intravitreally, although this difference was not significant (Fig. 2H). There was no recovery of degenerated photoreceptors after intravitreal or subretinal injection of Alde-Low EPCs at later stages (data not shown).
Growth Factor and Chemokine Expressions by EPC
To investigate which growth factors [21-23] or chemokines  might have functional roles in these protective effects, we determined the mRNA levels of candidate factors in Alde-Low EPC and Alde-High EPC (Fig. 3A). Quantitative RT-PCR results showed that Ang-2, PDGF-β, IGF-1, EGF, and CCL2 mRNAs were expressed at significantly higher levels in Alde-Low EPC as compared with Alde-High EPC. In agreement with previous reports that showed that CCL2, also known as monocyte chemotactic protein-1, recruited monocytes and macrophages, a greater number of macrophages was observed in the retinas injected with Alde-Low EPC (Fig. 3B, 3C).
Characterization of Migrated Macrophages
In a WT mouse, resident microglial cells located in the ganglion cell layer and in the inner plexiform layer had dendritic morphologies (Fig. 3E). When retinal degeneration occurs, resident and resting microglial cells migrate to the ONL and phagocytose apoptotic photoreceptors . In this study, microglial cells were observed in the ONL of an rd1 mouse without injecting cells or PBS and these microglial cells might have been involved in the phagocytosis of apoptotic photoreceptors, as in a previous study . The number of F4/80 positive cells (a marker for macrophages/microglial cells) was the highest after injecting Alde-Low EPC (Fig. 3B). F4/80-positive cells were observed in the ONL as well as on the surface and in the inner retina after injecting Alde-Low EPC (Fig. 3D). Confocal microscope images of a retinal flat mount clearly showed that recruited F4/80-positive cells had amoeboid-like shapes, which was different from the shapes of resident and resting microglial cells (Fig. 3B).
CCL2 Secreted by Alde-Low EPCs Induces Macrophage Migration and Prevents Vascular Attenuation and Neural Degeneration
To investigate whether CCL2 played an important role, we used shRNA to reduce CCL2 expression in Alde-Low EPC (Alde-Low CCL2kd EPC; Fig. 4A, 4B). When Alde-Low CCL2kd EPC were injected intravitreally, the numbers of migrated macrophages decreased dramatically and resulted in photoreceptor degeneration and a failure to rescue vascular attenuation (Fig. 4D, 4F–4H).
To determine how CCL2 was involved in the inhibiting vascular attenuation and neural degeneration, we overexpressed CCL2 in Alde-High EPC and HUVEC (Alde-High CCL2 EPC and HUVEC-CCL2 cells; Fig. 4A–4C). After intravitreal injection of Alde-High CCL2 EPC, the number of migrated macrophages was greater than after injecting Alde-High EPC (Fig. 4D, 4H). Although vascular attenuation and neural degeneration were retarded as compared with injecting Alde-High EPC alone, the same levels of protection were not observed as when injecting Alde-Low EPC (Fig. 4F, 4G, 4H). This suggested that Alde-Low EPC-derived factors other than CCL2 might have vascular and neuronal protective effects. In contrast to the results of Alde-High CCL2 EPC injection, the number of macrophages did not increase after transplanting HUVEC-CCL2 cells. Vascular attenuation and neural degeneration were not recovered by HUVEC-CCL2 cells, which suggested that CCL2 alone was not sufficient to attract those macrophages that might contribute to delaying retinal degeneration (Fig. 4E–4H).
Macrophage Migration Induced by Alde-Low EPCs Is Involved in Protection Against Vascular Attenuation and Neural Degeneration
To investigate how macrophages contributed to these protective effects in the retina, clodronate liposomes were injected to eliminate migrated macrophages . Peritoneal injection of clodronate liposomes completely abolished macrophages' migration (Fig. 4E, 4H) and also disrupted the vasoprotective and neuroprotective effects (Fig. 4E–4G). These results indicated that macrophages migrating to the retina played an important role in vasoprotection and neuroprotection rather than any direct effects by Alde-Low EPC. In contrast to injecting Alde-Low EPC, injecting CCL2-transfected Alde-Low EPC (Alde-Low CCL2 EPC) failed to promote any increase in the numbers of migrated macrophages and did not significantly change retinal vascular attenuation and photoreceptor degradation. This suggested that the number of migrated macrophages due to CCL2 was limited, and that the protective effects may have depended on the numbers of migrated macrophages. Taken together, these results indicated that migrated macrophages, recruited by Alde-Low EPCs via their secretion of CCL2 and other unknown factors, played an important role in delaying the progression of retinal degeneration.
Migrated Macrophages Are Different from Resident Retina Microglial Cells
To further investigate migrated macrophages, F4/80-positive cells were stained with other macrophage markers, CD11b and Isolectin-IB4. Macrophages were also stained with CX3CR1 (a myeloid lineage marker) [34-36] and Ly6c (a marker of bone marrow [BM]-derived monocytes)  after the injecting Alde-Low EPC (Fig. 5A). Migrated Ly6c-positive cells integrated not only into the ONL but also onto the surface and into the inner retinal layer (Fig. 5B). In contrast, resident microglial cells in the ONL were negative for Ly6c.
Intraperitoneal injection of clodronate liposomes clearly showed that these liposomes affected the recruitment of F4/80+/Ly6c+ macrophages; however, this effect was not observed for F4/80+/Ly6c− cells (Fig. 5C, 5D). F4/80+/Ly6c− cells were resident retinal microglial cells and had migrated to the ONL to phagocytose apoptotic photoreceptors [27, 38]. Thus, we speculated that these F4/80+/Ly6c+ cells were migrating monocyte-derived macrophages. We then characterized these macrophages by analyzing their mRNA expression.
Anti-inflammatory mediator IL-10 mRNA expression was found in migrated monocyte-derived macrophages (Ly6c+), whereas resident retinal microglial cells (Ly6c−) expressed inflammatory mediator TNF-α mRNA. The mRNA levels of growth factors, including TGF-β1 and IGF-1, were higher in migrated monocyte-derived macrophages (Ly6c+) than in resident retinal microglial cells (Ly6c−) (Fig. 5E, 5F). Of note, the mRNA level for CCR2, a chemokine receptor for CCL2, was higher in migrated monocyte-derived macrophages (Ly6c+) than in resident retinal microglial cells (Ly6c−). This suggested that transplanted Alde-Low EPCs secreted the chemokine CCL2 to attract monocyte-derived macrophages to the retina.
Monocyte-Derived Macrophages in Peripheral Blood Contribute to the Neuroprotective and Vasoprotective Effects
Mononuclear cells in the peripheral blood were isolated after intravitreal injection of PBS or cells. Flow cytometry analysis showed that there was a higher frequency of F4/80+/Ly6c+ cells in the peripheral blood of Alde-Low EPC injected mice as compared with other mice (Fig. 6A, 6B). This indicated that Alde-Low EPC induced the migration of F4/80+/Ly6c+ cells from the BM into the peripheral blood.
We then injected 1 × 104 F4/80+/Ly6c+ cells isolated from peripheral blood into the subretinal spaces of rd1 mice at P6 and analyzed the number of rescued cells. There were more rescued cells in the ONL as compared with subretinal injection of PBS. However, the neuroprotective effects were localized to the injection site (Fig. 6C–6E). Interestingly, injected F4/80+/IB4+ monocyte-derived macrophages remained at the rescued region and expressed IL-10 and IGF-1 mRNAs (Fig. 6F). Finally, subretinal injection of F4/80+/Ly6c+ cells clearly rescued the deep vascular layer construct as compared with the injecting PBS (Fig. 6G, 6H). However, this F4/80−Ly6C+ population showed no neuroprotective or vascular protective capabilities after injection. The F4/80−Ly6C+ population comprised neutrophils, ECs, or T-cell subsets based on flow cytometry analysis (Supporting Information Fig. 1).
In this study, we showed that Alde-Low EPC effectively recruited monocyte-derived macrophages by secreting CCL2 and these recruited cells exerted vasoprotective and neuroprotective effects in the retina. This remarkable vascular protective capability might be useful in neurodegenerative as well as in neural ischemic diseases. Alde-Low EPCs were effective in acute ischemic brain injury . Alde-Low EPCs also showed neuroprotective effects equivalent to that observed after injecting neurotrophic factors, such as ciliary neurotrophic factor (CNTF) (Supporting Information Fig. 2). CNTF is the only known factor with evident neuroprotective effects in the rd1 mouse . Thus, intravitreal CNTF injection was used as a positive control for neuroprotection.
However, in previous and the current studies, injected neurotrophic factor was used at a very high concentration, which would be difficult to adjust for actual clinical use  (Supporting Information Fig. 2, Supporting Information Methods). Moreover, injected Alde-Low did not produce any complications, such as retinal detachment, abnormal neovascularization, hemorrhage, or strong immune reactions.
A hierarchy of EPC based on their proliferative potential has been reported . The concepts of what constitutes EPCs have become increasingly complex and confusing and the definition of an EPC has been controversial. Thus, the methods used to isolate EPCs are highly variable among investigators . Given the therapeutic potential of EPCs, efficient isolation of highly proliferative EPCs is central to the generation of a reliable and safe cell-based therapy. We identified highly proliferative EPC based on their ALDH activity, which we referred to as “Alde-Low EPC.” By flow cytometry analysis, the frequency of Alde-Low EPC in umbilical cord blood was [mt]35% [12, 19]. It has also been reported that transplanted EPC not only induced the recruitment of monocytes/macrophages and promoted neovascularization but also accelerated dermal wound healing [15, 19, 40, 41].
Importantly, injecting CCL2-transfected HUVEC did not affect the migration of monocytes/macrophages and showed no vasoprotective/neuroprotective effects in the retina (Fig. 4E), which suggested that CCL2 alone was not sufficient for inducing the migration of monocytes/macrophages to the retina. In contrast, injecting CCL2-transfected Alde-High EPC induced an increased number of migrated monocytes/macrophages (Fig. 4D) which, in turn, had vasoprotective/neuroprotective effects in the retina. However, the protective effects achieved after transplanting CCL2-transfected Alde-High EPC were not as good as that achieved when transplanting Alde-Low EPC (Fig. 4F, 4G), which suggested that some unknown factors might be involved in providing effective protection. Transplanting CCL2-transfected Alde-Low EPCs did not increase the numbers of migrated monocytes/macrophages as compared with injecting Alde-Low EPC alone (Fig. 4D), possibly because the amount of secreted CCL2 needed for migration may have become saturated.
We also found that Alde-Low EPC recruited monocytes/macrophages (Fig. 3B) and resulted in the early recovery from vessel occlusion . Altogether, these findings suggest that EPC may be effective in two distinct ways: in a direct manner in which EPCs themselves are incorporated into neovessels and in an indirect manner in which EPC induce the migration of monocytes/macrophages to impaired sites. We found that Alde-Low EPC transplanted into the retina operated in the indirect rather than the direct manner.
Clodronate liposomes reduce the recruitment of BM-derived macrophages into the peripheral blood but do not reduce the numbers of tissue-resident microglial cells [26, 27]. Clodronate liposomes completely abolished the vasoprotective/neuroprotective effects of Alde-Low EPC transplantation (Fig. 4E), which suggested that migrated monocytes/macrophages from the BM were primarily responsible for the protection against retinal degeneration. It has been reported that EPCs are different from CD45−/CD14+ “colony forming unit-ECs” that differentiate into macrophages [19, 42]. Therefore, it was unlikely that transplanted EPC had differentiated into macrophages at the transplantation site.
Resident monocytes/macropahges play significant pathogenic roles in inflammatory diseases [43-46], including retinitis pigmentosa . However, these reports only investigated resident monocytes/macropahges and did not assess circulating monocytes or macrophages derived from the BM. Circulating monocytes/macrophages can be categorized into several subsets and there are two major monocyte/macrophage subsets: Ly-6c+/CX3CR1lo/CCR2+ cells, which infiltrate injured tissues; and Ly-6c−/CX3CR1hi/CCR2− cells, which primarily contribute to tissue-resident macrophages . In this study, migrated monocyte-derived macrophages (F4/80+/Ly6c+ cells) expressed IGF-1, one of the endogenous neurotrophic factors, to a greater extent than resident microglial cells (F4/80+/Ly6c− cells; Fig. 5E, 5F). Ly6c is expressed on monocytes in the BM and shortly after their migration into the circulation, which suggests that Ly6c might be a useful marker for isolating protective monocytes/macrophages from the peripheral blood (Fig. 7E).
CCL2 and CCR2 are important for recruiting Ly6c+ monocytes from the BM, as CCL2 and CCR2 deficiencies have been shown to disrupt Ly6c+ monocytes entry into the blood or inflammatory tissues [47, 48]. CCL2/CCR2 signaling is also crucial for upregulating IGF-1 mRNA expression in macrophages. Recently, disrupted recruitment of CCR2+/Ly6chi/CD11c+ monocytes was implicated as a mechanism underlying poor recovery in a model of spinal cord injury. Injecting large numbers of Ly6chi monocytes into the circulation enhanced recovery after spinal cord injury.
These monocytes express IL-10 and have an anti-inflammatory role , which suggests that migrating monocyte-derived macrophages that express IL-10 could also have an anti-inflammatory role. Abolishing the endogenous monocyte pool results in a reduced number of transcripts that encode for anti-inflammatory mediators, such as TGF-β1/2 and IL-10, and an increase in proinflammatory mediators . As compared with resident microglial cells, migrated monocyte-derived macrophages had higher mRNA expression levels for anti-inflammatory mediators and lower mRNA expression levels for proinflammatory factors (Fig. 5E, 5F). In retinitis pigmentosa, TNF-α expression is induced before the onset of photoreceptor cell death during retinal degeneration . Taken together, these findings indicate that migrated monocyte-derived cells are inflammation-resolving macrophages in the retina (Fig. 7).
In this study, Cyclosporine-A was used for immunosuppression, and we observed that Cyclosporine-A had no effect on retinal degeneration in rd1 mice neither on protection nor on getting worse. Cyclosporine-A exerts its immunomodulatory activities by its effects on T lymphocytes. However, previous studies reported that Cyclosporin-A had effects on macrophage, such as decreased IL-6 production . Thus, Cyclosporine-A could possibly have had some effects on microglial cells or macrophages.
In conclusion, intraocular injection of Alde-Low EPC effectively recruited inflammation-resolving and neurovascular protective monocyte-derived macrophages that delay the progression of neural degenerative disease.
This work was supported by Grants-in-Aid for Scientific Research from the Japanese Ministry of Education, Culture, Sports, Science, and Technology, Grant-in-Aid for JSPS Fellows, and a Research Grant from the Study Group on Chorioretinal Degeneration and Optic Atrophy, the Ministry of Health, Labor and Welfare, Japan. This article was presented in part at Association for Research in Vision and Ophthalmology Annual Meeting, May 2011 and 2012, Fort Lauderdale, FL.
Disclosure of Potential Conflicts of Interest
The authors indicate no potential conflicts of interest.