Immune stimulation recruits a subset of pro-regenerative macrophages to the retina that promotes axonal regrowth of injured neurons

Lien Andries; Daliya Kancheva; Luca Masin; Isabelle Scheyltjens; Hannah Van Hove; Karen De Vlaminck; Steven Bergmans; Marie Claes; Lies De Groef; Lieve Moons; Kiavash Movahedi

doi:10.1186/s40478-023-01580-3

Immune stimulation recruits a subset of pro-regenerative macrophages to the retina that promotes axonal regrowth of injured neurons

Andries, Lien; Kancheva, Daliya; Masin, Luca; Scheyltjens, Isabelle; Van Hove, Hannah; De Vlaminck, Karen; Bergmans, Steven; Claes, Marie; De Groef, Lies; Moons, Lieve; Movahedi, Kiavash 2023-05-24 00:00:00 The multifaceted nature of neuroinflammation is highlighted by its ability to both aggravate and promote neuronal health. While in mammals retinal ganglion cells (RGCs) are unable to regenerate following injury, acute inflammation can induce axonal regrowth. However, the nature of the cells, cellular states and signalling pathways that drive this inflammation-induced regeneration have remained elusive. Here, we investigated the functional significance of mac- rophages during RGC de- and regeneration, by characterizing the inflammatory cascade evoked by optic nerve crush (ONC) injury, with or without local inflammatory stimulation in the vitreous. By combining single-cell RNA sequenc- ing and fate mapping approaches, we elucidated the response of retinal microglia and recruited monocyte-derived macrophages (MDMs) to RGC injury. Importantly, inflammatory stimulation recruited large numbers of MDMs to the retina, which exhibited long-term engraftment and promoted axonal regrowth. Ligand-receptor analysis highlighted a subset of recruited macrophages that exhibited expression of pro-regenerative secreted factors, which were able to promote axon regrowth via paracrine signalling. Our work reveals how inflammation may promote CNS regenera- tion by modulating innate immune responses, providing a rationale for macrophage-centred strategies for driving neuronal repair following injury and disease. Lieve Moons and Kiavash Movahedi are co-senior author. *Correspondence: Lieve Moons lieve.moons@kuleuven.be Kiavash Movahedi kiavash.movahedi@vub.be Full list of author information is available at the end of the article © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 2 of 20 contribute to CNS de- and regeneration is the retina- Introduction brain connection. Over the past decades, multiple stud- The central nervous system (CNS) accommodates vari - ies, using the retinofugal pathway and the optic nerve ous populations of resident macrophages that are criti- crush (ONC) paradigm as a neurodegeneration model, cal regulators in brain development, homeostasis and have shown that induction of controlled inflamma - disease [1]. This includes microglia in the brain paren - tion in the retina induces retinal ganglion cell (RGC) chyma and border-associated macrophages (BAMs) in survival and axonal regrowth in rodents [26–32]. Pre- non-parenchymal border tissues. Microglia continuously vious research revealed that an inflammatory stimula - survey their microenvironment and interact with neu- tion results in infiltration of neutrophils and MDMs in rons to prune synapses, provide neurotrophic factors, both the vitreous and retina and leads to modulation of remove waste and sense danger [1]. Similarly, BAMs resident macrophages, as well as retinal astrocytes and play key roles in supporting healthy brain functions [2]. Müller glia [33, 34]. While it is established that an acute uTh s, resident CNS macrophages are highly special - inflammatory response is beneficial for survival and ized cells that play an active role in maintaining healthy axonal regrowth of damaged RGCs, controversy still brain physiology. Upon inflammation and disease, micro - prevails about which cells, cell states, molecules and glia and BAMs exit their homeostatic state and adopt pathways are functionally implicated. To investigate the new transcriptional modules. This has been thoroughly specific contribution of resident and recruited myeloid investigated for microglia during neurodegeneration cells during RGC de- and regeneration, we performed and subsequently in other disease and injury models, an in-depth characterization of the acute inflamma - where a specific disease-associated microglia (DAM) tory response evoked by optic nerve injury, with or state has been identified [3–12]. The disease-responses without a local inflammatory stimulation using a Toll- of microglia and BAMs can be beneficial or detrimen - like receptor 2 agonist. By combining single-cell RNA tal depending on the nature and/or stage of the disease. sequencing (scRNA-seq) and fate mapping approaches, For example, in Alzheimer’s disease DAMs may allevi- we elucidated the ontogeny, cell states and functional ate amyloid beta (Ab) pathology by compacting amyloid significance of the resident and recruited macrophage plaques but may also aggravate the disease following populations that react to RGC degeneration following the onset of tau pathology [13]. Importantly, inflamma - ONC injury. Our results show that inflammatory stim - tion and disease can also result in the recruitment of ulation recruits a subset of pro-regenerative MDMs to monocyte-derived macrophages (MDMs) to the CNS. the retina, which produce secreted proteins that can Emerging evidence indicates that recruited MDMs react promote axon regrowth of injured RGCs. differently to disease than resident macrophages [14–16]. Recruited MDMs may exert complementary functions Results and their interplay with resident brain macrophages can Injury to the optic nerve activates resident and recruited shape disease progression [15]. Therefore, inhibiting or myeloid cells in the retina promoting monocyte recruitment to the diseased brain To investigate the response of myeloid cells in the ret- may have significant therapeutic implications. ina to retinal ganglion cell (RGC) injury, we performed While the key role of brain macrophages in neurode- + + scRNA-seq on CD45 CD11b cells that were sorted generation is firmly established, their contribution to from healthy adult retinas or from retinas harvested at regeneration and repair has remained more elusive. Sev- 4 days post optic nerve crush (ONC) injury. The majority eral studies have reported that brain macrophages can + + of CD45 CD11b cells in healthy retinas were microglia, exert neuroprotective activities and contribute to healing identified based on their high expression of microglial and repair following neurodegeneration or CNS injury signature genes, including Sall1, Sparc and P2ry12 [17–20]. Different macrophage activation states, rang - (Fig. 1A, B). Naive retinas also contained small clusters of ing from more pro-inflammatory ones that are associ - + + Fcgr1 C1qa macrophages that did not express prototyp- ated with tissue damage, neuronal loss, axon retraction ical microglia genes, but exhibited enriched expression and demyelination to more anti-inflammatory pheno - of Ms4a7, Ms4a6c and Apoe (clusters BAM1-2) (Fig. 1A, types that are linked to neuroprotection [21] and axon B). Within the brain, this signature is associated with regrowth [22, 23], have been suggested to affect repair macrophages found in border tissues (border-associated after CNS injury [24, 25]. Nevertheless, while neuro- macrophages or BAMs), including the perivascular space protective and regenerative macrophage subtypes are [10, 35]. Therefore, these cells may represent retinal thought to exist, their molecular fingerprint remains BAMs, such as perivascular macrophages. We observed poorly characterized. two main clusters that showed differential expression One powerful model system to investigate the cellular of Mrc1, Cd163 and H2-Aa (Fig. 1B), reflecting BAM players and the molecules and signalling pathways that A ndries et al. Acta Neuropathologica Communications (2023) 11:85 Page 3 of 20 neurodegeneration. Mg3 represented a cluster of micro- heterogeneity in the brain [10]. Other C D11b cells in glia that expressed interferon (IFN)-induced genes (Addi- the naive retina were neutrophils (S100a9, Csf3r, Cd24a), tional File 1: Figure S1C) and was observed both in the classical monocytes (Ly6c2, Fn1), non-classical mono- naive and injured retina. These IFN response microglia cytes (Ear2, Ace), cDC2 dendritic cells (DCs) (Flt3, Ciita), are also observed in the healthy brain [41]. We also iden- migratory DCs (Flt3, Ccr7) and natural killer (NK) cells tified Mg4 as a cluster of proliferating microglia (Addi - (Klrb1c, Ncr1) (Fig. 1A, B). tional File 1: Figure S1D), which was most prominent in The immune composition in the retina was clearly the post ONC retina (Fig. 1C). The density and activa - altered in mice that underwent an ONC, with the latter tion of microglia in the different layers of the retina (e.g. showing an increase of peripheral myeloid cells (Fig. 1C). inner and outer plexiform layer) after ONC were further This was most notable for neutrophils, which showed an investigated via whole mount retinal staining. Confocal elevated infiltration in the retina at day 4 post ONC. We Z-stack images of the inner and outer plexiform layer also observed a large cluster of cells that expressed both revealed that microglia shifted from highly ramified cells macrophage and neutrophil markers (Additional File 1: to bigger more amoeboid cells with retracted processes, Figure S1A, B). These cells may represent macrophage- indicative for their reactive state (Fig. 1E, Additional File neutrophil aggregates that formed during in vitro single- 1: Figure S1E) [42]. cell processing. Alternatively, they may correspond to hi Cluster MDM1 represented a subset of Gpnmb Fab- microglia or macrophages that phagocytosed apoptotic hi neutrophils in vivo prior to retinal dissection and pro p5 macrophages that clustered distal from microglia and BAMs (Fig. 1A, B, Additional File 1: Figure S1F) cessing. For reasons of clarity, these cells were excluded and was restricted to the post ONC retina (Fig. 1C). Dif- from all further analyses. ferential gene expression analysis showed an absence Clusters Mg1–Mg4 expressed microglial signature of microglial signature genes in these cells, coupled to genes, including Sall1, which is reported to be restricted a high expression of genes associated with monocyte- to embryonically-derived microglia [36–38]. Addi- derived macrophages (MDMs), including Ms4a7, Lyz2 tionally, Mg1–Mg4 did not express genes related to and Clec12a (Fig. 1F). Therefore, this cluster may repre - BAMs or recruited monocyte-derived cells (e.g. Ms4a7, sent MDMs that were recruited to the retina following Clec12a) [36–38] and were therefore identified as micro - ONC. In line with this, we observed that microglia from glia. Interestingly, most retinal microglia from mice that the injured retina exhibited an elevated expression of the underwent ONC, clustered separately from their coun- monocyte chemoattractant Ccl2 (Fig. 1D). Furthermore, terparts observed in control mice (Fig. 1C), indicat- quantification of monocyte and macrophage infiltra - ing a change in microglial activation status. Microglia tion in the retina at days 2, 4, 6 and 8 post ONC via flow from ONC mice were mostly confined to cluster Mg2 cytometry, revealed a transient increase in the number of that, compared to Mg1, exhibited a downregulation of monocytes, while macrophage numbers peaked at later homeostatic microglial signature genes (P2ry12, Sall1, time points (Fig. 1G). A significant increase in the num - Tmem119) and an induction of prototypical disease- ber of monocytes was also observed in the injured optic associated microglia (DAM) markers (Cst7, Lpl, Fabp5) nerve (Fig. 1G). This shows that monocytes are attracted (Fig. 1D) [3, 39, 40]. This shows that the majority of reti - to both the retina and optic nerve following ONC nal microglia reacted to RGC axonal injury and that they injury, which is in line with MDM1 representing newly- exhibited gene expression changes that are compara- recruited monocyte-derived cells. Interestingly, the ble to those observed for brain microglia responding to (See figure on next page.) Fig. 1 Injury to the optic nerve activates resident and recruited myeloid cells in the retina. A UMAP and cluster annotation showing 7128 + + CD45 CD11b cells isolated from healthy retinas (naive mice) or injured retinas (mice receiving ONC). BAM border associated macrophage, cDC conventional dendritic cell, migDC migratory dendritic cell, MDM monocyte-derived macrophage, Mg microglia, MO monocyte, N neutrophils, NK natural killer cell. Data originate from retinas pooled from 32 naive mice and 10 ONC mice. B Dot plot corresponding to UMAP in (A), showing expression of the indicated genes. Dot size represents the percentage of cells expressing the gene and colour represents its average expression within a cell cluster. C UMAP showing 2819 cells of healthy retinas and 4309 cells of injured retinas with a pie chart showing the distribution of different immune cell populations present in the healthy and injured retinas. Numbers in the pie chart represent percentages of the cell subsets. D Volcano plot displaying differential expression between Mg2 (injured microglia) and Mg1 (healthy microglia). Genes with adjusted p-value < 0.01 and I Log2(FC) I > 1 are shown in red. E Retinal wholemounts stained for IBA1 (green) labelling the microglia in the retina of naive mice and mice at 4 dpi ONC. Scale bar 50 µm and 25 µm. Representative images of n = 3–4 mice per condition. F Volcano plot displaying differential expression between MDM1 and Mg2. Genes with adjusted p-value < 0.01 and I Log2(FC) I > 1 are shown in red. G Cell counts of monocytes and macrophages at different time points after ONC in the retina and optic nerve, as measured by flow cytometry. Data are shown as mean ± SEM. Repeated measures one-way ANOVA followed by Tukey’s multiple comparisons test, n = 3–6 biologically independent samples with retinas from 4 mice pooled per sample. * P < 0.05 Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 4 of 20 Fig. 1 (See legend on previous page.) A ndries et al. Acta Neuropathologica Communications (2023) 11:85 Page 5 of 20 MDM1 cluster exhibited a high expression of Apoe, Spp1 significantly different between the ONC and ONC + P3C and Igf1 (Additional File 1: Figure S1F), which have been conditions (Fig. 2C). This indicates that monocyte influx reported to be key genes involved in promoting RGC following intravitreal P3C injection is restricted to the regeneration following axonal injury [43]. This suggests retina. To spatially localize the infiltrating monocyte- that recruited MDMs can attain activation states that derived cells within the retina and vitreous, an immu- may promote RGC survival and/or axon regeneration. nostaining for IBA1 was performed on retinal sections However, as no axonal outgrowth is observed follow- of Lyz2-GFP mice subjected to ONC or ONC + P3C ing ONC, this response is potentially insufficient, or the (Fig. 2D). As shown in our scRNA-seq data, Lyz2 is number of recruited cells too low to induce regeneration. highly expressed in monocytes and monocyte-derived + + cells (Additional File 1: Figure S1F). IBA1 GFP cells Inflammatory stimulation upon ONC recruits thus likely represent monocytes and MDMs, although monocyte‑derived cells that exhibit long‑term we cannot rule out that a fraction of microglia may also engraftment in the retina and promote axonal upregulate Lyz2 following ONC and P3C treatment. P3C + + regeneration treatment induced a strong infiltration of IBA1 GFP As we observed the expression of potential pro-regener- cells that at day 2 were mostly observed in the vitre- ative genes in MDMs, we hypothesized that the reported ous and showed a round shape, indicative of monocytes regeneration of RGC axons following inflammatory stim - (Fig. 2D). Over time, these cells gradually changed their ulation may in part be driven by an increased recruitment morphology, adopting a more macrophage-like shape, of monocyte-derived cells in the retina. To investigate and infiltrated the ganglion cell layer, inner plexiform this, we combined ONC injury with intravitreal injection layer, inner nuclear layer and the outer plexiform layer of of the Toll-like receptor 2 agonist Pam3Cys combined the retina (Fig. 2D). with cAMP (P3C) and confirmed that this type of inflam - The increased number of macrophages observed in matory stimulation induced axonal regrowth of the dam- P3C treated mice may be driven by an increased recruit- aged RGCs (Additional File 2: Figure S2A, B) [28]. ment of monocytes and/or by an expansion of resident To assess the kinetics of peripheral myeloid cell infil - microglia/BAMs. To distinguish between these possibili- CreER tration, we performed flow cytometric analysis of the ties, we performed fate mapping using the Cx3cr1 : retina and optic nerve from C57BL/6 mice at days 2, 4, 6 R26-YFP model. Upon tamoxifen treatment in Cx3cr- CreER and 8 post ONC or ONC + P3C. P3C treatment induced 1 : R26-YFP mice, 95 ± 1% of retinal macrophages a large increase in the number of recruited neutrophils were YFP labelled (Fig. 3A, B). Four weeks following and monocytes in the retina, representing a ~ 360-fold tamoxifen administration, when YFP labelling in mono- and ~ 180-fold increase at day 2, respectively (Fig. 2A, B). cytes is lost [44], mice received ONC + P3C treatment. Neutrophil infiltration was very transient, with most cells Retinas were processed for flow cytometry at various disappearing by day 8 post treatment (Fig. 2B). Monocyte time points, ranging from 2 to 168 days post treatment. recruitment was similarly transient, but these cells gradu- We were able to distinguish resident microglia/BAMs ally differentiated into macrophages, as reflected by a loss from recruited MDMs based on their differential YFP of Ly6C and an increase in CX3CR1 expression (Fig. 2A). expression (Fig. 3A). This confirmed that the strong Coupled to this we observed a strong increase in the increase in retinal macrophages upon P3C treatment was number of retinal macrophages, which peaked around driven by the recruitment and differentiation of mono - day 6 post treatment (Fig. 2B). In contrast, the number of cytes (Fig. 3B, C). While the number of recruited mac- monocytes and macrophages in the optic nerve was not rophages progressively decreased, a substantial fraction (See figure on next page.) Fig. 2 Inflammatory stimulation upon optic nerve injury mobilizes the infiltration of monocyte-derived cells. A Representative flow cytometry plots low high high − high low showing the gating strategy used to identify neutrophils (Cx3cr1, Ly6G ), monocytes (Ly6C, Ly6G ) and macrophages (Cx3cr1, Ly6C ) in the retina and optic nerve of mice subjected to ONC and ONC combined with P3C treatment. Example plots were taken from retinas of mice at 2 dpi ONC and at 2 dpi ONC + P3C. Cells were pre-gated as live, single CD45 cells. B Cell counts of neutrophils, monocytes and macrophages at different time points after ONC (black) and ONC + P3C (coloured) in the retina, as measured by flow cytometry. Counts of the monocytes and macrophages after ONC correspond to the data shown in Fig. 1G. Data are shown as mean ± SEM. Statistical significance between ONC and ONC + P3C was evaluated via an unpaired t-test. n = 3–6 biologically independent samples with retinas from 4 mice pooled per sample. * P < 0.05 ** P < 0.01 C Cell counts of neutrophils, monocytes and macrophages at different time points after ONC (black) and ONC + P3C (coloured) in the optic nerve, as measured by flow cytometry. Counts of the monocytes and macrophages after ONC correspond to the data shown in Fig. 1G. Statistical significance between ONC and ONC + P3C was evaluated via an unpaired t-test. n = 3–6 biologically independent samples with optic nerves from 4 mice pooled per sample. D Retinal cryosections of Lyz2-GFP (green) mice stained for IBA1 (red). Sections are counterstained with DAPI (blue). Scale bar 50 µm. GCL ganglion cell layer, IPL inner plexiform layer, INL inner nuclear layer, OPL outer plexiform layer, ONL outer nuclear layer. Representative images of n = 3 mice per condition Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 6 of 20 Fig. 2 (See legend on previous page.) A ndries et al. Acta Neuropathologica Communications (2023) 11:85 Page 7 of 20 Fig. 3 The infiltrating monocyte-derived cells exhibit a long-term engraftment in the retina. A Representative flow cytometry plots showing + − the gating strategy for distinguishing resident macrophages (YFP ) from recruited monocyte-derived counterparts (YFP ) based on YFP CreER expression in retinas of Cx3cr1 :R26-YFP mice at various timepoints ranging from 2 to 168 dpi ONC + P3C. Cells were pre-gated as live single + + + + + − CD11b CD45 Ly6G-CX3CR1 F480 as shown in Fig. 2A. B Compiled flow cytometry data showing the percentage of YFP and YFP cells in the + + + + macrophage gate (live single CD11b CD45 Ly6G-CX3CR1 F480 ) at different time points after ONC + P3C ranging from 2 to 168 dpi ONC in the retina. Data are shown as mean ± SEM. n = 3–8 biologically independent samples containing 1 retina from 1 mouse. C Cell count of YFP resident macrophages (microglia and BAMs) and YFP recruited MDMs at different time points after ONC + P3C ranging from 2 to 168 dpi ONC in the retina, as measured by flow cytometry. Data are shown as mean ± SEM. Repeated measures one-way ANOVA followed by Tukey’s multiple comparisons test, statistical significance between different time points is indicated using different letters: conditions that share the same letter are not significantly different, while conditions with different letters are significantly different from each other. n = 3–8 biologically independent samples with 1 retina from 1 mouse showed long-term engraftment. At 70 days post treat- was strongly reduced in ONC + P3C treated Ccr2- − +/+ ment 75 ± 9% of macrophages were still recruited YFP deficient mice as compared to Ccr2 controls, while cells (Fig. 3B). However, at 168 days post treatment, the infiltration of neutrophils was unaltered (Fig. 4A, B). percentage of YFP retinal macrophages had dropped to Furthermore, while most macrophages in ONC + P3C +/+ hi −/− 30 ± 2%. Together, these data suggest that while a fraction treated Ccr2 retinas were CD45 , in Ccr2 retinas low of recruited MDMs were long-lived, their engraftment they were C D45 , indicative of microglia (Fig. 4A). −/− was still transient as they were progressively lost and/or These data thus reveal that in Ccr2 mice the replaced by embryonic microglia. recruitment of monocyte-derived macrophages to the −/− Next, we aimed to investigate whether MDMs that retina was strongly impaired. Importantly, Ccr2 infiltrate the retina upon inflammatory stimulation mice also showed a significantly reduced number of promote axonal regeneration. Therefore, we per- CTB regenerating axons (Fig. 4C). This indicates that formed ONC + P3C treatment in Ccr2-deficient mice, monocyte-derived macrophages that are recruited to which are known to have low numbers of blood mono- the retina upon P3C treatment, promote axonal regen- cytes [45]. Flow cytometry at day 4 post treatment eration of injured RGCs. confirmed that the number of infiltrating monocytes Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 8 of 20 Fig. 4 Recruited monocyte-derived macrophages that infiltrate the retina upon inflammatory stimulation promote axonal regeneration. A +/+ −/− Representative flow cytometry plots of cells from the retina of Ccr2 and Ccr2 mice at 4 dpi ONC + P3C. Cells were pre-gated as shown in hi low Fig. 2A. Percentages of monocytes and macrophages and of CD45 and CD45 macrophages are shown at 4 dpi ONC + P3C. B Cell counts +/+ −/− of neutrophils, monocytes and macrophages at 4 dpi ONC + P3C in the retina of Ccr2 (black) and Ccr2 (coloured) mice, as measured by +/+ −/− flow cytometry. Data are shown as mean ± SEM. Statistical significance between CCR2 and CCR2 was evaluated via the Mann–Whitney test. n = 3–5 biologically independent samples with 1 retina from 1 mouse. * P < 0.05. C Quantification of axonal regeneration on longitudinal +/+ −/− cryosections of the optic nerve of Ccr2 and Ccr2 mice at 14 dpi after ONC + P3C, analysed at various distances, starting from 150 µm after the ONC lesion site. Data are shown as mean ± SEM. Repeated measures two-way ANOVA followed by Tukey’s multiple comparisons test. n = 8–10 mice per condition. ****p < 0.0001 and *p < 0.5 Nerve injury combined with inflammatory stimulation (Fig. 5A). Additionally, to obtain insights into the tran- results in the hyperactivation of microglia which remain scription factors and gene regulatory networks that distinct from recruited monocyte‑derived macrophages shape the activation state of macrophages, we performed To profile the cell states and heterogeneity of mac - single-cell regulatory network inference and clustering rophages in the retina following inflammatory stimula - (SCENIC) analysis on all microglia and macrophage clus- + + tion, we performed scRNA-seq on CD45 CD11b cells ters [46, 47]. sorted from the retina at day 4 and 8 post ONC + P3C P3C treatment resulted in a large population of treatment. The obtained data were combined with those C1qb macrophages within the retina (Fig. 5A–C). of the naive and day 4 post ONC retina in a single dataset Most clusters expressed high levels of Ms4a7 and Lyz2 A ndries et al. Acta Neuropathologica Communications (2023) 11:85 Page 9 of 20 (Fig. 5C), suggesting that these cells were recruited MDMs from ONC + P3C treated retinas (MDM2-7) monocyte-derived cells (MDM2-MDM8). Ms4a7 and were distinct from MDMs observed in the ONC only Lyz2 were low in cluster Mg5, which also showed signa- condition (MDM1), indicating that P3C administra- ture expression of Fscn1, Nav3, Capn3 and Mgll, genes tion not only affected the level of MDM recruitment, that were shared with naive and ONC-only microglia but also altered their molecular state. Furthermore, (Fig. 5C). Mg5 was also the only macrophage cluster gene expression in ONC + P3C MDMs was dynamic in within the ONC + P3C-treated retinas that exhibited time, showing transcriptional divergence between cells Sall3 and Sall1 expression (Fig. 5C), genes known to profiled at day 4 versus day 8 post treatment (Fig. 5B). be highly restricted to embryonic microglia. SCENIC Interestingly, MDMs from ONC + P3C retinas exhib- analysis showed that the microglia-associated regulons ited a high level of heterogeneity (clusters MDM2-8). A Ets1, Etv1 and Klf3 were also active in Mg5 (Fig. 5D). fraction of these MDMs showed enriched Hexb expres- Together, this suggests that Mg5 represented micro- sion (MDM2-5), and these cells could be further subdi- hi hi glia, while MDM2-8 were recruited macrophages vided into a Trem2 (MDM2), Ch25h (MDM3-4) and hi low that remained transcriptionally distinct from micro- H2-Aa cluster (MDM5) (Fig. 5F). Within the Hexb glia. Cluster Mg5 represented 3% and 13% of the pro- macrophages, MDM6 expressed Ndrg1, Clec4b1 and + + filed CD11b cells at day 4 and 8 post ONC + P3C , Cd300e, while cells in MDM7 were Trem2 and showed respectively (Additional File 3: Figure S3). This was in an enriched expression of genes involved in phago- line with our previous fate mapping data, where we cytosis and lipid metabolism, including Fabp5 and observed 4 ± 1% and 16 ± 2% YFP microglia within Gpnmb (Fig. 5F). The latter genes were also enriched in CD11b cells at 4 and 8 days post treatment, respec- MDMs from the ONC-only retinas (MDM1). Addition- tively, as observed via flow cytometry in ONC + P3C ally, MDM7 expressed genes that are associated with CreER treated Cx3cr1 :R26-YFP mice. However, as we also hypoxia or HIF1a signalling, including Arg1 and Bnip3 observed enriched expression of genes that are related [49]. An active HIF1 regulon in MDM7 was also identi- to MDMs or BAMs, such as Clec12a, Clec4a1 and Itgal fied via SCENIC (Fig. 5D). MDMs also exhibited prolif- [10] (Fig. 5E), we cannot rule out that part of the Mg5 erative potential as represented by the MDM8 cluster. cluster is monocyte or BAM-derived. The putative Mg5 microglia from ONC + P3C treated retinas showed Identification of a pro‑regenerative gene signature many differentially expressed genes when compared to in recruited monocyte‑derived macrophages the Mg2 DAM cluster from ONC-only retinas (Fig. 5E), We wished to assess the nature of the crosstalk that suggesting a hyperactivation upon P3C treatment. This exists between macrophages and injured RGCs in included a further downregulation of homeostatic sig- ONC + P3C retinas and to identify important mol- nature genes in Mg5 as compared to Mg2 and an induc- ecules and pathways for axonal regrowth. Hereto, we tion of genes related to inflammatory activation, as relied on the dataset from Tran et al., who profiled highlighted by gene ontology (GO) analysis (Additional RGCs from naive and injured retinas at various time File 4: Figure S4). Mg5 cells also showed robust expres- points post ONC via scRNA-seq [50]. We merged their sion of the anti-inflammatory cytokine Il10 (Fig. 5C). dataset with ours and relied on the NicheNet algorithm This suggests that in the retina, microglia can produce [51] to screen for potential ligand-receptor interactions IL10 following nerve injury combined with TLR2 stim- between macrophages ("senders") and RGCs ("receiv- ulation, which is not observed in brain microglia upon ers"). We focused our analysis on intrinsically photosen- peripheral LPS challenge [48]. sitive RGCs (ipRGCs) and αRGCs [50], which are the (See figure on next page.) Fig. 5 Nerve injury combined with inflammatory stimulation results in the hyperactivation of microglia which remain distinct from recruited monocyte-derived macrophages. A UMAP and cluster annotation showing 14,963 macrophages of healthy, injured (4 dpi ONC) and regenerating (4 and 8 dpi ONC + P3C) retinas. BAM, border associated macrophage, MDM: monocyte-derived macrophage, Mg: microglia. Data originate from retinas pooled from 32 naive mice, 10 ONC mice, 4 ONC + P3C 4 dpi mice and 4 ONC + P3C 8dpi mice. B UMAP showing 3974 macrophages of regenerating retinas at 4 dpi ONC + P3C and 6957 macrophages of regenerating retinas at 8 dpi ONC + P3C with a pie chart showing the distribution of different macrophage populations present in the injured + P3C treated retinas. Numbers in the pie chart are indicating percentages of macrophage subsets. C Corresponding UMAPs revealing the expression of signature genes that differentiate between the multiple macrophage populations. The colour (grey, low expression; purple, high expression) represents the expression profile in the macrophage clusters. D Corresponding UMAPs showing the bimodal regulon activity of specific microglia and monocyte-derived regulons, with red dots indicating an active regulon in the corresponding cells. Regulon here refers to a module of co-expressed genes together with their corresponding transcription factor. E Volcano plot displaying differential expression between Mg5 and Mg2, Genes with adjusted p-value < 0.01 and I Log2(FC) I > 1 are shown in red. F Corresponding dot plot to the UMAP plot in Figure A, showing the expression of subset-specific genes, with the dot size representing the percentage of cells expressing the gene and the colour representing its average expression within a cluster Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 10 of 20 Fig. 5 (See legend on previous page.) A ndries et al. Acta Neuropathologica Communications (2023) 11:85 Page 11 of 20 subclasses most resilient to ONC injury and known to clusters (Additional File 5: Figure S5C), highlighting the have regenerative capacity [50, 52]. NicheNet was used crosstalk between these macrophage subsets with the to predict the ligand-receptor interactions that may drive injured RGCs. Interestingly, NicheNet identified the the gene expression changes observed in healthy versus ligands Spp1, Thbs1, Vegfa and Igf1 in MDM7 (Fig. 6B). injured RGCs at day 4 post ONC. Top ligands from the These are secreted proteins that are known to promote various macrophage clusters were selected and ranked RGC survival and/or axon regeneration following ONC based on their Pearson values between a ligand’s target injury [52–57]. Another predicted ligand was Nrg1, predictions and the observed transcriptional response which is involved in axon regeneration following periph- within the ONC + P3C retinas (Fig. 6A), together with eral nerve injury [58]. Besides these predicted ligands, an overview of the potential receptors (Additional File MDM7 also showed enriched gene expression for other 5: Figure S5A) and affected target genes in RGCs (Addi - secreted factors known to be involved in either RGC tional File 5: Figure S5B). Most ligand-receptor interac- or peripheral tissue regeneration, including Slpi, Sdc1 tions were predicted for the MDM6 and MDM7 sender and Fstl1 (Additional File 6: Figure S6A) [59–64]. These Fig. 6 Identification of a pro-regenerative gene signature in monocyte-derived macrophages. A Top ligands of each macrophage cluster were selected and ranked based on their ligand activity values. Each ligand was assigned to a cluster if it was expressed highest in this cluster compared to the remaining sender clusters. The colour (white, low expression; orange, high expression) represents the predicted activity of the ligands. The expression of the top ligands in each cluster is shown with the colour (blue, low expression; red, high expression) representing the scaled average expression in the corresponding cluster. B Circle plot of potential ligand-receptor pairs, that shows the links between predicted ligands from MDM cluster 7 with their associated receptors found on alpha- and intrinsically photosensitive retinal ganglion cells Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 12 of 20 ligands are currently not included in the NicheNet and provides evidence for a key role played by recruited ligand-target database and thus cannot be predicted, MDMs. but their expression further suggests a pro-regenerative Similar to the brain, the retina contains yolk-sac- signature in MDM7. This pro-regenerative phenotype derived microglia that self-renew and rely on CSF1 or may underlie the macrophage-mediated axonal regen- IL34 for their maintenance [40]. Our single-cell tran- eration that we observe in P3C treated retinas. Previous scriptomic profiling confirmed that homeostatic micro - work has suggested the involvement of oncomodulin and glia are the predominant myeloid cells in the healthy SDF1/CXCL12 in monocyte/macrophage mediated RGC retina and also revealed subsets of retinal BAMs that regeneration [65–69]. However, gene expression of Onc likely correspond to perivascular macrophages [35]. and Cxcl12 was not or hardly detected in the C D11b Upon ONC-induced RGC degeneration, resident mac- cells from our dataset (Additional File 6: Figure S6B). rophages changed their expression profile by downregu - lating homeostatic genes and upregulating genes related The pro‑regenerative factors identified to inflammatory activation. Notably, the specific neu - in monocyte‑derived macrophages can promote axon rodegenerative expression profile of retinal microglia in regeneration via paracrine signalling our nerve crush injury model was similar to the expres- Pro-regenerative factors such as THBS1, SPP1 or IGF1 sion profile of retinal DAMs observed under conditions have been described in the context of autocrine signal- of light-damage-induced photoreceptor degeneration ling, as these proteins are produced by surviving RGCs [40] or in glaucoma models [70]. It also resembles the [52, 53, 56]. We hypothesized that the secretion of these molecular signature of brain DAMs observed in amy- factors by macrophages may also contribute to RGC loid models of Alzheimer’s disease [3, 10, 71] and other regrowth via paracrine signalling. To investigate this, we pathological conditions of the CNS [5, 6, 8, 15, 72–74]. composed two mixtures of recombinant proteins based Therefore, the DAM phenotype is broadly similar in the on the pro-regenerative signature identified in cluster brain and retina and across multiple etiologically dis- MDM7. Mix 1 (THBS1, SLPI, VEGFA, SPP1 and IGF1) tinct diseases or injury models. Furthermore, it is also consisted of secreted proteins reported to induce axonal not strictly disease-associated, as a similar cell state is regeneration in the central nervous system. Mix 2 con- also observed for microglia from healthy young mice tained factors that have been shown to stimulate regen- that engulf myelin [9] or apoptotic neurons [4, 7] and eration upon their release in the periphery (SDC1, NRG1 for non-parenchymal microglia that live on the choroid and FSTL1), but have not yet been assessed in the con- plexus epithelium [10]. Therefore, microglia seem to text of RGC axonal regrowth. The individual mixes or the react in a similar way to many homeostatic disturbances. combination of both were injected in the vitreous imme- This may be a common feature of tissue-resident mac - diately after ONC, and at day 3 and 7 post ONC (Fig. 7A). rophages, as their highly specialized phenotypes require Control mice received intravitreal injections of PBS. tissue-imprinting that may limit their plasticity towards Axonal regeneration was assessed by quantifying the inflammatory insults [16]. The functional significance of number of CTB regrowing axons at day 14 post ONC the DAM response is dependent on the nature of the dis- on longitudinal optic nerve sections (Fig. 7B, C). Interest- turbance. In the retina it can be protective during pho- ingly, both mixes of recombinant proteins induced axonal toreceptor degeneration [40] but detrimental for RGC regeneration in the optic nerve. The largest increase in survival during glaucoma [70]. However, local P3C treat- the number of regenerating axons was observed for Mix ment did significantly alter microglial activation beyond 1, while a lower but still significant axonal regrowth was the DAM state, further reducing homeostatic signature observed for Mix 2. These results show that the recom - genes and driving inflammatory activation. This suggests binant proteins within mix 1 and 2 can stimulate the that strong TLR signalling induced by local P3C injection regeneration of RGCs when injected in the vitreous. This led to a hyperactivation of microglia. However, as these implies that the secretion of these pro-regenerative fac- cells also expressed genes related to recruited MDMs, we tors by MDMs can contribute to axonal regeneration via cannot rule out that they were partly monocyte derived. paracrine signalling in RGCs. It will be interesting to further assess the ontogeny and functional significance of these cells during RGC regen - Discussion eration in follow-up studies. The nature of the cell populations, cellular states, as well Monocytes that are recruited during disease may react as the molecules and signalling pathways that underly differently to the local inflammatory cues as compared to inflammation-induced axonal regrowth have remained resident macrophages. Previous myeloid cell fate map- elusive [33]. Our work now further highlights how ping studies, performed after ONC injury [35] and in inflammation and CNS regeneration are intertwined other retinal injury models, have highlighted a role for A ndries et al. Acta Neuropathologica Communications (2023) 11:85 Page 13 of 20 Fig. 7 The pro-regenerative factors identified in monocyte-derived macrophages can promote axon regeneration via paracrine signalling. A Schematic overview of the experimental setup of the different recombinant protein mixes intravitreally injected in the eye. B Representative images of regenerating axons that were CTB-traced on longitudinal cryosections of the optic nerve of mice at 14 dpi ONC and intravitreally injected with PBS, mix 1, mix 2 and mix 1 + 2. The ONC site is indicated with an asterisk. Scale bar 100 µm. C Quantification of axonal regeneration on longitudinal cryosections of the optic nerve of mice at 14 dpi ONC and intravitreal injection of PBS, mix 1, mix 2 and mix 1 + 2. Axonal number was counted at various distances starting at 150 µm from the ONC lesion site. Data are shown as mean ± SEM. Repeated measures two-way ANOVA followed by Tukey’s multiple comparisons test, statistical significance between different conditions at the same distance is indicated with different letters: conditions that share the same letter are not significantly different, while conditions with different letters are significantly different from each other, n = 4–5 mice per condition recruited MDMs [14, 75–77]. We observed that inflam - and brain ischemia [19], MDMs were observed to exert matory treatment resulted in a strong recruitment of a neuroprotective role and to facilitate repair, by dis- MDMs into the retina. ScRNA-seq analysis revealed playing multiple functions including anti-inflammatory that MDMs were transcriptionally distinct from resi- [18, 20, 80] and scar degrading roles [21, 81], as well as dent microglia and BAMs and exhibited transcriptional the ability to support axonal growth [20, 65, 80, 82, 83]. heterogeneity. Our data thus confirm that recruited Although our data highlight the importance of recruited MDMs exhibit substantial plasticity and show disease- MDMs, we do not exclude that also other immune specific adaptation. As inhibiting MDM recruitment also or non-immune cell types contribute to the observed impaired axonal outgrowth, this revealed the importance inflammation-induced axonal regeneration. For instance, of recruited MDMs in promoting RGC regeneration. An a recent study identified a subset of immature neutro - important contribution of myeloid cells to inflammation- phils with neuroprotective and regenerative properties induced optic nerve regeneration has long been debated, [84]. Furthermore, reactive macroglia (i.e. astrocytes and with conflicting views [25, 26, 65, 78, 79]. A possible Müller glia) may also add to the inflammation-enhanced explanation is the differential effect of resident versus axonal regeneration. Evidence indeed exists for recipro- recruited macrophages. Further dissecting the role of cal interactions between innate immune cells and mac- individual macrophage subsets or cell states will provide roglia in shaping the CNS response to injury and disease additional insights into the multifaceted role of innate [35, 85–88]. immunity in neurodegeneration versus protection and One of the first myeloid-cell-derived molecules repair. In animal models of acute spinal cord injury [20] reported to play a central role in RGC axonal regrowth Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 14 of 20 Tamoxifen treatment is oncomodulin. This small calcium-binding protein is CreER Three-to-four-week-old anesthetized Cx3cr1 : R26- reported to be secreted by macrophages and/or neutro- YFP mice were treated with tamoxifen (Sigma Aldrich, phils during Zymosan-driven ocular inflammation and to 2+ 20 mg/ml dissolved in corn oil (Sigma Aldrich)), which induce RGC survival and axonal regeneration via a Ca / was injected subcutaneously near the fore- and hind calmodulin kinase-dependant pathway [65–68]. Fur- limbs (4 × 50 µl). These injections were repeated three thermore, SDF1, also known as CXCL12, expressed by times at 48 h intervals. infiltrating monocytes/MDMs was reported to enhance oncomodulin activity [69]. Upon ONC + P3C treatment Intraorbital optic nerve crush model we did not identify Ocm or Cxcl12 gene expression in res- Optic nerve crush (ONC) was performed as previ- ident or recruited macrophages. We did identify a clus- ously described [93, 94]. Briefly, mice were anesthetized ter of MDMs showing enriched expression of multiple by intraperitoneal injection of a mixture of ketamine genes encoding proteins that have been shown to exert (Anesketin, Eurovet, 75 mg/kg body weight) and medeto- pro-regenerative effects in the CNS and are known to be midine (Domitor, Pfizer, 1 mg/kg body weight) diluted secreted. One of the most highly expressed genes encodes in saline (NaCl, Fischer Scientific, 0.9% in H O). and a for thrombospondin 1 (THBS1), a protein that is well- topical aesthetic ointment (oxybuprocaïnehydrochloride, known to mediate axon regeneration of RGCs in an auto- Unicaïne, Thea Pharma, 0,4%) was applied on the injured crine fashion [53]. Bray et al. showed that the observed eye. An incision in the temporal side of the conjunctiva effect of THBS1 depends on syndecan 1 (SDC1), a was made in the left eye. Then, the posterior side of the THBS1-binding protein (Bray et al. 2019). Autocrine eye was exposed, allowing visualization of the optic nerve. Sdc1 signalling has also been reported to mediate axon The exposed optic nerve was crushed approximately regrowth in the mouse PNS [59]. Other secreted proteins 1 mm from the optic nerve head with a cross-action for- expressed after the ONC + P3C treatment and known ceps for 5 s. Thereafter, a fundoscopy was performed and to promote regeneration of RGCs include secretory leu- animals with signs of ischemia were excluded. Eyes from kocyte protease inhibitor SLPI, osteopontin (SPP1) and uninjured mice were used as controls. insulin growth factor 1 (IGF1) [52, 56, 63]. Moreover, the putatively regenerative MDMs also expressed genes for Intravitreal injections secreted proteins that are known to be pro-regenerative Intravitreal injections were performed as previously in the PNS, such as VEGF [54], NRG1 [89] and FSTL1 described [93, 95]. Briefly, a Hamilton syringe equipped [64]. Our study now reveals that these previously iden- with a 34G Hamilton needle was inserted into the nasal tified pro-regenerative molecules are also produced by a part of the eye of anesthetised mice, at the limbus, under specific subcluster of MDMs in the regenerating retina a 45° angle to avoid damage to the lens. To induce an and that these factors can induce axonal regrowth of acute inflammatory stimulation, 2 µl of a combination of injured RGCs via non-cell autonomous paracrine sig- Pam3Cys (P3C, Sigma Aldrich, 2.5 µg/µl in sterile PBS nalling. SCENIC analysis identified HIF1A as a putative [96]) and chlorophenylthio-cyclic adenosine monophos- transcription factor that was driving this cell state. It will phate (CPT-cAMP, cAMP analogue, Sigma-Aldrich, be important to further identify the microenvironmental 50 µM in PBS) was injected immediately after the ONC signals and the gene regulatory networks that control the surgery. To trace regenerating RGC axons in the optic pro-regenerative phenotype of MDMs in future studies. nerve, 2 µl of cholera toxin subunit B conjugated to an This may pave the road for macrophage-centred strate - Alexa Fluor 488 fluorophore (CTB-Alexa488; Sigma gies for inducing and promoting neuroprotection and Aldrich, 5 µg/µl in sterile PBS containing dimethylsul- repair following injury and disease. foxide (DMSO, Sigma Aldrich, 0,5%)) was injected one day before sacrificing the mice. Recombinant proteins Material and methods (THBS1, SLPI, VEGFa, SPP1, IGF1, SDC1, NRG1 FSTL1, Animals R&D systems, 1 µg/µl in sterile PBS) were injected All experiments were performed using a combination 3 × 2 µl at 0, 3 and 7dpi ONC. of male and female 8–12-week-old mice of following CreER strains: C57BL/6 wild-type, Lyz2-GFP [90], Cx3cr1 −/− [91], R26-YFP [92] and Ccr2 [45] mice, as outlined Flow cytometry of myeloid inflammatory cells in Additional File 7: Table S1. All animal experiments Mice were euthanized with an intraperitoneal injection were approved by the Institutional Ethical Committees of an overdose of pentobarbital (Dolethal, Vetoquinol, for Animal Experimentation of KU Leuven and the Vrije 200 mg/kg body weight) and transcardially perfused with Universiteit Brussel and were conducted in strict accord- saline to remove all blood. Eyes were harvested and reti- ance with the European and Belgian legislation. nas and optic nerves dissected and transferred to Roswell A ndries et al. Acta Neuropathologica Communications (2023) 11:85 Page 15 of 20 Park Memorial Institute (RPMI) 1640 medium (Gibco). 1 µg/ml in MACS buffer) was used to exclude dead cells + + For the retinal samples, the retinal pigment epithelium and CD45 CD11b cells were sorted using a BD FACS was detached from the retina, but the vitreous was not ARIA III (BD Biosciences) equipped with a 100 µm noz- removed in order to include the infiltrating immune zle. Sorted cells were collected in ME medium (RPMI cells that were localized at the retina-vitreous interface. medium (Gibco) supplemented with heat-inactivated A single-cell suspension was obtained by mechanical FCS (Gibco, 20%), l-glutamine (Gibco, 300 μg/ml), pen- and enzymatic (collagenase I (Worthington, 10 U/ml), icillin (Gibco, 100 units/ml) and streptomycin (Gibco, collagenase IV (Worthington, 400 U/ml) and DNase I 100 μg/ml), non-essential amino acids (Gibco, 1 mM), (Worthington, 30 U/ml) diluted in Hank’s Balanced Salt sodium pyruvate (Gibco, 1 mM), 2-mercaptoethanol Solution (HBSS) medium (Gibco)) dissociation as previ- (Sigma Aldrich, 0.05 mM) and ActD (Sigma Aldrich, ously described (3 × 10 min at 37 °C) [10]. Afterwards, 3 μM)) for further processing in the 10 × genomics these cells were filtered, washed in MACS buffer (HBSS platform. medium (Gibco) supplemented with sterile filtered ethyl - enediaminetetraacetic acid (EDTA; Duchefa; 2 mM) and heat-inactivated fetal calf serum (FCS, Gibco, 2%)) and Single‑cell RNA sequencing using 10 × genomics platform blocked with anti-mouse CD16/CD32 (clone 2.4G2, BD The library construction for single-cell RNA sequencing Biosciences, 2 µg/µl in MACS buffer). Cells were stained (scRNA-seq) was performed as previously described [10]. with fluorescent antibodies in MACS buffer. The follow - Briefly, cellular suspensions of an estimated final concen - ing antibodies were used: F4/80 (BV421, clone BM8, Bio- tration of 1000 cells/µl were loaded on a GemCode Sin- legend), CD11c (PE/Cy7, BV510, clone N418, Biolegend), gle Cell Instrument (10 × Genomics) to partition them Ly6G (FITC, clone 1A8, Biolegend), Cx3cr1 (PE, clone into single-cell gel beads-in-emulsion (GEM). GEMs and SA011F11, Biolegend), CD11b (PE/Cy7, BV510, clone scRNA-seq libraries were prepared using the GemCode M1/70, Biolegend), Ly6C (APC, BV421, clone HK1.4, Single Cell 3ʹ Gel Bead and Library Kit (10 × Genom- Biolegend), CD45 (APC/Cy7, BV421, clone 30-F11, Bio- ics, No. 120237) and the Chromium i7 Multiplex Kit legend), MHCII (PerCP/Cy5.5, clone M5/114.15.2, Biole- (10 × Genomics, No. 120262) according to manufac- gend). Flow cytometry data were acquired using the BD turer’s instructions. Briefly, GEM reverse-transcription FACS CANTO II (BD Biosciences) and analysed using incubation was performed, followed by amplification of Flowjo v10.8 software. the full-length, barcoded cDNA, enzymatic fragmenta- tion, library construction by 5’ adaptor attachment to generate Illumina-ready sequencing libraries and eventu- Isolation of retinal CD11b + CD45 + cells for single‑cell RNA ally sample indexing. The cDNA content of pre-fragmen - sequencing tation and post-sample indexing was analysed using the Mice were euthanized and transcardially perfused with 2100 BioAnalyzer (Agilent). The libraries were sequenced saline. Retinas without retinal pigment epithelium on an Illumina HiSeq4000 flow cell with sequencing set - were transferred to RPMI (Gibco) containing actino- tings following the recommendations of 10 × Genomics mycin D (ActD, Sigma Aldrich, 30 μM) [97]. To obtain (read 1: 26 cycles; read 2: 98 cycles; index i7: eight cycles; sufficient number of cells, retinas were pooled from index i5: no cycles; 2.1 pM loading concentration). individual mice: 32 mice for the naïve sample (32 reti- nas), 10 mice for ONC sample, 4 mice for ONC + P3C Alignment and quantification of gene expression 4dpi sample, 4 mice for ONC + P3C 8dpi sample. The in single‑cell RNA sequencing data retinal samples underwent mechanical and enzymatic The Cell Ranger software (10 × Genomics) v.6.0.2 was (collagenase I (Worthington, 10 U/ml), collagenase used to perform sample demultiplexing and alignment of IV (Worthington, 400 U/ml) and DNase I (Roche, sequencing reads to the reference genome (Mus muscu- 30 U/ml) in HBSS medium (Gibco) containing ActD lus mm10), barcode processing, unique molecular identi- (Sigma Aldrich, 15 μM)) dissociation (3 × 10 min at fiers filtering and single-cell 3ʹgene counting. The average 37 °C). Afterwards, cells were filtered, resuspended in of the mean reads per cell was 49,780 ± 1629 SD, with an MACS buffer (HBSS medium (Gibco) supplemented average sequencing saturation metric of 59% ± 8% SD, as with sterile filtered EDTA (Duchefa; 2 mM) and heat- calculated by Cell Ranger. The further pre-processing and inactivated FCS (Gibco, 2%), containing ActD (Sigma analysis of the gene expression count matrices was per- Aldrich, 3 μM)) and blocked with anti-mouse CD16/ formed in R using Seurat v.3.2.3, DropletUtils v1.10.1.2, CD32 (clone 2.4G2, BD Biosciences, 2 µg/µl in MACS scater 1.18.3. The cellular barcodes, associated with low buffer). Cells were stained with CD45-APC (30-F11, quality “empty” droplets, were filtered out using the Biolegend) and CD11b-PE/Cy7 (M1/70, Biolegend) in “emptyDrops” function of the DropletUtils package with MACS. 4′,6-diamidino-2-phenylindole (DAPI, Dako, Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 16 of 20 the recommended FDR cutoff ≤ 0.1 for deviation from MDM2-7 were defined as sender, while alpha and ipRGC the ambient RNA profile. The gene expression matri - were defined as the receiver cell populations. Potential ces were further filtered for low quality cells, normal - ligands and receptors were identified as genes, expressed in ized and scaled, followed by selection of highly variable at least 10% of the sender/receiver population, respectively, genes, principal components analysis and clustering as and present in the prior interaction model. To prioritise the previously described (Scheyltjens et al., 2022). The genes, identified interactions, we performed NicheNet ligand activ - specifically expressed in each cluster, were identified via ity analysis, which ranks the ligands based on the presence differential expression analysis with the “FindMarkers” of their target genes in the gene set of interest, here defined function of Seurat (Wilcoxon Rank Sum test). The p-val - as the differentially expressed genes in the alpha and ipRGCs ues of differential expression were adjusted for multiple between the 4dpi ONC and the naive condition (adjusted p testing with Bonferroni correction. Clustering results value < 0.05). Next, we selected the top 40 ligands with high- were visualized using two-dimensional scatter plots est ligand activity (based on the Pearson score) and added with the Uniform Manifold Approximation and Projec- three ligands with lower ligand activity that had known neu- tion (UMAP) method. Several of the identified clusters roprotective effects (Thbs1, Nrg1 and Igf1). For the selected exhibited simultaneous expression of both macrophage 43 ligands, we inferred the top predicted receptors and target and neutrophil gene markers, e.g. C1qa, C1qb, P2ry12, genes in the receiver cells. For visualising the ligand—target Ms4a7; S100a8, S100a9, Retnlg, Csf3r. Additionally, those genes interactions, we showed the 110 most strongly pre- clusters showed a high doublet score, as calculated by the dicted targets of at least one of the selected ligands, that were scDblFinder package v.1.4.0, therefore they were assumed also part of the gene set of interest. to be macrophage–neutrophil aggregates and were excluded from further analysis. Immunohistochemistry on retinal whole mounts and cryosections of retina and optic nerve Single‑cell regulatory network inference and clustering Mice were euthanized as described above and transcar- using SCENIC dially perfused with saline followed by phosphate buff - We performed single-cell regulatory network inference ered paraformaldehyde (PFA, pH 7.4, Sigma Aldrich, analysis using SCENIC v1.2.4 [46] using the raw, untrans- 4% in PBS). For retinal whole mount stainings, the eyes formed UMI counts as input and following the proposed and subsequently the retinas were dissected, post-fixed workflow. The co-expression network was generated in PFA (pH 7.4, Sigma Aldrich, 4% in PBS) for 1 h and using GRNBoost2 via arboreto v0.1.5. For running GRN- rinsed in PBS. The retinas were incubated overnight Boost2, the expression matrix was filtered for genes with with the primary antibody (rabbit anti-IBA1, Wako, over 30 UMI counts and expressed in at least 40 cells. 1/2000 diluted in PBS supplemented with pre-immune The resulting transcription factor by gene targets matrix donkey serum (PID, Merck, 2%) and triton X-100 was imported in R and further analysed with the SCENIC (VWR, 2%)). After rinsing in PBS, the retinas were workflow with default parameters. The regulon activity, incubated for 2 h with a donkey anti-rabbit Alexa Fluor which identifies and scores gene regulatory networks or 488 secondary antibody (DAR488, Dako, 1/200 in PBS regulons in single cells, was calculated using AUCell as supplemented with PID (Merck, 2%) and triton X-100 previously described [46]. The better the gene targets of (VWR, 2%)). Mosaic pictures of the entire retinal whole a regulon match the highly expressed genes of a certain mounts were made using a confocal scanning micro- cell, the higher the AUC value (also named regulon activ- scope (Olympus FV 1000D). Microglia density, soma ity) of that regulon in that particular cell. The regulons size and roundness were analysed using a spatial statis- were visualized in a network using Cytoscape v.3.9.1 [98]. tics approach, all as previously described [99]. For retinal or optic nerve cryosections, complete Modelling the intercellular communication using NicheNet eyes and optic nerves were dissected, postfixed for We extracted gene expression matrices of RGCs of control 1 h at room temperature and cryoprotected through mice and mice 4 days post ONC using GSE137398 [50]. an ascending series of sucrose (Sigma Aldrich, 10%– The gene expression data was pre-processed as described 20%–30% in PBS). Afterwards, eyes or optic nerves above. The clusters "41_AlphaONT", "42_AlphaOFFS", were embedded in TissueTek (Sakura) and 14 µm "43_AlphaONS" and "45_AlphaOFFT" were grouped as thick sagittal sections of the eyes or longitudinal optic alphaRGCs, while the clusters "22_M5", "31_M2", "33_M1" nerve sections were made. For immunolabeling of the and "40_M1dup" were grouped as ipRGCs. For predict- cryosections of the eyes, epitope retrieval was accom- ing interactions between the macrophages and the RGCs, plished using citrate buffer (pH 6, citric acid (Chem- we applied the NicheNet package (v. 1.1.0), using the pre- lab, 10 mM) and Tween 20 (Sigma Aldrich, 0.05%) in build NicheNet prior model of ligand-receptor interactions. H O). Aspecific binding places were saturated with PID 2 A ndries et al. Acta Neuropathologica Communications (2023) 11:85 Page 17 of 20 (Merck, 20%) in Tris-sodium chloride blocking buffer (n). Statistically significant differences between multiple (TNB, triton X-100 (VWR, 1.5 mM %), Tris-HCl (Acros groups are specified using different letters. Conditions Organics, 0.1 M), NaCl (Fischer Scientific, 150 mM) with the same letter are not significantly different, while and blocking reagent (Perkin Elmer, 0.5%) in PBS)) conditions with different letters are significantly dif - and the primary antibodies (chicken anti-GFP, Abcam, ferent from each other. Statistical significance between 1/500 in TNB and rabbit anti-IBA1, Wako, 1/2000 in two groups were specified with **** for p < 0.0001, *** for TNB) were incubated overnight at room temperature. p < 0.001, ** for p < 0.01, and * for p < 0.5. After rinsing, the slides were incubated with, respec- tively, donkey anti-chicken Alexa Fluor 488 (DACh488, Supplementary Information Dako,1/200 in TNB) and donkey anti-rabbit biotin sec- The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s40478- 023- 01580-3. ondary antibody (DARbiotin, Dako, 1/300 in TNB), followed by subsequent incubation with streptavidin- + + Additional file 1. scRNA-seq of CD45 CD11b cells from naïve or ONC horse radish peroxidase (Strep-HRP, Dako, 1/100 in + retinas. A UMAP showing all CD11b cells profiled from the healthy and TNB) and tyramid signal amplification (TSA Cy3-Tyr, ONC retinas. BAM, border associated macrophage, cDC: conventional dendritic cell, migDC: migratory dendritic cell, MDM: monocyte-derived Thermofisher Scientific, 1/50 in amplification buffer). macrophage, Mg: microglia, MO, monocyte, N: neutrophils, NK: natural Finally, the slides were counterstained with DAPI killer cell. B UMAPs showing the expression of the indicated genes. Red (Dako, 1 µg/ml in PBS). Images of the mid-sagittal reti- line highlights the putative macrophage-neutrophil doublets. C Volcano plot displaying differential expression between Mg3 and Mg1. Genes with nal cryosections were taken using a Leica DM6 (Olym- adjusted p-value <0.01 and I Log2I >1 are shown in red. D Volcano plot pus) fluorescent microscope. For the optic nerves, displaying differential expression between Mg4 and Mg1. Genes with images of mid-longitudinal sections that contained the adjusted p-value <0.01 and I Log2I >1 are shown in red. E Quantification of the density and activityof microglia in retina at different timepoints ONC site were taken using a confocal scanning micro- after ONC corresponding with images shown in figure 1E. Data are shown scope (Olympus FV 1000D). as mean ± SEM. Repeated measures one-way ANOVA followed by Tukey’s multiple comparisons test, statistical significance between different time - points is indicated using different letters: conditions that share the same letter are not significantly different, while conditions with different letters Quantification of axonal growth are significantly different from each other. n=3-4 mice per condition. F Axon growth was quantified on three mid-longitudinal UMAPs showing the expression of the indicated genes, corresponding to the dataset shown in S1A. cryosections of the optic nerve by manually counting Additional file 2. Inflammatory treatment stimulates axonal initiation. the number of C TB axons every 150 µm (distance d) A Representative images of longitudinal cryosections of the optic nerve beyond the crush site, using ImageJ [100]. In addition, at showing regenerating axons that were CTB-traced at different timepoints each distance, the cross-sectional width of the nerve was after ONC and ONC+P3C. The ONC site is indicated by an asterisk. Scale bar 50µm. B Quantification of axonal regeneration in the optic nerve of measured along. The total estimated number of axons mice at different timepoints after ONC or ONC combined with P3C treat - in the optic nerve extending distance d from the ONC ment. The number of regrowing axons was analysed at various distances lesion site was calculated using following formula where starting at 150 µm from the ONC lesion site. Representative images of n = 3 mice per condition. Quantitative data after ONC+IS are shown as mean the radius of the optic nerve was set at r = 150 µm and the ± SEM. Repeated measures two-way ANOVA followed by Tukey’s multiple thickness of the sections was t = 14 µm, all as described comparisons test, statistical significance between different conditions previously [101]. at the same distance is indicated with different letters, n=3-5 mice per condition. Average(#axons/µ m of nerve width) Additional file 3. Full scRNA-seq dataset of naïve, ONC and ONC+P3C �a = πr . + + CD11b CD45 cells. A UMAP and cluster annotation showing 22081 cells of both healthy, injuredand regeneratingretinas. BAM, border associated macrophage, cDC: conventional dendritic cell, migDC: migratory dendritic The results obtained for each of the three sections per cell, MDM: monocyte-derived macrophage, Mg: microglia, MO, monocyte, nerve were averaged. N: neutrophils, NK: natural killer cell. B,C UMAP showing 6997 cells of retinas at 4dpi ONC+P3Cand 7956 cells of retinas at 8dpi ONC+P3C. Individual pie charts show the distribution of neutrophils, monocytes or Statistics alle immune populations. Numbers in the pie chart are percentages of the cells from the corresponding cluster. Statistical analyses were performed using GraphPad Prism 8 software (GraphPad Software). Normal distri- Additional file 4. Expression of pro-regenerative genes in cluster MDM7. Gene ontology analysis on the upregulated genes in Mg5 versus Mg2> 20; bution was evaluated using a Kolmogorov–Smirnov test log2>1) showing the top 20 enriched GO terms for Mg5. and parallel equal variance between groups was tested. Additional file 5. Nichenet analysis of MDMs against injured RGCs. A Outliers were identified and excluded based on a Grubb’s Overview of potential receptors on the retinal ganglion cells of the ligands test (extreme studentised deviate method). The values are expressed by the different macrophage clusters. The colourrepresents the regulatory potential of the receptors based on the prior model of ligand- expressed as mean values ± standard error (SEM). Sta- receptor interactions.Receptor expression in the different retinal ganglion tistical tests are specified in the figure legends, together cell populations is shown with the colourrepresenting the scaled average with the number of biologically independent samples Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 18 of 20 2. Drieu A, Du S, Storck SE, Rustenhoven J, Papadopoulos Z, Dykstra T et al expression in the corresponding cluster. B Overview of the predicted (2022) Parenchymal border macrophages regulate the flow dynamics target genes in the retinal ganglion cells of the ligands expressed by of the cerebrospinal fluid. Nature 611:585–593 the different macrophage clusters. The colourrepresents the regulatory 3. Keren-Shaul H, Spinrad A, Weiner A, Matcovitch-Natan O, Dvir-Sztern- potential of the target genes based on the prior model of ligand-target feld R, Ulland TK et al (2017) A unique microglia type associated with gene interactions. C Circle plot of potential ligand-receptor pairs. It shows restricting development of Alzheimer’s disease. Cell 169:1276–1290 the links between predicted ligands from the different monocyte-derived 4. Anderson SR, Roberts JM, Zhang J, Steele MR, Romero CO, Bosco A macrophage clusters of the regenerating retinawith their associated et al (2019) Developmental apoptosis promotes a disease-related gene receptors found on alpha- and intrinsically photosensitive retinal ganglion signature and independence from CSF1R signaling in retinal microglia. cells. Cell Rep 27:2002–2013 5. Hammond TR, Dufort C, Dissing-Olesen L, Giera S, Young A, Wysoker Additional file 6. Pro-regenerative gene signature in cluster MDM7. A A et al (2019) Single-cell RNA sequencing of microglia throughout Corresponding dot plot of the recruited monocyte-derived macrophage the mouse lifespan and in the injured brain reveals complex cell-state populations showing the expression of selected pro-regenerative genes, changes. Immunity 50:253–271 with the dot size representing the percentage of cells expressing the gene 6. Jordão MJC, Sankowski R, Brendecke SM, Locatelli G, Tai YH et al (2019) and the colour representing its average expression within a cluster. B Neuroimmunology: single-cell profiling identifies myeloid cell subsets UMAP plots showing expression of the indicated genes, Cxcl12 and Ocm, with distinct fates during neuroinflammation. Science 363:eaat7554 corresponding to the dataset shown in S3A. 7. Li Q, Cheng Z, Zhou L, Darmanis S, Neff NF, Okamoto J et al (2019) Additional file 7. Overview of mouse strains used in this study. Developmental heterogeneity of microglia and brain myeloid cells revealed by deep single-cell RNA sequencing. Neuron 101:207–223 8. Masuda T, Sankowski R, Staszewski O, Böttcher C, Amann L et al (2019) Acknowledgements Spatial and temporal heterogeneity of mouse and human microglia at We thank Veronique Brouwers, Marijke Christiaens and Lut Noterdaeme single-cell resolution. Nature 566:388–392 for their skilful technical assistance and Evelien Herinckx for her excellent 9. Safaiyan S, Besson-Girard S, Kaya T, Cantuti-Castelvetri L, Liu L, Ji H animal care taking. This research was funded by the Research Council of KU et al (2021) White matter aging drives microglial diversity. Neuron Leuven (KU Leuven BOF-OT/14/064), the Research Foundation Flanders (FWO, 109:1100–1117 G0B2315N and G053217N) and Innoviris (Attract Grant BB2B 2015-2). Luca 10. Van Hove H, Martens L, Scheyltjens I, De Vlaminck K, Pombo Antunes Masin, Steven Bergmans, Marie Claes and Lies De Groef were/are supported AR, De Prijck S et al (2019) A single-cell atlas of mouse brain mac- by a fellowship of the Research Foundation Flanders. rophages reveals unique transcriptional identities shaped by ontogeny and tissue environment. Nat Neurosci 22:1021–1035 Author contributions 11. Rashid K, Akhtar-Schaefer I, Langmann T (2019) Microglia in retinal LA together with LDG, LM and KM conceptualized the study and wrote, degeneration. Front Immunol 10:1975–1994 reviewed and edited the manuscript. LA collected the data with the help of 12. Choi S, Guo L, Cordeiro MF (2021) Retinal and brain microglia in multi- LM, IS, HH, KDV, SB and MC. LA, DK, IS and KM performed the bioinformatic ple sclerosis and neurodegeneration. Cells 10:1507–1528 analysis of the scRNA-seq data. All authors have read and agree to the pub- 13. Chen X, Holtzman DM (2022) Emerging roles of innate and adaptive lished version of the manuscript. immunity in Alzheimer’s disease. Immunity 55:2236–2254 14. Yu C, Roubeix C, Sennlaub F, Saban DR (2020) Microglia versus Data availability monocytes: distinct roles in degenerative diseases of the retina. Trends The scRNA-seq datasets described in this article will be made accessible Neurosci 43:433–449 through our interactive webserver at www. brain immun eatlas. org. Further- 15. De Vlaminck K, Van Hove H, Kancheva D, Scheyltjens I, Pombo Antunes more, all raw data and gene-cell count matrices are deposited at GEO (NCBI) AR, Bastos J et al (2022) Differential plasticity and fate of brain-resident with accession code GSE232470. and recruited macrophages during the onset and resolution of neuro- inflammation. Immunity 55:2085–2102 16. Guilliams M, Svedberg FR (2021) Does tissue imprinting restrict mac- Declaration rophage plasticity? Nat Immunol 22:118–127 17. Shechter R, Schwartz M (2013) Harnessing monocyte-derived mac- Competing interests rophages to control central nervous system pathologies: no longer ‘if ’ The authors declare no conflict of interest. but ‘how.’ J Pathol 229:332–346 18. Shechter R, Miller O, Yovel G, Rosenzweig N, London A, Ruckh J et al Author details (2013) Recruitment of beneficial M2 macrophages to injured spinal Neural Circuit Development and Regeneration Research Group, Animal cord is orchestrated by remote brain choroid plexus. Immunity Physiology and Neurobiology Division, Department of Biology, Leuven Brain 38:555–569 Institute, KU Leuven, Naamsestraat 61, Box 2464, 3000 Louvain, Belgium. 19. Wattananit S, Tornero D, Graubardt N, Memanishvili T, Monni E, Laboratory for Molecular and Cellular Therapy, Vrije Universiteit Brussel, Tatarishvili J et al (2016) Monocyte-derived macrophages contribute Laarbeeklaan 103, 1090 Brussels, Belgium. Myeloid Cell Immunology Lab, VIB to spontaneous long-term functional recovery after stroke in mice. J Center for Inflammation Research, Brussels, Belgium. Cellular Communication Neurosci 36:4182–4195 and Neurodegeneration Research Group, Animal Physiology and Neurobi- 20. Shechter R, London A, Varol C, Raposo C, Cusimano M, Yovel G et al ology Division, Department of Biology, Leuven Brain Institute, KU Leuven, (2009) Infiltrating blood-derived macrophages are vital cells playing an 3000 Louvain, Belgium. anti-inflammatory role in recovery from spinal cord injury in mice. PLoS Med 6:e1000113 Received: 29 March 2023 Accepted: 8 May 2023 21. London A, Itskovich E, Benhar I, Kalchenko V, Mack M, Jung S et al (2011) Neuroprotection and progenitor cell renewal in the injured adult murine retina requires healing monocyte-derived macrophages. J Exp Med 208:23–39 22. Lee-Liu D, Edwards-Faret G, Tapia VS, Larraín J (2013) Spinal cord References regeneration: lessons for mammals from non-mammalian vertebrates. 1. Prinz M, Masuda T, Wheeler MA, Quintana FJ (2021) Microglia and Genesis 51:529–544 central nervous system-associated macrophages mdash from origin to disease modulation. Annu Rev Immunol 39:251–277 A ndries et al. Acta Neuropathologica Communications (2023) 11:85 Page 19 of 20 23. David S, Kroner A (2011) Repertoire of microglial and macrophage cytokine responses in C-C chemokine receptor 2 knockout mice. J Clin responses after spinal cord injury. Nat Neurosci 12:388–399 Invest 100:2552–2561 24. Gensel JC, Zhang B (2015) Macrophage activation and its role in repair 46. Aibar S, González-Blas CB, Moerman T, Huynh-Thu VA, Imrichova H, and pathology after spinal cord injury. Brain Res 1619:1–11 Hulselmans G et al (2017) SCENIC: single-cell regulatory network infer- 25. Hilla AM, Diekmann H, Fischer D (2017) Microglia are irrelevant for neu- ence and clustering. Nat Methods 14:1083–1086 ronal degeneration and axon regeneration after acute injury. J Neurosci 47. Van de Sande B, Flerin C, Davie K, De Waegeneer M, Hulselmans G, 37:6113–6124 Aibar S et al (2020) A scalable SCENIC workflow for single-cell gene 26. Hauk TG, Müller A, Lee J, Schwendener R, Fischer D (2008) Neuroprotec- regulatory network analysis. Nat Protoc 15:2247–2276 tive and axon growth promoting effects of intraocular inflammation 48. Shemer A, Scheyltjens I, Frumer GR, Kim JS, Grozovski J, Ayanaw S et al do not depend on oncomodulin or the presence of large numbers of (2020) Interleukin-10 prevents pathological microglia hyperactivation activated macrophages. Exp Neurol 209:469–482 following peripheral endotoxin challenge. Immunity 53:1033–1049 27. Leibinger M, Andreadaki A, Gobrecht P, Levin E, Diekmann H, Fischer 49. Pombo Antunes AR, Scheyltjens I, Lodi F, Messiaen J, Antoranz A, Duer- D (2016) Boosting central nervous system axon regeneration by inck J et al (2021) Single-cell profiling of myeloid cells in glioblastoma circumventing limitations of natural cytokine signaling. Mol Ther across species and disease stage reveals macrophage competition and 24:1712–1725 specialization. Nat Neurosci 24:595–610 28. Hauk TG, Leibinger M, Müller A, Andreadaki A, Knippschild U, Fischer 50. Tran NM, Shekhar K, Whitney IE, Jacobi A, Benhar I, Hong G et al (2019) D (2010) Stimulation of axon regeneration in the mature optic nerve Single-cell profiles of retinal ganglion cells differing in resilience to by intravitreal application of the toll-like receptor 2 agonist Pam3Cys. injury reveal neuroprotective genes. Neuron 104:1039–1055 Invest Ophthalmol Vis Sci 51:459–464 51. Browaeys R, Saelens W, Saeys Y (2020) NicheNet: modeling intercel- 29. Diekmann H, Leibinger M, Fischer D (2013) Do growth-stimulated lular communication by linking ligands to target genes. Nat Methods retinal ganglion cell axons find their central targets after optic nerve 17:159–162 injury? New insights by three-dimensional imaging of the visual path- 52. Jacobi A, Tran NM, Yan W, Benhar I, Tian F, Schaffer R et al (2022) way. Exp Neurol 248:254–257 Overlapping transcriptional programs promote survival and axonal 30. Andries L, Masin L, Salinas-Navarro M, Zaunz S, Claes M, Bergmans S regeneration of injured retinal ganglion cells. Neuron 110:2625–2645 et al (2021) Mmp2 modulates inflammatory response during axonal 53. Bray ER, Yungher BJ, Levay K, Ribeiro M, Dvoryanchikov G, Ayupe AC regeneration in the murine visual system. Cells 10:1672–1691 et al (2019) Thrombospondin-1 mediates axon regeneration in retinal 31. Benowitz LI, Popovich PG (2011) Inflammation and axon regeneration. ganglion cells. Neuron 103:642–657 Curr Opin Neurol 24:577–583 54. Böcker-Meffert S, Rosenstiel P, Röhl C, Warneke N, Held-Feindt J, Sievers 32. Cui Q, Yin Y, Benowitz LI (2009) The role of macrophages in optic nerve J et al (2002) Erythropoietin and VEGF promote neural outgrowth from regeneration. Neuroscience 158:1039–1048 retinal explants in postnatal rats. IOVS 43:2021–2026 33. Andries L, De Groef L, Moons L (2020) Neuroinflammation and optic 55. Rosenstein JM, Mani NM, Khaibullina A, Krum JM (2003) Neurotrophic nerve regeneration: where do we stand in elucidating underlying cel- effects of vascular endothelial growth factor ( VEGF) on organo - lular and molecular players? Curr Eye Res 45:397–409 typic cortical explants and primary cortical neurons. J Neurosci 34. Wong KA, Benowitz LI (2022) Retinal Ganglion cell survival and axon 23:11036–11044 regeneration after optic nerve injury: role of inflammation and other 56. Duan X, Qiao M, Bei F, Kim I-JJ, He Z, Sanes JRR (2015) Subtype-specific factors. Int J Mol Sci 23:10179–10194 regeneration of retinal ganglion cells following axotomy: effects of 35. Benhar I, Ding J, Yan W, Whitney IE, Jacobi A, Sud M et al (2023) Tem- osteopontin and mTOR signaling. Neuron 85:1244–1256 poral single cell atlas of non-neuronal retinal cells reveals dynamic, 57. Joshi Y, Sória MG, Quadrato G, Inak G, Zhou L, Hervera A et al (2015) The coordinated multicellular responses to central nervous system injury. MDM4/MDM2-p53-IGF1 axis controls axonal regeneration, sprouting Nat Immunol 24:700–713 and functional recovery after CNS injury. Brain 138:1843–1862 36. Bennett FC, Bennett ML, Yaqoob F, Mulinyawe SB, Grant GA, Hayden 58. Fricker FR, Lago N, Balarajah S, Tsantoulas C, Tanna S, Zhu N et al (2011) Gephart M et al (2018) A combination of ontogeny and CNS environ- Axonally derived neuregulin-1 is required for remyelination and regen- ment establishes microglial identity. Neuron 98:1170–1183 eration after nerve injury in adulthood. J Neurosci 31:3225–3233 37. Shemer A, Grozovski J, Tay TL, Tao J, Volaski A, Süß P et al (2018) 59. Murakami K, Tanaka T, Bando Y, Yoshida S (2015) Nerve injury induces Engrafted parenchymal brain macrophages differ from microglia in the expression of syndecan-1 heparan sulfate proteoglycan in primary transcriptome, chromatin landscape and response to challenge. Nat sensory neurons. Neuroscience 300:338–350 Commun 9:1–16 60. Edwards TJ, Hammarlund M (2014) Syndecan promotes axon regenera- 38. Cronk JC, Filiano AJ, Louveau A, Marin I, Marsh R, Ji E et al (2018) Periph- tion by stabilizing growth cone migration. Cell Rep 8:272–283 erally derived macrophages can engraft the brain independent of 61. Taggart CC, Cryan SA, Weldon S, Gibbons A, Greene CM, Kelly E et al irradiation and maintain an identity distinct from microglia. J Exp Med (2005) Secretory leucoprotease inhibitor binds to NF-κB binding sites in 215:1627–1647 monocytes and inhibits p65 binding. J Exp Med 202:1659–1668 39. Deczkowska A, Keren-Shaul H, Weiner A, Colonna M, Schwartz M, Amit 62. Mueller AM, Pedré X, Stempfl T, Kleiter I, Couillard-Despres S, Aigner L I (2018) Disease-associated microglia: a universal immune sensor of et al (2008) Novel role for SLPI in MOG-induced EAE revealed by spinal neurodegeneration. Cell 173:1073–1081 cord expression analysis. J Neuroinflamm 5:1–20 40. O’Koren EG, Yu C, Klingeborn M, Wong AYW, Prigge CL, Mathew R et al 63. Ghasemlou N, Bouhy D, Yang J, López-Vales R, Haber M, Thuraisingam T (2019) Microglial function is distinct in different anatomical locations et al (2010) Beneficial effects of secretory leukocyte protease inhibitor during retinal homeostasis and degeneration. Immunity 50:723–737 after spinal cord injury. Brain 133:126–138 41. Sala Frigerio C, Wolfs L, Fattorelli N, Thrupp N, Voytyuk I, Schmidt I et al 64. Wei K, Serpooshan V, Hurtado C, Diez-Cunado M, Zhao M, Maruyama S (2019) The major risk factors for Alzheimer’s disease: age, sex, and genes et al (2015) Epicardial FSTL1 reconstitution regenerates the adult mam- modulate the microglia response to Aβ plaques. Cell Rep 27:1293–1306 malian heart. Nature 525:479–485 42. Heindl S, Gesierich B, Benakis C, Llovera G, Duering M, Liesz A (2018) 65. Yin Y, Henzl MT, Lorber B, Nakazawa T, Thomas TT, Jiang F et al (2006) Automated morphological analysis of microglia after stroke. Front Cell Oncomodulin is a macrophage-derived signal for axon regeneration in Neurosci 12:1–11 retinal ganglion cells. Nat Neurosci 9:843–852 43. Guo C, Cho KS, Li Y, Tchedre K, Antolik C, Ma J et al (2018) IGFBPL1 66. Yin Y, Cui Q, Gilbert HY, Yang Y, Yang Z, Berlinicke C et al (2009) Onco- regulates axon growth through IGF-1-mediated signaling cascades. Sci modulin links inflammation to optic nerve regeneration. Proc Natl Acad Rep 8:1–13 Sci 106:19587–19592 44. Goldmann T, Wieghofer P, Müller PF, Wolf Y, Varol D, Yona S et al (2013) A 67. Kurimoto T, Yin Y, Habboub G, Gilbert H-YY, Li Y, Nakao S et al (2013) new type of microglia gene targeting shows TAK1 to be pivotal in CNS Neutrophils express oncomodulin and promote optic nerve regenera- autoimmune inflammation. Nat Neurosci 16:1618–1626 tion. J Neurosci 33:14816–14824 45. Boring L, Gosling J, Chensue SW, Kunkel SL, Farese RV, Broxmeyer HE 68. Singh B, Plemel JR (2014) Neutrophil contribution in facilitating optic et al (1997) Impaired monocyte migration and reduced type 1 ( Th1) nerve regeneration. J Neurosci 34:1081–1082 Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 20 of 20 69. Xie L, Cen LP, Li Y, Gilbert HY, Strelko O, Berlinicke C et al (2022) 90. Faust N, Varas F, Kelly LM, Heck S, Graf T (2000) Insertion of enhanced Monocyte-derived SDF1 supports optic nerve regeneration and alters green fluorescent protein into the lysozyme gene creates mice with retinal ganglion cells’ response to Pten deletion. Proc Natl Acad Sci U S green fluorescent granulocytes and macrophages. Blood 96:719–726 A 119:e2113751119 91. Yona S, Kim KW, Wolf Y, Mildner A, Varol D, Breker M et al (2013) Fate 70. Margeta MA, Yin Z, Madore C, Pitts KM, Letcher SM, Tang J et al mapping reveals origins and dynamics of monocytes and tissue mac- (2022) Apolipoprotein E4 impairs the response of neurodegenerative rophages under homeostasis. Immunity 38:79–91 retinal microglia and prevents neuronal loss in glaucoma. Immunity 92. Srinivas S, Watanabe T, Lin CS, William CM, Tanabe Y, Jessell TM et al 55:1627–1644 (2001) Cre reporter strains produced by targeted insertion of EYFP and 71. Krasemann S, Madore C, Cialic R, Baufeld C, Calcagno N, El Fatimy R et al ECFP into the ROSA26 locus. BMC Dev Biol 1:1–8 (2017) The TREM2-APOE pathway drives the transcriptional phenotype 93. De Groef L, Dekeyster E, Geeraerts E, Lefevere E, Stalmans I, Salinas of dysfunctional microglia in neurodegenerative diseases. Immunity Navarro M et al (2016) Differential visual system organization and 47:566–581 susceptibility to experimental models of optic neuropathies in three 72. Sousa C, Golebiewska A, Poovathingal SK, Kaoma T, Pires-Afonso commonly used mouse strains. Exp Eye Res 145:235–247 Y, Martina S et al (2018) Single-cell transcriptomics reveals distinct 94. Lefevere E, Salinas-Navarro M, Andries L, Noterdaeme L, Etienne I, Van inflammation-induced microglia signatures. EMBO Rep 19:e46171 WE et al (2020) Tightening the retinal glia limitans attenuates neuroin- 73. Lodder C, Scheyltjens I, Stancu IC, Botella Lucena P, Gutiérrez de Ravé flammation after optic nerve injury. Glia 68:2643–2660 M, Vanherle S et al (2021) CSF1R inhibition rescues tau pathology and 95. Devoldere J, Peynshaert K, Dewitte H, Vanhove C, De Groef L, Moons neurodegeneration in an A/T/N model with combined AD pathologies, L et al (2019) Non-viral delivery of chemically modified mRNA to the while preserving plaque associated microglia. Acta Neuropathol Com- retina: subretinal versus intravitreal administration. J Control Release mun 9:1–21 307:315–330 74. Beuker C, Schafflick D, Strecker JK, Heming M, Li X, Wolbert J et al (2022) 96. Pernet V, Schwab ME (2012) The role of Nogo-A in axonal plasticity, Stroke induces disease-specific myeloid cells in the brain parenchyma regrowth and repair. Cell Tissue Res 349:97–104 and pia. Nat Commun 13:1–14 97. Wu YE, Pan L, Zuo Y, Li X, Hong W (2017) Detecting activated cell popu- 75. Reyes NJ, Mathew R, Saban DR (2019) Fate mapping in vivo to lations using single-cell RNA-Seq. Neuron 96:313–329 distinguish bona fide microglia versus recruited monocyte-derived 98. Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D et al macrophages in retinal disease. Methods Mol Biol 1834:153–164 (2003) Cytoscape: a software environment for integrated models of 76. O’Koren EG, Mathew R, Saban DR (2016) Fate mapping reveals biomolecular interaction networks. Genome Res 13:2498–2504 that microglia and recruited monocyte-derived macrophages are 99. Davis BM, Salinas-Navarro M, Cordeiro MF, Moons L, De GL (2017) Char- definitively distinguishable by phenotype in the retina. Sci Rep acterizing microglia activation: a spatial statistics approach to maximize 6:20636–20648 information extraction. Sci Rep 7:1–12 77. Wieghofer P, Hagemeyer N, Sankowski R, Schlecht A, Staszewski O, 100. Schneider CA, Rasband WS, Eliceiri KW (2012) NIH image to ImageJ: 25 Amann L et al (2021) Mapping the origin and fate of myeloid cells years of image analysis. Nat Methods 9:671–675 in distinct compartments of the eye by single-cell profiling. EMBO J 101. Leon S, Yin Y, Nguyen J, Irwin N, Benowitz LI (2000) Lens injury 40:e105123 stimulates axon regeneration in the mature rat optic nerve. J Neurosci 78. Yin Y, Cui Q, Li Y, Irwin N, Fischer D, Harvey AR et al (2003) Macrophage- 20:4615–4626 derived factors stimulate optic nerve regeneration. J Neurosci 23:2284–2293 Publisher’s Note 79. Gensel JC, Nakamura S, Guan Z, Van RN, Ankeny DP, Popovich PG (2009) Springer Nature remains neutral with regard to jurisdictional claims in pub- Macrophages promote axon regeneration with concurrent neurotoxic- lished maps and institutional affiliations. ity. J Neurosci 29:3956–3968 80. Kigerl KA, Gensel JC, Ankeny DP, Alexander JK, Donnelly DJ, Popovich PG (2009) Identification of two distinct macrophage subsets with diver - gent effects causing either neurotoxicity or regeneration in the injured mouse spinal cord. J Neurosci 29:13435–13444 81. Rolls A, Shechter R, London A, Segev Y, Jacob-Hirsch J, Amariglio N et al (2008) Two faces of chondroitin sulfate proteoglycan in spinal cord repair: a role in microglia/macrophage activation. PLoS Med 5:1262–1277 82. David S, Greenhalgh AD, Kroner A (2015) Macrophage and microglial plasticity in the injured spinal cord. Neuroscience 307:311–318 83. Schwartz M, Ramos JMP, Ben-Yehuda H (2020) A 20-Year journey from axonal injury to neurodegenerative diseases and the prospect of immu- notherapy for combating Alzheimer’s disease. J Immunol 204:243–250 84. Sas AR, Carbajal KS, Jerome AD, Menon R, Yoon C, Kalinski AL et al (2020) A new neutrophil subset promotes CNS neuron survival and axon regeneration. Nat Immunol 21:1496–1505 85. Batlle M, Ferri L, Andrade C, Ortega F-J, Vidal-Taboada JM, Pugliese M Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : et al (2015) Astroglia-microglia cross talk during neurodegeneration in the rat hippocampus. Biomed Res Int 2015:102419–102434 fast, convenient online submission 86. Kirkley KS, Popichak KA, Afzali MF, Legare ME, Tjalkens RB (2017) thorough peer review by experienced researchers in your ﬁeld Microglia amplify inflammatory activation of astrocytes in manganese neurotoxicity. J Neuroinflamm 14:1–17 rapid publication on acceptance 87. Xu C, Su Z (2015) Identification of cell types from single-cell transcrip - support for research data, including large and complex data types tomes using a novel clustering method. Bioinformatics 31:1974–1980 • gold Open Access which fosters wider collaboration and increased citations 88. Van Dyck A, Bollaerts I, Beckers A, Vanhunsel S, Glorian N, Houcke J et al (2021) Müller glia–myeloid cell crosstalk accelerates optic nerve maximum visibility for your research: over 100M website views per year regeneration in the adult zebrafish. Glia 69:1444–1463 89. Gambarotta G, Fregnan F, Gnavi S, Perroteau I (2013) Neuregulin 1 At BMC, research is always in progress. role in Schwann cell regulation and potential applications to promote Learn more biomedcentral.com/submissions peripheral nerve regeneration. Int Rev Neurobiol 108:223–256 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Acta Neuropathologica Communications Springer Journals http://www.deepdyve.com/lp/springer-journals/immune-stimulation-recruits-a-subset-of-pro-regenerative-macrophages-gWOS3Ho0Db

Loading next page...

References (109)

C. Lodder, Isabelle Scheyltjens, I. Stancu, Pablo Lucena, Manuel Ravé, Sarah Vanherle, T. Vanmierlo, Niels Cremers, Hannah Vanrusselt, B. Brône, B. Hanseeuw, J. Octave, A. Bottelbergs, K. Movahedi, I. Dewachter (2021)
CSF1R inhibition rescues tau pathology and neurodegeneration in an A/T/N model with combined AD pathologies, while preserving plaque associated microglia
Acta Neuropathologica Communications, 9
Y Yin, Q Cui, Y Li, N Irwin, D Fischer, AR Harvey (2003)
Macrophage-derived factors stimulate optic nerve regeneration
J Neurosci, 23
R. Shechter, Omer Miller, Gili Yovel, N. Rosenzweig, A. London, J. Ruckh, Ki-Wook Kim, E. Klein, V. Kalchenko, P. Bendel, S. Lira, Steffen Jung, M. Schwartz (2013)
Recruitment of beneficial M2 macrophages to injured spinal cord is orchestrated by remote brain choroid plexus.
Immunity, 38 3
S. Yona, Ki-Wook Kim, Yochai Wolf, A. Mildner, Diana Varol, Michal Breker, D. Strauss‐Ayali, Sergey Viukov, M. Guilliams, A. Misharin, D. Hume, H. Perlman, B. Malissen, E. Zelzer, Steffen Jung (2013)
Fate mapping reveals origins and dynamics of monocytes and tissue macrophages under homeostasis.
Immunity, 38 1
J. Rosenstein, N. Mani, A. Khaibullina, J. Krum (2003)
Neurotrophic Effects of Vascular Endothelial Growth Factor on Organotypic Cortical Explants and Primary Cortical Neurons
The Journal of Neuroscience, 23
J. Gensel, Bei Zhang (2015)
Macrophage activation and its role in repair and pathology after spinal cord injury
Brain Research, 1619
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations
F. Bennett, Mariko Bennett, Fazeela Yaqoob, Sara Mulinyawe, Gerald Grant, Melanie Gephart, E. Plowey, Ben Barres (2018)
A Combination of Ontogeny and CNS Environment Establishes Microglial Identity
Neuron, 98
S. Aibar, C. González-Blas, Thomas Moerman, V. Huynh-Thu, Hana Imrichová, Gert Hulselmans, F. Rambow, J. Marine, P. Geurts, J. Aerts, Joost Oord, Z. Atak, J. Wouters, S. Aerts (2017)
SCENIC: Single-cell regulatory network inference and clustering
Nature methods, 14
Anat Shemer, Isabelle Scheyltjens, G. Frumer, Jung-Seok Kim, Jonathan Grozovski, Serkalem Ayanaw, B. Dassa, Hannah Hove, Louise Chappell-Maor, S. Boura-Halfon, D. Leshkowitz, W. Mueller, N. Maggio, K. Movahedi, Steffen Jung (2020)
Interleukin-10 Prevents Pathological Microglia Hyperactivation following Peripheral Endotoxin Challenge.
Immunity
E. Lefevere, M. Salinas-Navarro, Lien Andries, L. Noterdaeme, Isabelle Etienne, E. Wonterghem, S. Vinckier, B. Davis, T. Bergen, I. Hove, K. Movahedi, R. Vandenbroucke, L. Moons, L. Groef (2020)
Tightening the retinal glia limitans attenuates neuroinflammation after optic nerve injury
Glia, 68
Bram Sande, Christopher Flerin, K. Davie, M. Waegeneer, Gert Hulselmans, S. Aibar, R. Seurinck, Wouter Saelens, Robrecht Cannoodt, Quentin Rouchon, T. Verbeiren, D. Maeyer, J. Reumers, Y. Saeys, S. Aerts (2020)
A scalable SCENIC workflow for single-cell gene regulatory network analysis
Nature Protocols, 15
Carole Sousa, A. Gołȩbiewska, S. Poovathingal, T. Kaoma, Y. Pires-Afonso, S. Martina, D. Coowar, F. Azuaje, A. Skupin, R. Balling, K. Biber, S. Niclou, A. Michelucci (2018)
Single‐cell transcriptomics reveals distinct inflammation‐induced microglia signatures
EMBO Reports, 19
T. Goldmann, Peter Wieghofer, P. Müller, Yochai Wolf, Diana Varol, S. Yona, S. Brendecke, Katrin Kierdorf, O. Staszewski, M. Datta, T. Luedde, M. Heikenwalder, Steffen Jung, M. Prinz (2013)
A new type of microglia gene targeting shows TAK1 to be pivotal in CNS autoimmune inflammation
Nature Neuroscience, 16
Karen Vlaminck, Hannah Hove, Daliya Kancheva, Isabelle Scheyltjens, Ana Antunes, Jonathan Bastos, Mónica Vara-Pérez, Leen Ali, M. Mampay, L. Deneyer, Juliana Miranda, Ruiyao Cai, L. Bouwens, Dimitri Bundel, G. Caljon, B. Stijlemans, A. Massie, J. Ginderachter, R. Vandenbroucke, K. Movahedi (2022)
Differential plasticity and fate of brain-resident and recruited macrophages during the onset and resolution of neuroinflammation.
Immunity
Takuji Kurimoto, Yuqin Yin, Ghaith Habboub, Hui-Ya Gilbert, Yiqing Li, S. Nakao, A. Hafezi-Moghadam, L. Benowitz (2013)
Neutrophils Express Oncomodulin and Promote Optic Nerve Regeneration
The Journal of Neuroscience, 33
R. Shechter, M. Schwartz (2013)
Harnessing monocyte‐derived macrophages to control central nervous system pathologies: no longer ‘if’ but ‘how’
The Journal of Pathology, 229
A. Deczkowska, Hadas Keren-Shaul, A. Weiner, M. Colonna, M. Schwartz, I. Amit (2018)
Disease-Associated Microglia: A Universal Immune Sensor of Neurodegeneration
Cell, 173
Qingyun Li, Zuolin Cheng, Lu Zhou, S. Darmanis, N. Neff, Jennifer Okamoto, G. Gulati, Mariko Bennett, Lu Sun, L. Clarke, J. Marschallinger, Guoqiang Yu, S. Quake, T. Wyss-Coray, B. Barres (2018)
Developmental Heterogeneity of Microglia and Brain Myeloid Cells Revealed by Deep Single-Cell RNA Sequencing
Neuron, 101
L. Benowitz, P. Popovich (2011)
Inflammation and axon regeneration.
Current opinion in neurology, 24 6
T Masuda, R Sankowski, O Staszewski, C Böttcher, L Amann (2019)
Spatial and temporal heterogeneity of mouse and human microglia at single-cell resolution
Nature, 566
Inbal Benhar, Jiarui Ding, Wenjun Yan, Irene Whitney, A. Jacobi, Malika Sud, Grace Burgin, K. Shekhar, N. Tran, Chen Wang, Zhigang He, J. Sanes, A. Regev (2022)
Temporal single-cell atlas of non-neuronal retinal cells reveals dynamic, coordinated multicellular responses to central nervous system injury
Nature Immunology, 24
James Cronk, A. Filiano, A. Louveau, Ioana Marin, R. Marsh, Emily Ji, Dylan Goldman, I. Smirnov, Nicholas Geraci, S. Acton, Christopher Overall, J. Kipnis (2018)
Peripherally derived macrophages can engraft the brain independent of irradiation and maintain an identity distinct from microglia
The Journal of Experimental Medicine, 215
Lili Xie, Ling-Ping Cen, Yiqing Li, Hui-Ya Gilbert, Oleksandr Strelko, C. Berlinicke, M. Stavarache, Madeline Ma, Yongting Wang, Q. Cui, M. Kaplitt, D. Zack, L. Benowitz, Yuqin Yin (2022)
Monocyte-derived SDF1 supports optic nerve regeneration and alters retinal ganglion cells’ response to Pten deletion
Proceedings of the National Academy of Sciences of the United States of America, 119
S. Safaiyan, Simon Besson-Girard, Tugberk Kaya, Ludovico Cantuti-Castelvetri, Lu Liu, Hao Ji, M. Schifferer, Garyfallia Gouna, Fumere Usifo, N. Kannaiyan, D. Fitzner, X. Xiang, M. Rossner, M. Brendel, O. Gokce, M. Simons (2021)
White matter aging drives microglial diversity
Neuron, 109
Ke Wei, V. Serpooshan, C. Hurtado, Marta Diez-Cuñado, Mingming Zhao, Sonomi Maruyama, Wenhong Zhu, G. Fajardo, M. Noseda, Kazuto Nakamura, Xueying Tian, Qiaozhen Liu, Andrew Wang, Y. Matsuura, P. Bushway, Wenqing Cai, A. Savchenko, M. Mahmoudi, M. Schneider, M. Hoff, M. Butte, Phillip Yang, K. Walsh, Bin Zhou, D. Bernstein, M. Mercola, P. Ruiz‐Lozano (2015)
Epicardial FSTL1 reconstitution regenerates the adult mammalian heart
Nature, 525
C. Frigerio, L. Wolfs, Nicola Fattorelli, N. Thrupp, I. Voytyuk, Inga Schmidt, R. Mancuso, Wei-Ting Chen, Maya Woodbury, G. Srivastava, T. Möller, E. Hudry, Sudeshna Das, T. Saido, E. Karran, B. Hyman, V. Perry, M. Fiers, B. Strooper (2019)
The Major Risk Factors for Alzheimer’s Disease: Age, Sex, and Genes Modulate the Microglia Response to Aβ Plaques
Cell reports, 27
Milica Margeta, Zhuoran Yin, C. Madore, Kristen Pitts, Sophia Letcher, Jing Tang, Shuhong Jiang, C. Gauthier, Sebastian Silveira, Caitlin Schroeder, E. Lad, A. Proia, R. Tanzi, D. Holtzman, S. Krasemann, D. Chen, O. Butovsky (2022)
Apolipoprotein E4 impairs the response of neurodegenerative retinal microglia and prevents neuronal loss in glaucoma.
Immunity
S. Srinivas, Tomoko Watanabe, Chyuan-Sheng Lin, C. William, Y. Tanabe, T. Jessell, F. Costantini (2001)
Cre reporter strains produced by targeted insertion of EYFP and ECFP into the ROSA26 locus
BMC Developmental Biology, 1
A. Dyck, I. Bollaerts, An Beckers, Sophie Vanhunsel, Nynke Glorian, J. houcke, T. Ham, L. Groef, Lien Andries, L. Moons (2021)
Müller glia–myeloid cell crosstalk accelerates optic nerve regeneration in the adult zebrafish
Glia, 69
S. Krasemann, C. Madore, R. Cialic, Caroline Baufeld, Narghes Calcagno, R. Fatimy, L. Beckers, E. O’Loughlin, Yang Xu, Zain Fanek, David Greco, Scott Smith, G. Tweet, Zachary Humulock, T. Zrzavy, Patricia Conde-Sanroman, Mar Gacias, Z. Weng, Hao Chen, Emily Tjon, F. Mazaheri, K. Hartmann, A. Madi, J. Ulrich, M. Glatzel, A. Worthmann, J. Heeren, B. Budnik, C. Lemere, T. Ikezu, F. Heppner, V. Litvak, D. Holtzman, H. Lassmann, H. Weiner, J. Ochando, C. Haass, O. Butovsky (2017)
The TREM2-APOE Pathway Drives the Transcriptional Phenotype of Dysfunctional Microglia in Neurodegenerative Diseases.
Immunity, 47 3
Emily O’Koren, Chen Yu, Mikael Klingeborn, Alicia Wong, Cameron Prigge, R. Mathew, J. Kalnitsky, Rasha Msallam, Aymeric Silvin, J. Kay, C. Rickman, V. Arshavsky, F. Ginhoux, M. Merad, D. Saban (2019)
Microglial Function Is Distinct in Different Anatomical Locations during Retinal Homeostasis and Degeneration
Immunity, 50
L. Groef, E. Dekeyster, E. Geeraerts, E. Lefevere, I. Stalmans, M. Salinas-Navarro, L. Moons (2016)
Differential visual system organization and susceptibility to experimental models of optic neuropathies in three commonly used mouse strains.
Experimental eye research, 145
Antoine Drieu, Siling Du, Steffen Storck, Justin Rustenhoven, Zachary Papadopoulos, Taitea Dykstra, Fenghe Zhong, Kyungdeok Kim, S. Blackburn, Tornike Mamuladze, O. Harari, C. Karch, R. Bateman, R. Perrin, M. Farlow, J. Chhatwal, Jared Benzing, J. Brosch, Jill Buck, M. Farlow, B. Ghetti, S. Adams, Nicolas Barthélemy, T. Benzinger, S. Brandon, V. Buckles, Lisa Cash, Charlie Chen, J. Chua, C. Cruchaga, Darcy Denner, Aylin Dincer, Tamara Donahue, A. Fagan, B. Feldman, S. Flores, E. Franklin, N. Joseph-Mathurin, Alyssa Gonzalez, B. Gordon, Julia Gray, Emily Gremminger, Alexandrer Groves, J. Hassenstab, Cortaiga Hellm, Elizabeth Herries, Laura Hoechst-Swisher, D. Holtzman, R. Hornbeck, Gina Jerome, S. Keefe, Deborah Koudelis, Yan Li, Jacob Marsh, Rita Martinez, K. Mawuenyega, Austin McCullough, E. McDade, J. Morris, J. Norton, Kristine Shady, Wendy Sigurdson, Jennifer Smith, Peter Wang, Qing Wang, C. Xiong, Jinbin Xu, Xiong Xu, R. Allegri, P. Mendez, Noelia Egido, Akihiko Araki, T. Ikeuchi, K. Ishii, K. Kasuga, Jacob Bechara, W. Brooks, P. Schofield, S. Berman, S. Goldberg, S. Ikonomović, W. Klunk, Oscar Lopez, J. Mountz, N. Nadkarni, Riddhi Patira, Lori Smith, B. Snitz, S. Thompson, E. Weamer, Courtney Bodge, S. Salloway, Kathleen Carter, Duc Duong, Erik Johnson, Allan Levey, Lingyan Ping, N. Seyfried, Colleen Fitzpatrick, Helena Chui, J. Ringman, G. Day, N. Graff-Radford, M. Graham, Sochenda Stephens, Chrismary Cruz, J. Goldman, Arlene Mejia, Katie Neimeyer, James Noble, Anna Diffenbacher, I. Yakushev, J. Levin, J. Vöglein, J. Douglas, Nick Fox, M. Grilo, C. Mummery, Antoinette O’Connor, Bianca Esposito, Alison Goate, A. Renton, H. Fujii, M. Senda, H. Shimada, S. Gardener, Ralph Martins, H. Sohrabi, K. Taddei, Susanne Gräber‐Sultan, Lisa Häsler, Anna Hofmann, M. Jucker, S. Käser, Elke Kuder-Buletta, C. Laske, O. Preische, C. Haass, E. Morenas-Rodriguez, B. Nuscher, R. Ihara, Akem Nagamatsu, Y. Niimi, Clifford Jack, R. Koeppe, N. Mason, C. Masters, U. Obermüller, Song Hu, G. Randolph, I. Smirnov, J. Kipnis (2022)
Parenchymal border macrophages regulate the flow dynamics of the cerebrospinal fluid
Nature, 611
S. David, A. Kroner (2011)
Repertoire of microglial and macrophage responses after spinal cord injury
Nature Reviews Neuroscience, 12
J. Gensel, Satoshi Nakamura, Z. Guan, N. Rooijen, Daniel Ankeny, P. Popovich (2009)
Macrophages Promote Axon Regeneration with Concurrent Neurotoxicity
The Journal of Neuroscience, 29
A. London, É. Itskovich, Inbal Benhar, V. Kalchenko, M. Mack, Steffen Jung, M. Schwartz (2011)
Neuroprotection and progenitor cell renewal in the injured adult murine retina requires healing monocyte-derived macrophages
The Journal of Experimental Medicine, 208
A. Sas, K. Carbajal, A. Jerome, R. Menon, C. Yoon, Ashley Kalinski, R. Giger, B. Segal (2020)
A novel neutrophil subset promotes CNS neuron survival and axon regeneration
Nature immunology, 21
Marco Leibinger, A. Andreadaki, P. Gobrecht, Evgeny Levin, H. Diekmann, D. Fischer (2016)
Boosting Central Nervous System Axon Regeneration by Circumventing Limitations of Natural Cytokine Signaling.
Molecular therapy : the journal of the American Society of Gene Therapy, 24 10
Montserrat Batlle, Lorenzo Ferri, C. Andrade, Francisco Ortega, J. Vidal-Taboada, M. Pugliese, N. Mahy, M. Rodríguez (2015)
Astroglia-Microglia Cross Talk during Neurodegeneration in the Rat Hippocampus
BioMed Research International, 2015
D. Selkoe (2011)
Alzheimer's disease.
Cold Spring Harbor perspectives in biology, 3 7
Hannah Hove, L. Martens, Isabelle Scheyltjens, Karen Vlaminck, Ana Antunes, Sofie Prijck, N. Vandamme, Sebastiaan Schepper, G. Isterdael, Charlotte Scott, J. Aerts, G. Berx, G. Boeckxstaens, R. Vandenbroucke, L. Vereecke, D. Moechars, M. Guilliams, J. Ginderachter, Y. Saeys, K. Movahedi (2019)
A single-cell atlas of mouse brain macrophages reveals unique transcriptional identities shaped by ontogeny and tissue environment
Nature Neuroscience, 22
A. Jacobi, N. Tran, Wenjun Yan, Inbal Benhar, Feng Tian, Rebecca Schaffer, Zhigang He, J. Sanes (2022)
Overlapping transcriptional programs promote survival and axonal regeneration of injured retinal ganglion cells
Neuron, 110
Chenying Guo, Kin-Sang Cho, Yingqian Li, Kissauo Tchedre, C. Antolik, Jie Ma, Justin Chew, T. Utheim, Xizhong Huang, Honghua Yu, M. Malik, Nada Anzak, D. Chen (2018)
IGFBPL1 Regulates Axon Growth through IGF-1-mediated Signaling Cascades
Scientific Reports, 8
Bhagat Singh, Jason Plemel (2014)
Neutrophil Contribution in Facilitating Optic Nerve Regeneration
The Journal of Neuroscience, 34
A. Rolls, R. Shechter, A. London, Y. Segev, J. Jacob-Hirsch, N. Amariglio, G. Rechavi, M. Schwartz (2008)
Two Faces of Chondroitin Sulfate Proteoglycan in Spinal Cord Repair: A Role in Microglia/Macrophage Activation
PLoS Medicine, 5
A. Mueller, X. Pedré, T. Stempfl, I. Kleiter, S. Couillard-Després, L. Aigner, G. Giegerich, A. Steinbrecher (2008)
Novel role for SLPI in MOG-induced EAE revealed by spinal cord expression analysis
Journal of Neuroinflammation, 5
Chen Xu, Z. Su (2015)
Identification of cell types from single-cell transcriptomes using a novel clustering method
Bioinformatics, 31 12
V. Pernet, M. Schwab (2012)
The role of Nogo-A in axonal plasticity, regrowth and repair
Cell and Tissue Research, 349
C. Beuker, D. Schafflick, Jan-Kolja Strecker, M. Heming, Xiaolin Li, Jolien Wolbert, Antje Schmidt-Pogoda, Christian Thomas, T. Kuhlmann, Irene Aranda-Pardos, Noelia A-Gonzalez, Praveen Kumar, Yves Werner, E. Kılıç, D. Hermann, H. Wiendl, R. Stumm, G. Hörste, J. Minnerup (2022)
Stroke induces disease-specific myeloid cells in the brain parenchyma and pia
Nature Communications, 13
Soyoung Choi, Li Guo, M. Cordeiro (2021)
Retinal and Brain Microglia in Multiple Sclerosis and Neurodegeneration
Cells, 10
M. Jordão, R. Sankowski, S. Brendecke, Sagar, G. Locatelli, Yi-Heng Tai, Tuan Tay, Eva Schramm, S. Armbruster, N. Hagemeyer, O. Gross, Dominic Mai, Özgün Çiçek, Thorsten Falk, M. Kerschensteiner, Dominic Grün, M. Prinz (2019)
Single-cell profiling identifies myeloid cell subsets with distinct fates during neuroinflammation
Science, 363
Lien Andries, L. Masin, M. Navarro, S. Zaunz, Marie Claes, Steven Bergmans, Véronique Brouwers, E. Lefevere, C. Verfaillie, K. Movahedi, L. Groef, L. Moons (2021)
MMP2 Modulates Inflammatory Response during Axonal Regeneration in the Murine Visual System
Cells, 10
Kelly Kirkley, Katriana Popichak, Maryam Afzali, M. Légaré, R. Tjalkens (2017)
Microglia amplify inflammatory activation of astrocytes in manganese neurotoxicity
Journal of Neuroinflammation, 14
T. Hauk, A. Müller, Jieun Lee, R. Schwendener, D. Fischer (2008)
Neuroprotective and axon growth promoting effects of intraocular inflammation do not depend on oncomodulin or the presence of large numbers of activated macrophages
Experimental Neurology, 209
Xiaoying Chen, D. Holtzman (2022)
Emerging roles of innate and adaptive immunity in Alzheimer's disease.
Immunity
Chen Yu, C. Roubeix, F. Sennlaub, D. Saban (2020)
Microglia versus Monocytes: Distinct Roles in Degenerative Diseases of the Retina
Trends in Neurosciences, 43
Peter Wieghofer, N. Hagemeyer, R. Sankowski, A. Schlecht, O. Staszewski, Lukas Amann, M. Gruber, Jana Koch, Annika Hausmann, Peipei Zhang, S. Boneva, Takahiro Masuda, I. Hilgendorf, T. Goldmann, C. Böttcher, J. Priller, F. Rossi, C. Lange, M. Prinz (2021)
Mapping the origin and fate of myeloid cells in distinct compartments of the eye by single‐cell profiling
The EMBO Journal, 40
Qi Cui, Y. Yin, Larry Benowitz (2009)
The role of macrophages in optic nerve regeneration
Neuroscience, 158
Eric Bray, Benjamin Yungher, K. Levay, Márcio Ribeiro, G. Dvoryanchikov, A. Ayupe, Kinjal Thakor, Victoria Marks, M. Randolph, M. Danzi, Tiffany Schmidt, Nirupa Chaudhari, V. Lemmon, S. Hattar, Kevin Park (2019)
Thrombospondin-1 Mediates Axon Regeneration in Retinal Ganglion Cells
Neuron, 103
N. Reyes, R. Mathew, D. Saban (2019)
Fate Mapping In Vivo to Distinguish Bona Fide Microglia Versus Recruited Monocyte-Derived Macrophages in Retinal Disease.
Methods in molecular biology, 1834
G. Gambarotta, F. Fregnan, S. Gnavi, I. Perroteau (2013)
Neuregulin 1 role in Schwann cell regulation and potential applications to promote peripheral nerve regeneration.
International review of neurobiology, 108
S. Wattananit, D. Tornero, N. Graubardt, Tamar Memanishvili, Emanuela Monni, J. Tatarishvili, G. Miskinyte, Ruimin Ge, H. Ahlenius, O. Lindvall, M. Schwartz, Z. Kokaia (2016)
Monocyte-Derived Macrophages Contribute to Spontaneous Long-Term Functional Recovery after Stroke in Mice
The Journal of Neuroscience, 36
Kimberly Wong, L. Benowitz (2022)
Retinal Ganglion Cell Survival and Axon Regeneration after Optic Nerve Injury: Role of Inflammation and Other Factors
International Journal of Molecular Sciences, 23
Yuqin Yin, Q. Cui, Hui-Ya Gilbert, Yang Yang, Zhiyong Yang, C. Berlinicke, Zhi-wai Li, Camila Zaverucha-do-Valle, Huamei He, V. Petkova, D. Zack, L. Benowitz (2009)
Oncomodulin links inflammation to optic nerve regeneration
Proceedings of the National Academy of Sciences, 106
Yuqin Yin, M. Henzl, B. Lorber, T. Nakazawa, T. Thomas, F. Jiang, R. Langer, L. Benowitz (2006)
Oncomodulin is a macrophage-derived signal for axon regeneration in retinal ganglion cells
Nature Neuroscience, 9
T. Edwards, Marc Hammarlund (2014)
Syndecan promotes axon regeneration by stabilizing growth cone migration.
Cell reports, 8 1
Elisabeth Müller (2021)
Muller
Encyclopedic Dictionary of Archaeology
Robin Browaeys, Wouter Saelens, Y. Saeys (2019)
NicheNet: modeling intercellular communication by linking ligands to target genes
Nature Methods, 17
Edinburgh Research Explorer Spatial and temporal heterogeneity of mouse and human microglia at single-cell resolution
Anat Shemer, Jonathan Grozovski, Tuan Tay, Jenhan Tao, Alon Volaski, P. Süß, A. Ardura-Fabregat, Mor Gross-Vered, Jung-Seok Kim, Eyal David, Louise Chappell-Maor, L. Thielecke, C. Glass, K. Cornils, M. Prinz, Steffen Jung (2018)
Engrafted parenchymal brain macrophages differ from microglia in transcriptome, chromatin landscape and response to challenge
Nature Communications, 9
K. Rashid, Isha Akhtar-Schaefer, T. Langmann (2019)
Microglia in Retinal Degeneration
Frontiers in Immunology, 10
Emily O’Koren, R. Mathew, D. Saban (2016)
Fate mapping reveals that microglia and recruited monocyte-derived macrophages are definitively distinguishable by phenotype in the retina
Scientific Reports, 6
Yashashree Joshi, M. Sória, Giorgia Quadrato, Gizem Inak, Luming Zhou, A. Hervera, Khizr Rathore, M. Elnaggar, M. Cucchiarini, J. Marine, R. Puttagunta, S. Giovanni (2015)
The MDM4/MDM2-p53-IGF1 axis controls axonal regeneration, sprouting and functional recovery after CNS injury.
Brain : a journal of neurology, 138 Pt 7
Steven Leon, Yuqin Yin, Jennifer Nguyen, N. Irwin, L. Benowitz (2000)
Lens Injury Stimulates Axon Regeneration in the Mature Rat Optic Nerve
The Journal of Neuroscience, 20
N. Faust, F. Varas, Louise Kelly, S. Heck, T. Graf (2000)
Insertion of enhanced green fluorescent protein into the lysozyme gene creates mice with green fluorescent granulocytes and macrophages.
Blood, 96 2
Y. Wu, Lin Pan, Yanning Zuo, Xinmin Li, Weizhe Hong (2017)
Detecting Activated Cell Populations Using Single-Cell RNA-Seq
Neuron, 96
L. Boring, J. Gosling, S. Chensue, S. Kunkel, Robert Farese, H. Broxmeyer, I. Charo (1997)
Impaired monocyte migration and reduced type 1 (Th1) cytokine responses in C-C chemokine receptor 2 knockout mice.
The Journal of clinical investigation, 100 10
Timothy Hammond, Connor Dufort, Lasse Dissing‐Olesen, Stefanie Giera, Adam Young, Alec Wysoker, A. Walker, F. Gergits, Michael Segel, J. Nemesh, S. Marsh, Arpiar Saunders, Evan Macosko, F. Ginhoux, Jinmiao Chen, R. Franklin, X. Piao, S. Mccarroll, B. Stevens (2019)
Single‐Cell RNA Sequencing of Microglia throughout the Mouse Lifespan and in the Injured Brain Reveals Complex Cell‐State Changes
Immunity, 50
Kristina Kigerl, J. Gensel, Daniel Ankeny, Jessica Alexander, Dustin Donnelly, P. Popovich (2009)
Identification of Two Distinct Macrophage Subsets with Divergent Effects Causing either Neurotoxicity or Regeneration in the Injured Mouse Spinal Cord
The Journal of Neuroscience, 29
N. Ghasemlou, D. Bouhy, Jingxuan Yang, Rubèn López-Vales, M. Haber, T. Thuraisingam, Guoan He, D. Radzioch, A. Ding, S. David (2010)
Beneficial effects of secretory leukocyte protease inhibitor after spinal cord injury.
Brain : a journal of neurology, 133 Pt 1
J. Devoldere, K. Peynshaert, H. Dewitte, C. Vanhove, L. Groef, L. Moons, Sinem Özcan, D. Dalkara, S. Smedt, K. Remaut (2019)
Non-viral delivery of chemically modified mRNA to the retina: Subretinal versus intravitreal administration.
Journal of controlled release : official journal of the Controlled Release Society
Lien Andries, L. Groef, L. Moons (2020)
Neuroinflammation and Optic Nerve Regeneration: Where Do We Stand in Elucidating Underlying Cellular and Molecular Players?
Current Eye Research, 45
M. Prinz, Takahiro Masuda, Michael Wheeler, F. Quintana (2021)
Microglia and Central Nervous System-Associated Macrophages-From Origin to Disease Modulation.
Annual review of immunology
N. Tran, K. Shekhar, Irene Whitney, A. Jacobi, Inbal Benhar, Guosong Hong, Wenjun Yan, X. Adiconis, McKinzie Arnold, Jung Lee, J. Levin, Dingchang Lin, Chen Wang, Charles Lieber, A. Regev, Zhigang He, J. Sanes (2019)
Single-Cell Profiles of Retinal Ganglion Cells Differing in Resilience to Injury Reveal Neuroprotective Genes
Neuron, 104
S Aibar, CB González-Blas, T Moerman, VA Huynh-Thu, H Imrichova, G Hulselmans (2017)
SCENIC: single-cell regulatory network inference and clustering
Nat Methods, 14
H. Togt (2003)
Publisher's Note
J. Netw. Comput. Appl., 26
M. Schwartz, Javier Ramos, H. Ben-Yehuda (2020)
A 20-Year Journey from Axonal Injury to Neurodegenerative Diseases and the Prospect of Immunotherapy for Combating Alzheimer’s Disease
The Journal of Immunology, 204
Sarah Anderson, Jacqueline Roberts, Jianmin Zhang, Michael Steele, Cesar Romero, Alejandra Bosco, M. Vetter (2019)
Developmental Apoptosis Promotes a Disease-Related Gene Signature and Independence from CSF1R Signaling in Retinal Microglia
Cell reports, 27
B. Davis, Manual Salinas-Navarro, M. Cordeiro, L. Moons, L. Groef (2017)
Characterizing microglia activation: a spatial statistics approach to maximize information extraction
Scientific Reports, 7
C. Schneider, W. Rasband, K. Eliceiri (2012)
NIH Image to ImageJ: 25 years of image analysis
Nature Methods, 9
A. Hilla, H. Diekmann, D. Fischer (2017)
Microglia Are Irrelevant for Neuronal Degeneration and Axon Regeneration after Acute Injury
The Journal of Neuroscience, 37
T. Hauk, Marco Leibinger, A. Müller, A. Andreadaki, U. Knippschild, D. Fischer (2010)
Stimulation of axon regeneration in the mature optic nerve by intravitreal application of the toll-like receptor 2 agonist Pam3Cys.
Investigative ophthalmology & visual science, 51 1
Samuel David, A. Greenhalgh, Antje Kroner (2015)
Macrophage and microglial plasticity in the injured spinal cord
Neuroscience, 307
Paul Shannon, Andrew Markiel, Owen Ozier, N. Baliga, Jonathan Wang, Daniel Ramage, Nada Amin, Benno Schwikowski, T. Ideker (2003)
Cytoscape: a software environment for integrated models of biomolecular interaction networks.
Genome research, 13 11
C Guo (2018)
1
Sci Rep, 8
Dasfne Lee-Liu, Gabriela Edwards-Faret, Victor Tapia, J. Larraín (2013)
Spinal cord regeneration: Lessons for mammals from non‐mammalian vertebrates
genesis, 51
F. Fricker, N. Lago, Sharmili Balarajah, C. Tsantoulas, Shamil Tanna, N. Zhu, S. Fageiry, M. Jenkins, A. Garratt, C. Birchmeier, D. Bennett (2011)
Axonally Derived Neuregulin-1 Is Required for Remyelination and Regeneration after Nerve Injury in Adulthood
The Journal of Neuroscience, 31
C. Taggart, S. Cryan, S. Weldon, A. Gibbons, C. Greene, E. Kelly, T. Low, S. O’Neill, N. McElvaney (2005)
Secretory leucoprotease inhibitor binds to NF-κB binding sites in monocytes and inhibits p65 binding
The Journal of Experimental Medicine, 202
S Böcker-Meffert (2002)
2021
IOVS, 43
K. Murakami, T. Tanaka, Y. Bando, S. Yoshida (2015)
Nerve injury induces the expression of syndecan-1 heparan sulfate proteoglycan in primary sensory neurons
Neuroscience, 300
M. Guilliams, F. Svedberg (2021)
Does tissue imprinting restrict macrophage plasticity?
Nature Immunology, 22
H. Diekmann, Marco Leibinger, D. Fischer (2013)
Do growth-stimulated retinal ganglion cell axons find their central targets after optic nerve injury? New insights by three-dimensional imaging of the visual pathway
Experimental Neurology, 248
Xin Duan, Mu Qiao, F. Bei, In-Jung Kim, Zhigang He, J. Sanes (2015)
Subtype-Specific Regeneration of Retinal Ganglion Cells following Axotomy: Effects of Osteopontin and mTOR Signaling
Neuron, 85
S. Heindl, B. Gesierich, Corinne Benakis, G. Llovera, M. Duering, A. Liesz (2018)
Automated Morphological Analysis of Microglia After Stroke
Frontiers in Cellular Neuroscience, 12
R. Shechter, A. London, C. Varol, Catarina Rapôso, Melania Cusimano, Gili Yovel, A. Rolls, M. Mack, S. Pluchino, G. Martino, Steffen Jung, M. Schwartz (2009)
Infiltrating Blood-Derived Macrophages Are Vital Cells Playing an Anti-inflammatory Role in Recovery from Spinal Cord Injury in Mice
PLoS Medicine, 6
Hadas Keren-Shaul, A. Spinrad, A. Weiner, O. Matcovitch-Natan, Raz Dvir-Szternfeld, Tyler Ulland, Eyal David, K. Baruch, David Lara-Astaiso, Beáta Tóth, S. Itzkovitz, M. Colonna, M. Schwartz, I. Amit (2017)
A Unique Microglia Type Associated with Restricting Development of Alzheimer’s Disease
Cell, 169
Ana Antunes, Isabelle Scheyltjens, F. Lodi, J. Messiaen, A. Antoranz, J. Duerinck, Daliya Kancheva, L. Martens, Karen Vlaminck, Hannah Hove, Signe Hansen, F. Bosisio, K. Borght, S. Vleeschouwer, R. Sciot, L. Bouwens, M. Verfaillie, N. Vandamme, R. Vandenbroucke, O. Wever, Y. Saeys, M. Guilliams, C. Gysemans, B. Neyns, F. Smet, D. Lambrechts, J. Ginderachter, K. Movahedi (2021)
Single-cell profiling of myeloid cells in glioblastoma across species and disease stage reveals macrophage competition and specialization
Nature Neuroscience, 24
S. Böcker-Meffert, P. Rosenstiel, C. Röhl, Nils Warneke, J. Held-Feindt, J. Sievers, R. Lucius (2002)
Erythropoietin and VEGF promote neural outgrowth from retinal explants in postnatal rats.
Investigative ophthalmology & visual science, 43 6

Publisher: Springer Journals
Copyright: Copyright © The Author(s) 2023
eISSN: 2051-5960
DOI: 10.1186/s40478-023-01580-3
Publisher site: See Article on Publisher Site

Abstract

The multifaceted nature of neuroinflammation is highlighted by its ability to both aggravate and promote neuronal health. While in mammals retinal ganglion cells (RGCs) are unable to regenerate following injury, acute inflammation can induce axonal regrowth. However, the nature of the cells, cellular states and signalling pathways that drive this inflammation-induced regeneration have remained elusive. Here, we investigated the functional significance of mac- rophages during RGC de- and regeneration, by characterizing the inflammatory cascade evoked by optic nerve crush (ONC) injury, with or without local inflammatory stimulation in the vitreous. By combining single-cell RNA sequenc- ing and fate mapping approaches, we elucidated the response of retinal microglia and recruited monocyte-derived macrophages (MDMs) to RGC injury. Importantly, inflammatory stimulation recruited large numbers of MDMs to the retina, which exhibited long-term engraftment and promoted axonal regrowth. Ligand-receptor analysis highlighted a subset of recruited macrophages that exhibited expression of pro-regenerative secreted factors, which were able to promote axon regrowth via paracrine signalling. Our work reveals how inflammation may promote CNS regenera- tion by modulating innate immune responses, providing a rationale for macrophage-centred strategies for driving neuronal repair following injury and disease. Lieve Moons and Kiavash Movahedi are co-senior author. *Correspondence: Lieve Moons lieve.moons@kuleuven.be Kiavash Movahedi kiavash.movahedi@vub.be Full list of author information is available at the end of the article © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 2 of 20 contribute to CNS de- and regeneration is the retina- Introduction brain connection. Over the past decades, multiple stud- The central nervous system (CNS) accommodates vari - ies, using the retinofugal pathway and the optic nerve ous populations of resident macrophages that are criti- crush (ONC) paradigm as a neurodegeneration model, cal regulators in brain development, homeostasis and have shown that induction of controlled inflamma - disease [1]. This includes microglia in the brain paren - tion in the retina induces retinal ganglion cell (RGC) chyma and border-associated macrophages (BAMs) in survival and axonal regrowth in rodents [26–32]. Pre- non-parenchymal border tissues. Microglia continuously vious research revealed that an inflammatory stimula - survey their microenvironment and interact with neu- tion results in infiltration of neutrophils and MDMs in rons to prune synapses, provide neurotrophic factors, both the vitreous and retina and leads to modulation of remove waste and sense danger [1]. Similarly, BAMs resident macrophages, as well as retinal astrocytes and play key roles in supporting healthy brain functions [2]. Müller glia [33, 34]. While it is established that an acute uTh s, resident CNS macrophages are highly special - inflammatory response is beneficial for survival and ized cells that play an active role in maintaining healthy axonal regrowth of damaged RGCs, controversy still brain physiology. Upon inflammation and disease, micro - prevails about which cells, cell states, molecules and glia and BAMs exit their homeostatic state and adopt pathways are functionally implicated. To investigate the new transcriptional modules. This has been thoroughly specific contribution of resident and recruited myeloid investigated for microglia during neurodegeneration cells during RGC de- and regeneration, we performed and subsequently in other disease and injury models, an in-depth characterization of the acute inflamma - where a specific disease-associated microglia (DAM) tory response evoked by optic nerve injury, with or state has been identified [3–12]. The disease-responses without a local inflammatory stimulation using a Toll- of microglia and BAMs can be beneficial or detrimen - like receptor 2 agonist. By combining single-cell RNA tal depending on the nature and/or stage of the disease. sequencing (scRNA-seq) and fate mapping approaches, For example, in Alzheimer’s disease DAMs may allevi- we elucidated the ontogeny, cell states and functional ate amyloid beta (Ab) pathology by compacting amyloid significance of the resident and recruited macrophage plaques but may also aggravate the disease following populations that react to RGC degeneration following the onset of tau pathology [13]. Importantly, inflamma - ONC injury. Our results show that inflammatory stim - tion and disease can also result in the recruitment of ulation recruits a subset of pro-regenerative MDMs to monocyte-derived macrophages (MDMs) to the CNS. the retina, which produce secreted proteins that can Emerging evidence indicates that recruited MDMs react promote axon regrowth of injured RGCs. differently to disease than resident macrophages [14–16]. Recruited MDMs may exert complementary functions Results and their interplay with resident brain macrophages can Injury to the optic nerve activates resident and recruited shape disease progression [15]. Therefore, inhibiting or myeloid cells in the retina promoting monocyte recruitment to the diseased brain To investigate the response of myeloid cells in the ret- may have significant therapeutic implications. ina to retinal ganglion cell (RGC) injury, we performed While the key role of brain macrophages in neurode- + + scRNA-seq on CD45 CD11b cells that were sorted generation is firmly established, their contribution to from healthy adult retinas or from retinas harvested at regeneration and repair has remained more elusive. Sev- 4 days post optic nerve crush (ONC) injury. The majority eral studies have reported that brain macrophages can + + of CD45 CD11b cells in healthy retinas were microglia, exert neuroprotective activities and contribute to healing identified based on their high expression of microglial and repair following neurodegeneration or CNS injury signature genes, including Sall1, Sparc and P2ry12 [17–20]. Different macrophage activation states, rang - (Fig. 1A, B). Naive retinas also contained small clusters of ing from more pro-inflammatory ones that are associ - + + Fcgr1 C1qa macrophages that did not express prototyp- ated with tissue damage, neuronal loss, axon retraction ical microglia genes, but exhibited enriched expression and demyelination to more anti-inflammatory pheno - of Ms4a7, Ms4a6c and Apoe (clusters BAM1-2) (Fig. 1A, types that are linked to neuroprotection [21] and axon B). Within the brain, this signature is associated with regrowth [22, 23], have been suggested to affect repair macrophages found in border tissues (border-associated after CNS injury [24, 25]. Nevertheless, while neuro- macrophages or BAMs), including the perivascular space protective and regenerative macrophage subtypes are [10, 35]. Therefore, these cells may represent retinal thought to exist, their molecular fingerprint remains BAMs, such as perivascular macrophages. We observed poorly characterized. two main clusters that showed differential expression One powerful model system to investigate the cellular of Mrc1, Cd163 and H2-Aa (Fig. 1B), reflecting BAM players and the molecules and signalling pathways that A ndries et al. Acta Neuropathologica Communications (2023) 11:85 Page 3 of 20 neurodegeneration. Mg3 represented a cluster of micro- heterogeneity in the brain [10]. Other C D11b cells in glia that expressed interferon (IFN)-induced genes (Addi- the naive retina were neutrophils (S100a9, Csf3r, Cd24a), tional File 1: Figure S1C) and was observed both in the classical monocytes (Ly6c2, Fn1), non-classical mono- naive and injured retina. These IFN response microglia cytes (Ear2, Ace), cDC2 dendritic cells (DCs) (Flt3, Ciita), are also observed in the healthy brain [41]. We also iden- migratory DCs (Flt3, Ccr7) and natural killer (NK) cells tified Mg4 as a cluster of proliferating microglia (Addi - (Klrb1c, Ncr1) (Fig. 1A, B). tional File 1: Figure S1D), which was most prominent in The immune composition in the retina was clearly the post ONC retina (Fig. 1C). The density and activa - altered in mice that underwent an ONC, with the latter tion of microglia in the different layers of the retina (e.g. showing an increase of peripheral myeloid cells (Fig. 1C). inner and outer plexiform layer) after ONC were further This was most notable for neutrophils, which showed an investigated via whole mount retinal staining. Confocal elevated infiltration in the retina at day 4 post ONC. We Z-stack images of the inner and outer plexiform layer also observed a large cluster of cells that expressed both revealed that microglia shifted from highly ramified cells macrophage and neutrophil markers (Additional File 1: to bigger more amoeboid cells with retracted processes, Figure S1A, B). These cells may represent macrophage- indicative for their reactive state (Fig. 1E, Additional File neutrophil aggregates that formed during in vitro single- 1: Figure S1E) [42]. cell processing. Alternatively, they may correspond to hi Cluster MDM1 represented a subset of Gpnmb Fab- microglia or macrophages that phagocytosed apoptotic hi neutrophils in vivo prior to retinal dissection and pro p5 macrophages that clustered distal from microglia and BAMs (Fig. 1A, B, Additional File 1: Figure S1F) cessing. For reasons of clarity, these cells were excluded and was restricted to the post ONC retina (Fig. 1C). Dif- from all further analyses. ferential gene expression analysis showed an absence Clusters Mg1–Mg4 expressed microglial signature of microglial signature genes in these cells, coupled to genes, including Sall1, which is reported to be restricted a high expression of genes associated with monocyte- to embryonically-derived microglia [36–38]. Addi- derived macrophages (MDMs), including Ms4a7, Lyz2 tionally, Mg1–Mg4 did not express genes related to and Clec12a (Fig. 1F). Therefore, this cluster may repre - BAMs or recruited monocyte-derived cells (e.g. Ms4a7, sent MDMs that were recruited to the retina following Clec12a) [36–38] and were therefore identified as micro - ONC. In line with this, we observed that microglia from glia. Interestingly, most retinal microglia from mice that the injured retina exhibited an elevated expression of the underwent ONC, clustered separately from their coun- monocyte chemoattractant Ccl2 (Fig. 1D). Furthermore, terparts observed in control mice (Fig. 1C), indicat- quantification of monocyte and macrophage infiltra - ing a change in microglial activation status. Microglia tion in the retina at days 2, 4, 6 and 8 post ONC via flow from ONC mice were mostly confined to cluster Mg2 cytometry, revealed a transient increase in the number of that, compared to Mg1, exhibited a downregulation of monocytes, while macrophage numbers peaked at later homeostatic microglial signature genes (P2ry12, Sall1, time points (Fig. 1G). A significant increase in the num - Tmem119) and an induction of prototypical disease- ber of monocytes was also observed in the injured optic associated microglia (DAM) markers (Cst7, Lpl, Fabp5) nerve (Fig. 1G). This shows that monocytes are attracted (Fig. 1D) [3, 39, 40]. This shows that the majority of reti - to both the retina and optic nerve following ONC nal microglia reacted to RGC axonal injury and that they injury, which is in line with MDM1 representing newly- exhibited gene expression changes that are compara- recruited monocyte-derived cells. Interestingly, the ble to those observed for brain microglia responding to (See figure on next page.) Fig. 1 Injury to the optic nerve activates resident and recruited myeloid cells in the retina. A UMAP and cluster annotation showing 7128 + + CD45 CD11b cells isolated from healthy retinas (naive mice) or injured retinas (mice receiving ONC). BAM border associated macrophage, cDC conventional dendritic cell, migDC migratory dendritic cell, MDM monocyte-derived macrophage, Mg microglia, MO monocyte, N neutrophils, NK natural killer cell. Data originate from retinas pooled from 32 naive mice and 10 ONC mice. B Dot plot corresponding to UMAP in (A), showing expression of the indicated genes. Dot size represents the percentage of cells expressing the gene and colour represents its average expression within a cell cluster. C UMAP showing 2819 cells of healthy retinas and 4309 cells of injured retinas with a pie chart showing the distribution of different immune cell populations present in the healthy and injured retinas. Numbers in the pie chart represent percentages of the cell subsets. D Volcano plot displaying differential expression between Mg2 (injured microglia) and Mg1 (healthy microglia). Genes with adjusted p-value < 0.01 and I Log2(FC) I > 1 are shown in red. E Retinal wholemounts stained for IBA1 (green) labelling the microglia in the retina of naive mice and mice at 4 dpi ONC. Scale bar 50 µm and 25 µm. Representative images of n = 3–4 mice per condition. F Volcano plot displaying differential expression between MDM1 and Mg2. Genes with adjusted p-value < 0.01 and I Log2(FC) I > 1 are shown in red. G Cell counts of monocytes and macrophages at different time points after ONC in the retina and optic nerve, as measured by flow cytometry. Data are shown as mean ± SEM. Repeated measures one-way ANOVA followed by Tukey’s multiple comparisons test, n = 3–6 biologically independent samples with retinas from 4 mice pooled per sample. * P < 0.05 Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 4 of 20 Fig. 1 (See legend on previous page.) A ndries et al. Acta Neuropathologica Communications (2023) 11:85 Page 5 of 20 MDM1 cluster exhibited a high expression of Apoe, Spp1 significantly different between the ONC and ONC + P3C and Igf1 (Additional File 1: Figure S1F), which have been conditions (Fig. 2C). This indicates that monocyte influx reported to be key genes involved in promoting RGC following intravitreal P3C injection is restricted to the regeneration following axonal injury [43]. This suggests retina. To spatially localize the infiltrating monocyte- that recruited MDMs can attain activation states that derived cells within the retina and vitreous, an immu- may promote RGC survival and/or axon regeneration. nostaining for IBA1 was performed on retinal sections However, as no axonal outgrowth is observed follow- of Lyz2-GFP mice subjected to ONC or ONC + P3C ing ONC, this response is potentially insufficient, or the (Fig. 2D). As shown in our scRNA-seq data, Lyz2 is number of recruited cells too low to induce regeneration. highly expressed in monocytes and monocyte-derived + + cells (Additional File 1: Figure S1F). IBA1 GFP cells Inflammatory stimulation upon ONC recruits thus likely represent monocytes and MDMs, although monocyte‑derived cells that exhibit long‑term we cannot rule out that a fraction of microglia may also engraftment in the retina and promote axonal upregulate Lyz2 following ONC and P3C treatment. P3C + + regeneration treatment induced a strong infiltration of IBA1 GFP As we observed the expression of potential pro-regener- cells that at day 2 were mostly observed in the vitre- ative genes in MDMs, we hypothesized that the reported ous and showed a round shape, indicative of monocytes regeneration of RGC axons following inflammatory stim - (Fig. 2D). Over time, these cells gradually changed their ulation may in part be driven by an increased recruitment morphology, adopting a more macrophage-like shape, of monocyte-derived cells in the retina. To investigate and infiltrated the ganglion cell layer, inner plexiform this, we combined ONC injury with intravitreal injection layer, inner nuclear layer and the outer plexiform layer of of the Toll-like receptor 2 agonist Pam3Cys combined the retina (Fig. 2D). with cAMP (P3C) and confirmed that this type of inflam - The increased number of macrophages observed in matory stimulation induced axonal regrowth of the dam- P3C treated mice may be driven by an increased recruit- aged RGCs (Additional File 2: Figure S2A, B) [28]. ment of monocytes and/or by an expansion of resident To assess the kinetics of peripheral myeloid cell infil - microglia/BAMs. To distinguish between these possibili- CreER tration, we performed flow cytometric analysis of the ties, we performed fate mapping using the Cx3cr1 : retina and optic nerve from C57BL/6 mice at days 2, 4, 6 R26-YFP model. Upon tamoxifen treatment in Cx3cr- CreER and 8 post ONC or ONC + P3C. P3C treatment induced 1 : R26-YFP mice, 95 ± 1% of retinal macrophages a large increase in the number of recruited neutrophils were YFP labelled (Fig. 3A, B). Four weeks following and monocytes in the retina, representing a ~ 360-fold tamoxifen administration, when YFP labelling in mono- and ~ 180-fold increase at day 2, respectively (Fig. 2A, B). cytes is lost [44], mice received ONC + P3C treatment. Neutrophil infiltration was very transient, with most cells Retinas were processed for flow cytometry at various disappearing by day 8 post treatment (Fig. 2B). Monocyte time points, ranging from 2 to 168 days post treatment. recruitment was similarly transient, but these cells gradu- We were able to distinguish resident microglia/BAMs ally differentiated into macrophages, as reflected by a loss from recruited MDMs based on their differential YFP of Ly6C and an increase in CX3CR1 expression (Fig. 2A). expression (Fig. 3A). This confirmed that the strong Coupled to this we observed a strong increase in the increase in retinal macrophages upon P3C treatment was number of retinal macrophages, which peaked around driven by the recruitment and differentiation of mono - day 6 post treatment (Fig. 2B). In contrast, the number of cytes (Fig. 3B, C). While the number of recruited mac- monocytes and macrophages in the optic nerve was not rophages progressively decreased, a substantial fraction (See figure on next page.) Fig. 2 Inflammatory stimulation upon optic nerve injury mobilizes the infiltration of monocyte-derived cells. A Representative flow cytometry plots low high high − high low showing the gating strategy used to identify neutrophils (Cx3cr1, Ly6G ), monocytes (Ly6C, Ly6G ) and macrophages (Cx3cr1, Ly6C ) in the retina and optic nerve of mice subjected to ONC and ONC combined with P3C treatment. Example plots were taken from retinas of mice at 2 dpi ONC and at 2 dpi ONC + P3C. Cells were pre-gated as live, single CD45 cells. B Cell counts of neutrophils, monocytes and macrophages at different time points after ONC (black) and ONC + P3C (coloured) in the retina, as measured by flow cytometry. Counts of the monocytes and macrophages after ONC correspond to the data shown in Fig. 1G. Data are shown as mean ± SEM. Statistical significance between ONC and ONC + P3C was evaluated via an unpaired t-test. n = 3–6 biologically independent samples with retinas from 4 mice pooled per sample. * P < 0.05 ** P < 0.01 C Cell counts of neutrophils, monocytes and macrophages at different time points after ONC (black) and ONC + P3C (coloured) in the optic nerve, as measured by flow cytometry. Counts of the monocytes and macrophages after ONC correspond to the data shown in Fig. 1G. Statistical significance between ONC and ONC + P3C was evaluated via an unpaired t-test. n = 3–6 biologically independent samples with optic nerves from 4 mice pooled per sample. D Retinal cryosections of Lyz2-GFP (green) mice stained for IBA1 (red). Sections are counterstained with DAPI (blue). Scale bar 50 µm. GCL ganglion cell layer, IPL inner plexiform layer, INL inner nuclear layer, OPL outer plexiform layer, ONL outer nuclear layer. Representative images of n = 3 mice per condition Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 6 of 20 Fig. 2 (See legend on previous page.) A ndries et al. Acta Neuropathologica Communications (2023) 11:85 Page 7 of 20 Fig. 3 The infiltrating monocyte-derived cells exhibit a long-term engraftment in the retina. A Representative flow cytometry plots showing + − the gating strategy for distinguishing resident macrophages (YFP ) from recruited monocyte-derived counterparts (YFP ) based on YFP CreER expression in retinas of Cx3cr1 :R26-YFP mice at various timepoints ranging from 2 to 168 dpi ONC + P3C. Cells were pre-gated as live single + + + + + − CD11b CD45 Ly6G-CX3CR1 F480 as shown in Fig. 2A. B Compiled flow cytometry data showing the percentage of YFP and YFP cells in the + + + + macrophage gate (live single CD11b CD45 Ly6G-CX3CR1 F480 ) at different time points after ONC + P3C ranging from 2 to 168 dpi ONC in the retina. Data are shown as mean ± SEM. n = 3–8 biologically independent samples containing 1 retina from 1 mouse. C Cell count of YFP resident macrophages (microglia and BAMs) and YFP recruited MDMs at different time points after ONC + P3C ranging from 2 to 168 dpi ONC in the retina, as measured by flow cytometry. Data are shown as mean ± SEM. Repeated measures one-way ANOVA followed by Tukey’s multiple comparisons test, statistical significance between different time points is indicated using different letters: conditions that share the same letter are not significantly different, while conditions with different letters are significantly different from each other. n = 3–8 biologically independent samples with 1 retina from 1 mouse showed long-term engraftment. At 70 days post treat- was strongly reduced in ONC + P3C treated Ccr2- − +/+ ment 75 ± 9% of macrophages were still recruited YFP deficient mice as compared to Ccr2 controls, while cells (Fig. 3B). However, at 168 days post treatment, the infiltration of neutrophils was unaltered (Fig. 4A, B). percentage of YFP retinal macrophages had dropped to Furthermore, while most macrophages in ONC + P3C +/+ hi −/− 30 ± 2%. Together, these data suggest that while a fraction treated Ccr2 retinas were CD45 , in Ccr2 retinas low of recruited MDMs were long-lived, their engraftment they were C D45 , indicative of microglia (Fig. 4A). −/− was still transient as they were progressively lost and/or These data thus reveal that in Ccr2 mice the replaced by embryonic microglia. recruitment of monocyte-derived macrophages to the −/− Next, we aimed to investigate whether MDMs that retina was strongly impaired. Importantly, Ccr2 infiltrate the retina upon inflammatory stimulation mice also showed a significantly reduced number of promote axonal regeneration. Therefore, we per- CTB regenerating axons (Fig. 4C). This indicates that formed ONC + P3C treatment in Ccr2-deficient mice, monocyte-derived macrophages that are recruited to which are known to have low numbers of blood mono- the retina upon P3C treatment, promote axonal regen- cytes [45]. Flow cytometry at day 4 post treatment eration of injured RGCs. confirmed that the number of infiltrating monocytes Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 8 of 20 Fig. 4 Recruited monocyte-derived macrophages that infiltrate the retina upon inflammatory stimulation promote axonal regeneration. A +/+ −/− Representative flow cytometry plots of cells from the retina of Ccr2 and Ccr2 mice at 4 dpi ONC + P3C. Cells were pre-gated as shown in hi low Fig. 2A. Percentages of monocytes and macrophages and of CD45 and CD45 macrophages are shown at 4 dpi ONC + P3C. B Cell counts +/+ −/− of neutrophils, monocytes and macrophages at 4 dpi ONC + P3C in the retina of Ccr2 (black) and Ccr2 (coloured) mice, as measured by +/+ −/− flow cytometry. Data are shown as mean ± SEM. Statistical significance between CCR2 and CCR2 was evaluated via the Mann–Whitney test. n = 3–5 biologically independent samples with 1 retina from 1 mouse. * P < 0.05. C Quantification of axonal regeneration on longitudinal +/+ −/− cryosections of the optic nerve of Ccr2 and Ccr2 mice at 14 dpi after ONC + P3C, analysed at various distances, starting from 150 µm after the ONC lesion site. Data are shown as mean ± SEM. Repeated measures two-way ANOVA followed by Tukey’s multiple comparisons test. n = 8–10 mice per condition. ****p < 0.0001 and *p < 0.5 Nerve injury combined with inflammatory stimulation (Fig. 5A). Additionally, to obtain insights into the tran- results in the hyperactivation of microglia which remain scription factors and gene regulatory networks that distinct from recruited monocyte‑derived macrophages shape the activation state of macrophages, we performed To profile the cell states and heterogeneity of mac - single-cell regulatory network inference and clustering rophages in the retina following inflammatory stimula - (SCENIC) analysis on all microglia and macrophage clus- + + tion, we performed scRNA-seq on CD45 CD11b cells ters [46, 47]. sorted from the retina at day 4 and 8 post ONC + P3C P3C treatment resulted in a large population of treatment. The obtained data were combined with those C1qb macrophages within the retina (Fig. 5A–C). of the naive and day 4 post ONC retina in a single dataset Most clusters expressed high levels of Ms4a7 and Lyz2 A ndries et al. Acta Neuropathologica Communications (2023) 11:85 Page 9 of 20 (Fig. 5C), suggesting that these cells were recruited MDMs from ONC + P3C treated retinas (MDM2-7) monocyte-derived cells (MDM2-MDM8). Ms4a7 and were distinct from MDMs observed in the ONC only Lyz2 were low in cluster Mg5, which also showed signa- condition (MDM1), indicating that P3C administra- ture expression of Fscn1, Nav3, Capn3 and Mgll, genes tion not only affected the level of MDM recruitment, that were shared with naive and ONC-only microglia but also altered their molecular state. Furthermore, (Fig. 5C). Mg5 was also the only macrophage cluster gene expression in ONC + P3C MDMs was dynamic in within the ONC + P3C-treated retinas that exhibited time, showing transcriptional divergence between cells Sall3 and Sall1 expression (Fig. 5C), genes known to profiled at day 4 versus day 8 post treatment (Fig. 5B). be highly restricted to embryonic microglia. SCENIC Interestingly, MDMs from ONC + P3C retinas exhib- analysis showed that the microglia-associated regulons ited a high level of heterogeneity (clusters MDM2-8). A Ets1, Etv1 and Klf3 were also active in Mg5 (Fig. 5D). fraction of these MDMs showed enriched Hexb expres- Together, this suggests that Mg5 represented micro- sion (MDM2-5), and these cells could be further subdi- hi hi glia, while MDM2-8 were recruited macrophages vided into a Trem2 (MDM2), Ch25h (MDM3-4) and hi low that remained transcriptionally distinct from micro- H2-Aa cluster (MDM5) (Fig. 5F). Within the Hexb glia. Cluster Mg5 represented 3% and 13% of the pro- macrophages, MDM6 expressed Ndrg1, Clec4b1 and + + filed CD11b cells at day 4 and 8 post ONC + P3C , Cd300e, while cells in MDM7 were Trem2 and showed respectively (Additional File 3: Figure S3). This was in an enriched expression of genes involved in phago- line with our previous fate mapping data, where we cytosis and lipid metabolism, including Fabp5 and observed 4 ± 1% and 16 ± 2% YFP microglia within Gpnmb (Fig. 5F). The latter genes were also enriched in CD11b cells at 4 and 8 days post treatment, respec- MDMs from the ONC-only retinas (MDM1). Addition- tively, as observed via flow cytometry in ONC + P3C ally, MDM7 expressed genes that are associated with CreER treated Cx3cr1 :R26-YFP mice. However, as we also hypoxia or HIF1a signalling, including Arg1 and Bnip3 observed enriched expression of genes that are related [49]. An active HIF1 regulon in MDM7 was also identi- to MDMs or BAMs, such as Clec12a, Clec4a1 and Itgal fied via SCENIC (Fig. 5D). MDMs also exhibited prolif- [10] (Fig. 5E), we cannot rule out that part of the Mg5 erative potential as represented by the MDM8 cluster. cluster is monocyte or BAM-derived. The putative Mg5 microglia from ONC + P3C treated retinas showed Identification of a pro‑regenerative gene signature many differentially expressed genes when compared to in recruited monocyte‑derived macrophages the Mg2 DAM cluster from ONC-only retinas (Fig. 5E), We wished to assess the nature of the crosstalk that suggesting a hyperactivation upon P3C treatment. This exists between macrophages and injured RGCs in included a further downregulation of homeostatic sig- ONC + P3C retinas and to identify important mol- nature genes in Mg5 as compared to Mg2 and an induc- ecules and pathways for axonal regrowth. Hereto, we tion of genes related to inflammatory activation, as relied on the dataset from Tran et al., who profiled highlighted by gene ontology (GO) analysis (Additional RGCs from naive and injured retinas at various time File 4: Figure S4). Mg5 cells also showed robust expres- points post ONC via scRNA-seq [50]. We merged their sion of the anti-inflammatory cytokine Il10 (Fig. 5C). dataset with ours and relied on the NicheNet algorithm This suggests that in the retina, microglia can produce [51] to screen for potential ligand-receptor interactions IL10 following nerve injury combined with TLR2 stim- between macrophages ("senders") and RGCs ("receiv- ulation, which is not observed in brain microglia upon ers"). We focused our analysis on intrinsically photosen- peripheral LPS challenge [48]. sitive RGCs (ipRGCs) and αRGCs [50], which are the (See figure on next page.) Fig. 5 Nerve injury combined with inflammatory stimulation results in the hyperactivation of microglia which remain distinct from recruited monocyte-derived macrophages. A UMAP and cluster annotation showing 14,963 macrophages of healthy, injured (4 dpi ONC) and regenerating (4 and 8 dpi ONC + P3C) retinas. BAM, border associated macrophage, MDM: monocyte-derived macrophage, Mg: microglia. Data originate from retinas pooled from 32 naive mice, 10 ONC mice, 4 ONC + P3C 4 dpi mice and 4 ONC + P3C 8dpi mice. B UMAP showing 3974 macrophages of regenerating retinas at 4 dpi ONC + P3C and 6957 macrophages of regenerating retinas at 8 dpi ONC + P3C with a pie chart showing the distribution of different macrophage populations present in the injured + P3C treated retinas. Numbers in the pie chart are indicating percentages of macrophage subsets. C Corresponding UMAPs revealing the expression of signature genes that differentiate between the multiple macrophage populations. The colour (grey, low expression; purple, high expression) represents the expression profile in the macrophage clusters. D Corresponding UMAPs showing the bimodal regulon activity of specific microglia and monocyte-derived regulons, with red dots indicating an active regulon in the corresponding cells. Regulon here refers to a module of co-expressed genes together with their corresponding transcription factor. E Volcano plot displaying differential expression between Mg5 and Mg2, Genes with adjusted p-value < 0.01 and I Log2(FC) I > 1 are shown in red. F Corresponding dot plot to the UMAP plot in Figure A, showing the expression of subset-specific genes, with the dot size representing the percentage of cells expressing the gene and the colour representing its average expression within a cluster Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 10 of 20 Fig. 5 (See legend on previous page.) A ndries et al. Acta Neuropathologica Communications (2023) 11:85 Page 11 of 20 subclasses most resilient to ONC injury and known to clusters (Additional File 5: Figure S5C), highlighting the have regenerative capacity [50, 52]. NicheNet was used crosstalk between these macrophage subsets with the to predict the ligand-receptor interactions that may drive injured RGCs. Interestingly, NicheNet identified the the gene expression changes observed in healthy versus ligands Spp1, Thbs1, Vegfa and Igf1 in MDM7 (Fig. 6B). injured RGCs at day 4 post ONC. Top ligands from the These are secreted proteins that are known to promote various macrophage clusters were selected and ranked RGC survival and/or axon regeneration following ONC based on their Pearson values between a ligand’s target injury [52–57]. Another predicted ligand was Nrg1, predictions and the observed transcriptional response which is involved in axon regeneration following periph- within the ONC + P3C retinas (Fig. 6A), together with eral nerve injury [58]. Besides these predicted ligands, an overview of the potential receptors (Additional File MDM7 also showed enriched gene expression for other 5: Figure S5A) and affected target genes in RGCs (Addi - secreted factors known to be involved in either RGC tional File 5: Figure S5B). Most ligand-receptor interac- or peripheral tissue regeneration, including Slpi, Sdc1 tions were predicted for the MDM6 and MDM7 sender and Fstl1 (Additional File 6: Figure S6A) [59–64]. These Fig. 6 Identification of a pro-regenerative gene signature in monocyte-derived macrophages. A Top ligands of each macrophage cluster were selected and ranked based on their ligand activity values. Each ligand was assigned to a cluster if it was expressed highest in this cluster compared to the remaining sender clusters. The colour (white, low expression; orange, high expression) represents the predicted activity of the ligands. The expression of the top ligands in each cluster is shown with the colour (blue, low expression; red, high expression) representing the scaled average expression in the corresponding cluster. B Circle plot of potential ligand-receptor pairs, that shows the links between predicted ligands from MDM cluster 7 with their associated receptors found on alpha- and intrinsically photosensitive retinal ganglion cells Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 12 of 20 ligands are currently not included in the NicheNet and provides evidence for a key role played by recruited ligand-target database and thus cannot be predicted, MDMs. but their expression further suggests a pro-regenerative Similar to the brain, the retina contains yolk-sac- signature in MDM7. This pro-regenerative phenotype derived microglia that self-renew and rely on CSF1 or may underlie the macrophage-mediated axonal regen- IL34 for their maintenance [40]. Our single-cell tran- eration that we observe in P3C treated retinas. Previous scriptomic profiling confirmed that homeostatic micro - work has suggested the involvement of oncomodulin and glia are the predominant myeloid cells in the healthy SDF1/CXCL12 in monocyte/macrophage mediated RGC retina and also revealed subsets of retinal BAMs that regeneration [65–69]. However, gene expression of Onc likely correspond to perivascular macrophages [35]. and Cxcl12 was not or hardly detected in the C D11b Upon ONC-induced RGC degeneration, resident mac- cells from our dataset (Additional File 6: Figure S6B). rophages changed their expression profile by downregu - lating homeostatic genes and upregulating genes related The pro‑regenerative factors identified to inflammatory activation. Notably, the specific neu - in monocyte‑derived macrophages can promote axon rodegenerative expression profile of retinal microglia in regeneration via paracrine signalling our nerve crush injury model was similar to the expres- Pro-regenerative factors such as THBS1, SPP1 or IGF1 sion profile of retinal DAMs observed under conditions have been described in the context of autocrine signal- of light-damage-induced photoreceptor degeneration ling, as these proteins are produced by surviving RGCs [40] or in glaucoma models [70]. It also resembles the [52, 53, 56]. We hypothesized that the secretion of these molecular signature of brain DAMs observed in amy- factors by macrophages may also contribute to RGC loid models of Alzheimer’s disease [3, 10, 71] and other regrowth via paracrine signalling. To investigate this, we pathological conditions of the CNS [5, 6, 8, 15, 72–74]. composed two mixtures of recombinant proteins based Therefore, the DAM phenotype is broadly similar in the on the pro-regenerative signature identified in cluster brain and retina and across multiple etiologically dis- MDM7. Mix 1 (THBS1, SLPI, VEGFA, SPP1 and IGF1) tinct diseases or injury models. Furthermore, it is also consisted of secreted proteins reported to induce axonal not strictly disease-associated, as a similar cell state is regeneration in the central nervous system. Mix 2 con- also observed for microglia from healthy young mice tained factors that have been shown to stimulate regen- that engulf myelin [9] or apoptotic neurons [4, 7] and eration upon their release in the periphery (SDC1, NRG1 for non-parenchymal microglia that live on the choroid and FSTL1), but have not yet been assessed in the con- plexus epithelium [10]. Therefore, microglia seem to text of RGC axonal regrowth. The individual mixes or the react in a similar way to many homeostatic disturbances. combination of both were injected in the vitreous imme- This may be a common feature of tissue-resident mac - diately after ONC, and at day 3 and 7 post ONC (Fig. 7A). rophages, as their highly specialized phenotypes require Control mice received intravitreal injections of PBS. tissue-imprinting that may limit their plasticity towards Axonal regeneration was assessed by quantifying the inflammatory insults [16]. The functional significance of number of CTB regrowing axons at day 14 post ONC the DAM response is dependent on the nature of the dis- on longitudinal optic nerve sections (Fig. 7B, C). Interest- turbance. In the retina it can be protective during pho- ingly, both mixes of recombinant proteins induced axonal toreceptor degeneration [40] but detrimental for RGC regeneration in the optic nerve. The largest increase in survival during glaucoma [70]. However, local P3C treat- the number of regenerating axons was observed for Mix ment did significantly alter microglial activation beyond 1, while a lower but still significant axonal regrowth was the DAM state, further reducing homeostatic signature observed for Mix 2. These results show that the recom - genes and driving inflammatory activation. This suggests binant proteins within mix 1 and 2 can stimulate the that strong TLR signalling induced by local P3C injection regeneration of RGCs when injected in the vitreous. This led to a hyperactivation of microglia. However, as these implies that the secretion of these pro-regenerative fac- cells also expressed genes related to recruited MDMs, we tors by MDMs can contribute to axonal regeneration via cannot rule out that they were partly monocyte derived. paracrine signalling in RGCs. It will be interesting to further assess the ontogeny and functional significance of these cells during RGC regen - Discussion eration in follow-up studies. The nature of the cell populations, cellular states, as well Monocytes that are recruited during disease may react as the molecules and signalling pathways that underly differently to the local inflammatory cues as compared to inflammation-induced axonal regrowth have remained resident macrophages. Previous myeloid cell fate map- elusive [33]. Our work now further highlights how ping studies, performed after ONC injury [35] and in inflammation and CNS regeneration are intertwined other retinal injury models, have highlighted a role for A ndries et al. Acta Neuropathologica Communications (2023) 11:85 Page 13 of 20 Fig. 7 The pro-regenerative factors identified in monocyte-derived macrophages can promote axon regeneration via paracrine signalling. A Schematic overview of the experimental setup of the different recombinant protein mixes intravitreally injected in the eye. B Representative images of regenerating axons that were CTB-traced on longitudinal cryosections of the optic nerve of mice at 14 dpi ONC and intravitreally injected with PBS, mix 1, mix 2 and mix 1 + 2. The ONC site is indicated with an asterisk. Scale bar 100 µm. C Quantification of axonal regeneration on longitudinal cryosections of the optic nerve of mice at 14 dpi ONC and intravitreal injection of PBS, mix 1, mix 2 and mix 1 + 2. Axonal number was counted at various distances starting at 150 µm from the ONC lesion site. Data are shown as mean ± SEM. Repeated measures two-way ANOVA followed by Tukey’s multiple comparisons test, statistical significance between different conditions at the same distance is indicated with different letters: conditions that share the same letter are not significantly different, while conditions with different letters are significantly different from each other, n = 4–5 mice per condition recruited MDMs [14, 75–77]. We observed that inflam - and brain ischemia [19], MDMs were observed to exert matory treatment resulted in a strong recruitment of a neuroprotective role and to facilitate repair, by dis- MDMs into the retina. ScRNA-seq analysis revealed playing multiple functions including anti-inflammatory that MDMs were transcriptionally distinct from resi- [18, 20, 80] and scar degrading roles [21, 81], as well as dent microglia and BAMs and exhibited transcriptional the ability to support axonal growth [20, 65, 80, 82, 83]. heterogeneity. Our data thus confirm that recruited Although our data highlight the importance of recruited MDMs exhibit substantial plasticity and show disease- MDMs, we do not exclude that also other immune specific adaptation. As inhibiting MDM recruitment also or non-immune cell types contribute to the observed impaired axonal outgrowth, this revealed the importance inflammation-induced axonal regeneration. For instance, of recruited MDMs in promoting RGC regeneration. An a recent study identified a subset of immature neutro - important contribution of myeloid cells to inflammation- phils with neuroprotective and regenerative properties induced optic nerve regeneration has long been debated, [84]. Furthermore, reactive macroglia (i.e. astrocytes and with conflicting views [25, 26, 65, 78, 79]. A possible Müller glia) may also add to the inflammation-enhanced explanation is the differential effect of resident versus axonal regeneration. Evidence indeed exists for recipro- recruited macrophages. Further dissecting the role of cal interactions between innate immune cells and mac- individual macrophage subsets or cell states will provide roglia in shaping the CNS response to injury and disease additional insights into the multifaceted role of innate [35, 85–88]. immunity in neurodegeneration versus protection and One of the first myeloid-cell-derived molecules repair. In animal models of acute spinal cord injury [20] reported to play a central role in RGC axonal regrowth Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 14 of 20 Tamoxifen treatment is oncomodulin. This small calcium-binding protein is CreER Three-to-four-week-old anesthetized Cx3cr1 : R26- reported to be secreted by macrophages and/or neutro- YFP mice were treated with tamoxifen (Sigma Aldrich, phils during Zymosan-driven ocular inflammation and to 2+ 20 mg/ml dissolved in corn oil (Sigma Aldrich)), which induce RGC survival and axonal regeneration via a Ca / was injected subcutaneously near the fore- and hind calmodulin kinase-dependant pathway [65–68]. Fur- limbs (4 × 50 µl). These injections were repeated three thermore, SDF1, also known as CXCL12, expressed by times at 48 h intervals. infiltrating monocytes/MDMs was reported to enhance oncomodulin activity [69]. Upon ONC + P3C treatment Intraorbital optic nerve crush model we did not identify Ocm or Cxcl12 gene expression in res- Optic nerve crush (ONC) was performed as previ- ident or recruited macrophages. We did identify a clus- ously described [93, 94]. Briefly, mice were anesthetized ter of MDMs showing enriched expression of multiple by intraperitoneal injection of a mixture of ketamine genes encoding proteins that have been shown to exert (Anesketin, Eurovet, 75 mg/kg body weight) and medeto- pro-regenerative effects in the CNS and are known to be midine (Domitor, Pfizer, 1 mg/kg body weight) diluted secreted. One of the most highly expressed genes encodes in saline (NaCl, Fischer Scientific, 0.9% in H O). and a for thrombospondin 1 (THBS1), a protein that is well- topical aesthetic ointment (oxybuprocaïnehydrochloride, known to mediate axon regeneration of RGCs in an auto- Unicaïne, Thea Pharma, 0,4%) was applied on the injured crine fashion [53]. Bray et al. showed that the observed eye. An incision in the temporal side of the conjunctiva effect of THBS1 depends on syndecan 1 (SDC1), a was made in the left eye. Then, the posterior side of the THBS1-binding protein (Bray et al. 2019). Autocrine eye was exposed, allowing visualization of the optic nerve. Sdc1 signalling has also been reported to mediate axon The exposed optic nerve was crushed approximately regrowth in the mouse PNS [59]. Other secreted proteins 1 mm from the optic nerve head with a cross-action for- expressed after the ONC + P3C treatment and known ceps for 5 s. Thereafter, a fundoscopy was performed and to promote regeneration of RGCs include secretory leu- animals with signs of ischemia were excluded. Eyes from kocyte protease inhibitor SLPI, osteopontin (SPP1) and uninjured mice were used as controls. insulin growth factor 1 (IGF1) [52, 56, 63]. Moreover, the putatively regenerative MDMs also expressed genes for Intravitreal injections secreted proteins that are known to be pro-regenerative Intravitreal injections were performed as previously in the PNS, such as VEGF [54], NRG1 [89] and FSTL1 described [93, 95]. Briefly, a Hamilton syringe equipped [64]. Our study now reveals that these previously iden- with a 34G Hamilton needle was inserted into the nasal tified pro-regenerative molecules are also produced by a part of the eye of anesthetised mice, at the limbus, under specific subcluster of MDMs in the regenerating retina a 45° angle to avoid damage to the lens. To induce an and that these factors can induce axonal regrowth of acute inflammatory stimulation, 2 µl of a combination of injured RGCs via non-cell autonomous paracrine sig- Pam3Cys (P3C, Sigma Aldrich, 2.5 µg/µl in sterile PBS nalling. SCENIC analysis identified HIF1A as a putative [96]) and chlorophenylthio-cyclic adenosine monophos- transcription factor that was driving this cell state. It will phate (CPT-cAMP, cAMP analogue, Sigma-Aldrich, be important to further identify the microenvironmental 50 µM in PBS) was injected immediately after the ONC signals and the gene regulatory networks that control the surgery. To trace regenerating RGC axons in the optic pro-regenerative phenotype of MDMs in future studies. nerve, 2 µl of cholera toxin subunit B conjugated to an This may pave the road for macrophage-centred strate - Alexa Fluor 488 fluorophore (CTB-Alexa488; Sigma gies for inducing and promoting neuroprotection and Aldrich, 5 µg/µl in sterile PBS containing dimethylsul- repair following injury and disease. foxide (DMSO, Sigma Aldrich, 0,5%)) was injected one day before sacrificing the mice. Recombinant proteins Material and methods (THBS1, SLPI, VEGFa, SPP1, IGF1, SDC1, NRG1 FSTL1, Animals R&D systems, 1 µg/µl in sterile PBS) were injected All experiments were performed using a combination 3 × 2 µl at 0, 3 and 7dpi ONC. of male and female 8–12-week-old mice of following CreER strains: C57BL/6 wild-type, Lyz2-GFP [90], Cx3cr1 −/− [91], R26-YFP [92] and Ccr2 [45] mice, as outlined Flow cytometry of myeloid inflammatory cells in Additional File 7: Table S1. All animal experiments Mice were euthanized with an intraperitoneal injection were approved by the Institutional Ethical Committees of an overdose of pentobarbital (Dolethal, Vetoquinol, for Animal Experimentation of KU Leuven and the Vrije 200 mg/kg body weight) and transcardially perfused with Universiteit Brussel and were conducted in strict accord- saline to remove all blood. Eyes were harvested and reti- ance with the European and Belgian legislation. nas and optic nerves dissected and transferred to Roswell A ndries et al. Acta Neuropathologica Communications (2023) 11:85 Page 15 of 20 Park Memorial Institute (RPMI) 1640 medium (Gibco). 1 µg/ml in MACS buffer) was used to exclude dead cells + + For the retinal samples, the retinal pigment epithelium and CD45 CD11b cells were sorted using a BD FACS was detached from the retina, but the vitreous was not ARIA III (BD Biosciences) equipped with a 100 µm noz- removed in order to include the infiltrating immune zle. Sorted cells were collected in ME medium (RPMI cells that were localized at the retina-vitreous interface. medium (Gibco) supplemented with heat-inactivated A single-cell suspension was obtained by mechanical FCS (Gibco, 20%), l-glutamine (Gibco, 300 μg/ml), pen- and enzymatic (collagenase I (Worthington, 10 U/ml), icillin (Gibco, 100 units/ml) and streptomycin (Gibco, collagenase IV (Worthington, 400 U/ml) and DNase I 100 μg/ml), non-essential amino acids (Gibco, 1 mM), (Worthington, 30 U/ml) diluted in Hank’s Balanced Salt sodium pyruvate (Gibco, 1 mM), 2-mercaptoethanol Solution (HBSS) medium (Gibco)) dissociation as previ- (Sigma Aldrich, 0.05 mM) and ActD (Sigma Aldrich, ously described (3 × 10 min at 37 °C) [10]. Afterwards, 3 μM)) for further processing in the 10 × genomics these cells were filtered, washed in MACS buffer (HBSS platform. medium (Gibco) supplemented with sterile filtered ethyl - enediaminetetraacetic acid (EDTA; Duchefa; 2 mM) and heat-inactivated fetal calf serum (FCS, Gibco, 2%)) and Single‑cell RNA sequencing using 10 × genomics platform blocked with anti-mouse CD16/CD32 (clone 2.4G2, BD The library construction for single-cell RNA sequencing Biosciences, 2 µg/µl in MACS buffer). Cells were stained (scRNA-seq) was performed as previously described [10]. with fluorescent antibodies in MACS buffer. The follow - Briefly, cellular suspensions of an estimated final concen - ing antibodies were used: F4/80 (BV421, clone BM8, Bio- tration of 1000 cells/µl were loaded on a GemCode Sin- legend), CD11c (PE/Cy7, BV510, clone N418, Biolegend), gle Cell Instrument (10 × Genomics) to partition them Ly6G (FITC, clone 1A8, Biolegend), Cx3cr1 (PE, clone into single-cell gel beads-in-emulsion (GEM). GEMs and SA011F11, Biolegend), CD11b (PE/Cy7, BV510, clone scRNA-seq libraries were prepared using the GemCode M1/70, Biolegend), Ly6C (APC, BV421, clone HK1.4, Single Cell 3ʹ Gel Bead and Library Kit (10 × Genom- Biolegend), CD45 (APC/Cy7, BV421, clone 30-F11, Bio- ics, No. 120237) and the Chromium i7 Multiplex Kit legend), MHCII (PerCP/Cy5.5, clone M5/114.15.2, Biole- (10 × Genomics, No. 120262) according to manufac- gend). Flow cytometry data were acquired using the BD turer’s instructions. Briefly, GEM reverse-transcription FACS CANTO II (BD Biosciences) and analysed using incubation was performed, followed by amplification of Flowjo v10.8 software. the full-length, barcoded cDNA, enzymatic fragmenta- tion, library construction by 5’ adaptor attachment to generate Illumina-ready sequencing libraries and eventu- Isolation of retinal CD11b + CD45 + cells for single‑cell RNA ally sample indexing. The cDNA content of pre-fragmen - sequencing tation and post-sample indexing was analysed using the Mice were euthanized and transcardially perfused with 2100 BioAnalyzer (Agilent). The libraries were sequenced saline. Retinas without retinal pigment epithelium on an Illumina HiSeq4000 flow cell with sequencing set - were transferred to RPMI (Gibco) containing actino- tings following the recommendations of 10 × Genomics mycin D (ActD, Sigma Aldrich, 30 μM) [97]. To obtain (read 1: 26 cycles; read 2: 98 cycles; index i7: eight cycles; sufficient number of cells, retinas were pooled from index i5: no cycles; 2.1 pM loading concentration). individual mice: 32 mice for the naïve sample (32 reti- nas), 10 mice for ONC sample, 4 mice for ONC + P3C Alignment and quantification of gene expression 4dpi sample, 4 mice for ONC + P3C 8dpi sample. The in single‑cell RNA sequencing data retinal samples underwent mechanical and enzymatic The Cell Ranger software (10 × Genomics) v.6.0.2 was (collagenase I (Worthington, 10 U/ml), collagenase used to perform sample demultiplexing and alignment of IV (Worthington, 400 U/ml) and DNase I (Roche, sequencing reads to the reference genome (Mus muscu- 30 U/ml) in HBSS medium (Gibco) containing ActD lus mm10), barcode processing, unique molecular identi- (Sigma Aldrich, 15 μM)) dissociation (3 × 10 min at fiers filtering and single-cell 3ʹgene counting. The average 37 °C). Afterwards, cells were filtered, resuspended in of the mean reads per cell was 49,780 ± 1629 SD, with an MACS buffer (HBSS medium (Gibco) supplemented average sequencing saturation metric of 59% ± 8% SD, as with sterile filtered EDTA (Duchefa; 2 mM) and heat- calculated by Cell Ranger. The further pre-processing and inactivated FCS (Gibco, 2%), containing ActD (Sigma analysis of the gene expression count matrices was per- Aldrich, 3 μM)) and blocked with anti-mouse CD16/ formed in R using Seurat v.3.2.3, DropletUtils v1.10.1.2, CD32 (clone 2.4G2, BD Biosciences, 2 µg/µl in MACS scater 1.18.3. The cellular barcodes, associated with low buffer). Cells were stained with CD45-APC (30-F11, quality “empty” droplets, were filtered out using the Biolegend) and CD11b-PE/Cy7 (M1/70, Biolegend) in “emptyDrops” function of the DropletUtils package with MACS. 4′,6-diamidino-2-phenylindole (DAPI, Dako, Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 16 of 20 the recommended FDR cutoff ≤ 0.1 for deviation from MDM2-7 were defined as sender, while alpha and ipRGC the ambient RNA profile. The gene expression matri - were defined as the receiver cell populations. Potential ces were further filtered for low quality cells, normal - ligands and receptors were identified as genes, expressed in ized and scaled, followed by selection of highly variable at least 10% of the sender/receiver population, respectively, genes, principal components analysis and clustering as and present in the prior interaction model. To prioritise the previously described (Scheyltjens et al., 2022). The genes, identified interactions, we performed NicheNet ligand activ - specifically expressed in each cluster, were identified via ity analysis, which ranks the ligands based on the presence differential expression analysis with the “FindMarkers” of their target genes in the gene set of interest, here defined function of Seurat (Wilcoxon Rank Sum test). The p-val - as the differentially expressed genes in the alpha and ipRGCs ues of differential expression were adjusted for multiple between the 4dpi ONC and the naive condition (adjusted p testing with Bonferroni correction. Clustering results value < 0.05). Next, we selected the top 40 ligands with high- were visualized using two-dimensional scatter plots est ligand activity (based on the Pearson score) and added with the Uniform Manifold Approximation and Projec- three ligands with lower ligand activity that had known neu- tion (UMAP) method. Several of the identified clusters roprotective effects (Thbs1, Nrg1 and Igf1). For the selected exhibited simultaneous expression of both macrophage 43 ligands, we inferred the top predicted receptors and target and neutrophil gene markers, e.g. C1qa, C1qb, P2ry12, genes in the receiver cells. For visualising the ligand—target Ms4a7; S100a8, S100a9, Retnlg, Csf3r. Additionally, those genes interactions, we showed the 110 most strongly pre- clusters showed a high doublet score, as calculated by the dicted targets of at least one of the selected ligands, that were scDblFinder package v.1.4.0, therefore they were assumed also part of the gene set of interest. to be macrophage–neutrophil aggregates and were excluded from further analysis. Immunohistochemistry on retinal whole mounts and cryosections of retina and optic nerve Single‑cell regulatory network inference and clustering Mice were euthanized as described above and transcar- using SCENIC dially perfused with saline followed by phosphate buff - We performed single-cell regulatory network inference ered paraformaldehyde (PFA, pH 7.4, Sigma Aldrich, analysis using SCENIC v1.2.4 [46] using the raw, untrans- 4% in PBS). For retinal whole mount stainings, the eyes formed UMI counts as input and following the proposed and subsequently the retinas were dissected, post-fixed workflow. The co-expression network was generated in PFA (pH 7.4, Sigma Aldrich, 4% in PBS) for 1 h and using GRNBoost2 via arboreto v0.1.5. For running GRN- rinsed in PBS. The retinas were incubated overnight Boost2, the expression matrix was filtered for genes with with the primary antibody (rabbit anti-IBA1, Wako, over 30 UMI counts and expressed in at least 40 cells. 1/2000 diluted in PBS supplemented with pre-immune The resulting transcription factor by gene targets matrix donkey serum (PID, Merck, 2%) and triton X-100 was imported in R and further analysed with the SCENIC (VWR, 2%)). After rinsing in PBS, the retinas were workflow with default parameters. The regulon activity, incubated for 2 h with a donkey anti-rabbit Alexa Fluor which identifies and scores gene regulatory networks or 488 secondary antibody (DAR488, Dako, 1/200 in PBS regulons in single cells, was calculated using AUCell as supplemented with PID (Merck, 2%) and triton X-100 previously described [46]. The better the gene targets of (VWR, 2%)). Mosaic pictures of the entire retinal whole a regulon match the highly expressed genes of a certain mounts were made using a confocal scanning micro- cell, the higher the AUC value (also named regulon activ- scope (Olympus FV 1000D). Microglia density, soma ity) of that regulon in that particular cell. The regulons size and roundness were analysed using a spatial statis- were visualized in a network using Cytoscape v.3.9.1 [98]. tics approach, all as previously described [99]. For retinal or optic nerve cryosections, complete Modelling the intercellular communication using NicheNet eyes and optic nerves were dissected, postfixed for We extracted gene expression matrices of RGCs of control 1 h at room temperature and cryoprotected through mice and mice 4 days post ONC using GSE137398 [50]. an ascending series of sucrose (Sigma Aldrich, 10%– The gene expression data was pre-processed as described 20%–30% in PBS). Afterwards, eyes or optic nerves above. The clusters "41_AlphaONT", "42_AlphaOFFS", were embedded in TissueTek (Sakura) and 14 µm "43_AlphaONS" and "45_AlphaOFFT" were grouped as thick sagittal sections of the eyes or longitudinal optic alphaRGCs, while the clusters "22_M5", "31_M2", "33_M1" nerve sections were made. For immunolabeling of the and "40_M1dup" were grouped as ipRGCs. For predict- cryosections of the eyes, epitope retrieval was accom- ing interactions between the macrophages and the RGCs, plished using citrate buffer (pH 6, citric acid (Chem- we applied the NicheNet package (v. 1.1.0), using the pre- lab, 10 mM) and Tween 20 (Sigma Aldrich, 0.05%) in build NicheNet prior model of ligand-receptor interactions. H O). Aspecific binding places were saturated with PID 2 A ndries et al. Acta Neuropathologica Communications (2023) 11:85 Page 17 of 20 (Merck, 20%) in Tris-sodium chloride blocking buffer (n). Statistically significant differences between multiple (TNB, triton X-100 (VWR, 1.5 mM %), Tris-HCl (Acros groups are specified using different letters. Conditions Organics, 0.1 M), NaCl (Fischer Scientific, 150 mM) with the same letter are not significantly different, while and blocking reagent (Perkin Elmer, 0.5%) in PBS)) conditions with different letters are significantly dif - and the primary antibodies (chicken anti-GFP, Abcam, ferent from each other. Statistical significance between 1/500 in TNB and rabbit anti-IBA1, Wako, 1/2000 in two groups were specified with **** for p < 0.0001, *** for TNB) were incubated overnight at room temperature. p < 0.001, ** for p < 0.01, and * for p < 0.5. After rinsing, the slides were incubated with, respec- tively, donkey anti-chicken Alexa Fluor 488 (DACh488, Supplementary Information Dako,1/200 in TNB) and donkey anti-rabbit biotin sec- The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s40478- 023- 01580-3. ondary antibody (DARbiotin, Dako, 1/300 in TNB), followed by subsequent incubation with streptavidin- + + Additional file 1. scRNA-seq of CD45 CD11b cells from naïve or ONC horse radish peroxidase (Strep-HRP, Dako, 1/100 in + retinas. A UMAP showing all CD11b cells profiled from the healthy and TNB) and tyramid signal amplification (TSA Cy3-Tyr, ONC retinas. BAM, border associated macrophage, cDC: conventional dendritic cell, migDC: migratory dendritic cell, MDM: monocyte-derived Thermofisher Scientific, 1/50 in amplification buffer). macrophage, Mg: microglia, MO, monocyte, N: neutrophils, NK: natural Finally, the slides were counterstained with DAPI killer cell. B UMAPs showing the expression of the indicated genes. Red (Dako, 1 µg/ml in PBS). Images of the mid-sagittal reti- line highlights the putative macrophage-neutrophil doublets. C Volcano plot displaying differential expression between Mg3 and Mg1. Genes with nal cryosections were taken using a Leica DM6 (Olym- adjusted p-value <0.01 and I Log2I >1 are shown in red. D Volcano plot pus) fluorescent microscope. For the optic nerves, displaying differential expression between Mg4 and Mg1. Genes with images of mid-longitudinal sections that contained the adjusted p-value <0.01 and I Log2I >1 are shown in red. E Quantification of the density and activityof microglia in retina at different timepoints ONC site were taken using a confocal scanning micro- after ONC corresponding with images shown in figure 1E. Data are shown scope (Olympus FV 1000D). as mean ± SEM. Repeated measures one-way ANOVA followed by Tukey’s multiple comparisons test, statistical significance between different time - points is indicated using different letters: conditions that share the same letter are not significantly different, while conditions with different letters Quantification of axonal growth are significantly different from each other. n=3-4 mice per condition. F Axon growth was quantified on three mid-longitudinal UMAPs showing the expression of the indicated genes, corresponding to the dataset shown in S1A. cryosections of the optic nerve by manually counting Additional file 2. Inflammatory treatment stimulates axonal initiation. the number of C TB axons every 150 µm (distance d) A Representative images of longitudinal cryosections of the optic nerve beyond the crush site, using ImageJ [100]. In addition, at showing regenerating axons that were CTB-traced at different timepoints each distance, the cross-sectional width of the nerve was after ONC and ONC+P3C. The ONC site is indicated by an asterisk. Scale bar 50µm. B Quantification of axonal regeneration in the optic nerve of measured along. The total estimated number of axons mice at different timepoints after ONC or ONC combined with P3C treat - in the optic nerve extending distance d from the ONC ment. The number of regrowing axons was analysed at various distances lesion site was calculated using following formula where starting at 150 µm from the ONC lesion site. Representative images of n = 3 mice per condition. Quantitative data after ONC+IS are shown as mean the radius of the optic nerve was set at r = 150 µm and the ± SEM. Repeated measures two-way ANOVA followed by Tukey’s multiple thickness of the sections was t = 14 µm, all as described comparisons test, statistical significance between different conditions previously [101]. at the same distance is indicated with different letters, n=3-5 mice per condition. Average(#axons/µ m of nerve width) Additional file 3. Full scRNA-seq dataset of naïve, ONC and ONC+P3C �a = πr . + + CD11b CD45 cells. A UMAP and cluster annotation showing 22081 cells of both healthy, injuredand regeneratingretinas. BAM, border associated macrophage, cDC: conventional dendritic cell, migDC: migratory dendritic The results obtained for each of the three sections per cell, MDM: monocyte-derived macrophage, Mg: microglia, MO, monocyte, nerve were averaged. N: neutrophils, NK: natural killer cell. B,C UMAP showing 6997 cells of retinas at 4dpi ONC+P3Cand 7956 cells of retinas at 8dpi ONC+P3C. Individual pie charts show the distribution of neutrophils, monocytes or Statistics alle immune populations. Numbers in the pie chart are percentages of the cells from the corresponding cluster. Statistical analyses were performed using GraphPad Prism 8 software (GraphPad Software). Normal distri- Additional file 4. Expression of pro-regenerative genes in cluster MDM7. Gene ontology analysis on the upregulated genes in Mg5 versus Mg2> 20; bution was evaluated using a Kolmogorov–Smirnov test log2>1) showing the top 20 enriched GO terms for Mg5. and parallel equal variance between groups was tested. Additional file 5. Nichenet analysis of MDMs against injured RGCs. A Outliers were identified and excluded based on a Grubb’s Overview of potential receptors on the retinal ganglion cells of the ligands test (extreme studentised deviate method). The values are expressed by the different macrophage clusters. The colourrepresents the regulatory potential of the receptors based on the prior model of ligand- expressed as mean values ± standard error (SEM). Sta- receptor interactions.Receptor expression in the different retinal ganglion tistical tests are specified in the figure legends, together cell populations is shown with the colourrepresenting the scaled average with the number of biologically independent samples Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 18 of 20 2. Drieu A, Du S, Storck SE, Rustenhoven J, Papadopoulos Z, Dykstra T et al expression in the corresponding cluster. B Overview of the predicted (2022) Parenchymal border macrophages regulate the flow dynamics target genes in the retinal ganglion cells of the ligands expressed by of the cerebrospinal fluid. Nature 611:585–593 the different macrophage clusters. The colourrepresents the regulatory 3. Keren-Shaul H, Spinrad A, Weiner A, Matcovitch-Natan O, Dvir-Sztern- potential of the target genes based on the prior model of ligand-target feld R, Ulland TK et al (2017) A unique microglia type associated with gene interactions. C Circle plot of potential ligand-receptor pairs. It shows restricting development of Alzheimer’s disease. Cell 169:1276–1290 the links between predicted ligands from the different monocyte-derived 4. Anderson SR, Roberts JM, Zhang J, Steele MR, Romero CO, Bosco A macrophage clusters of the regenerating retinawith their associated et al (2019) Developmental apoptosis promotes a disease-related gene receptors found on alpha- and intrinsically photosensitive retinal ganglion signature and independence from CSF1R signaling in retinal microglia. cells. Cell Rep 27:2002–2013 5. Hammond TR, Dufort C, Dissing-Olesen L, Giera S, Young A, Wysoker Additional file 6. Pro-regenerative gene signature in cluster MDM7. A A et al (2019) Single-cell RNA sequencing of microglia throughout Corresponding dot plot of the recruited monocyte-derived macrophage the mouse lifespan and in the injured brain reveals complex cell-state populations showing the expression of selected pro-regenerative genes, changes. Immunity 50:253–271 with the dot size representing the percentage of cells expressing the gene 6. Jordão MJC, Sankowski R, Brendecke SM, Locatelli G, Tai YH et al (2019) and the colour representing its average expression within a cluster. B Neuroimmunology: single-cell profiling identifies myeloid cell subsets UMAP plots showing expression of the indicated genes, Cxcl12 and Ocm, with distinct fates during neuroinflammation. Science 363:eaat7554 corresponding to the dataset shown in S3A. 7. Li Q, Cheng Z, Zhou L, Darmanis S, Neff NF, Okamoto J et al (2019) Additional file 7. Overview of mouse strains used in this study. Developmental heterogeneity of microglia and brain myeloid cells revealed by deep single-cell RNA sequencing. Neuron 101:207–223 8. Masuda T, Sankowski R, Staszewski O, Böttcher C, Amann L et al (2019) Acknowledgements Spatial and temporal heterogeneity of mouse and human microglia at We thank Veronique Brouwers, Marijke Christiaens and Lut Noterdaeme single-cell resolution. Nature 566:388–392 for their skilful technical assistance and Evelien Herinckx for her excellent 9. Safaiyan S, Besson-Girard S, Kaya T, Cantuti-Castelvetri L, Liu L, Ji H animal care taking. This research was funded by the Research Council of KU et al (2021) White matter aging drives microglial diversity. Neuron Leuven (KU Leuven BOF-OT/14/064), the Research Foundation Flanders (FWO, 109:1100–1117 G0B2315N and G053217N) and Innoviris (Attract Grant BB2B 2015-2). Luca 10. Van Hove H, Martens L, Scheyltjens I, De Vlaminck K, Pombo Antunes Masin, Steven Bergmans, Marie Claes and Lies De Groef were/are supported AR, De Prijck S et al (2019) A single-cell atlas of mouse brain mac- by a fellowship of the Research Foundation Flanders. rophages reveals unique transcriptional identities shaped by ontogeny and tissue environment. Nat Neurosci 22:1021–1035 Author contributions 11. Rashid K, Akhtar-Schaefer I, Langmann T (2019) Microglia in retinal LA together with LDG, LM and KM conceptualized the study and wrote, degeneration. Front Immunol 10:1975–1994 reviewed and edited the manuscript. LA collected the data with the help of 12. Choi S, Guo L, Cordeiro MF (2021) Retinal and brain microglia in multi- LM, IS, HH, KDV, SB and MC. LA, DK, IS and KM performed the bioinformatic ple sclerosis and neurodegeneration. Cells 10:1507–1528 analysis of the scRNA-seq data. All authors have read and agree to the pub- 13. Chen X, Holtzman DM (2022) Emerging roles of innate and adaptive lished version of the manuscript. immunity in Alzheimer’s disease. Immunity 55:2236–2254 14. Yu C, Roubeix C, Sennlaub F, Saban DR (2020) Microglia versus Data availability monocytes: distinct roles in degenerative diseases of the retina. Trends The scRNA-seq datasets described in this article will be made accessible Neurosci 43:433–449 through our interactive webserver at www. brain immun eatlas. org. Further- 15. De Vlaminck K, Van Hove H, Kancheva D, Scheyltjens I, Pombo Antunes more, all raw data and gene-cell count matrices are deposited at GEO (NCBI) AR, Bastos J et al (2022) Differential plasticity and fate of brain-resident with accession code GSE232470. and recruited macrophages during the onset and resolution of neuro- inflammation. Immunity 55:2085–2102 16. Guilliams M, Svedberg FR (2021) Does tissue imprinting restrict mac- Declaration rophage plasticity? Nat Immunol 22:118–127 17. Shechter R, Schwartz M (2013) Harnessing monocyte-derived mac- Competing interests rophages to control central nervous system pathologies: no longer ‘if ’ The authors declare no conflict of interest. but ‘how.’ J Pathol 229:332–346 18. Shechter R, Miller O, Yovel G, Rosenzweig N, London A, Ruckh J et al Author details (2013) Recruitment of beneficial M2 macrophages to injured spinal Neural Circuit Development and Regeneration Research Group, Animal cord is orchestrated by remote brain choroid plexus. Immunity Physiology and Neurobiology Division, Department of Biology, Leuven Brain 38:555–569 Institute, KU Leuven, Naamsestraat 61, Box 2464, 3000 Louvain, Belgium. 19. Wattananit S, Tornero D, Graubardt N, Memanishvili T, Monni E, Laboratory for Molecular and Cellular Therapy, Vrije Universiteit Brussel, Tatarishvili J et al (2016) Monocyte-derived macrophages contribute Laarbeeklaan 103, 1090 Brussels, Belgium. Myeloid Cell Immunology Lab, VIB to spontaneous long-term functional recovery after stroke in mice. J Center for Inflammation Research, Brussels, Belgium. Cellular Communication Neurosci 36:4182–4195 and Neurodegeneration Research Group, Animal Physiology and Neurobi- 20. Shechter R, London A, Varol C, Raposo C, Cusimano M, Yovel G et al ology Division, Department of Biology, Leuven Brain Institute, KU Leuven, (2009) Infiltrating blood-derived macrophages are vital cells playing an 3000 Louvain, Belgium. anti-inflammatory role in recovery from spinal cord injury in mice. PLoS Med 6:e1000113 Received: 29 March 2023 Accepted: 8 May 2023 21. London A, Itskovich E, Benhar I, Kalchenko V, Mack M, Jung S et al (2011) Neuroprotection and progenitor cell renewal in the injured adult murine retina requires healing monocyte-derived macrophages. J Exp Med 208:23–39 22. Lee-Liu D, Edwards-Faret G, Tapia VS, Larraín J (2013) Spinal cord References regeneration: lessons for mammals from non-mammalian vertebrates. 1. Prinz M, Masuda T, Wheeler MA, Quintana FJ (2021) Microglia and Genesis 51:529–544 central nervous system-associated macrophages mdash from origin to disease modulation. Annu Rev Immunol 39:251–277 A ndries et al. Acta Neuropathologica Communications (2023) 11:85 Page 19 of 20 23. David S, Kroner A (2011) Repertoire of microglial and macrophage cytokine responses in C-C chemokine receptor 2 knockout mice. J Clin responses after spinal cord injury. Nat Neurosci 12:388–399 Invest 100:2552–2561 24. Gensel JC, Zhang B (2015) Macrophage activation and its role in repair 46. Aibar S, González-Blas CB, Moerman T, Huynh-Thu VA, Imrichova H, and pathology after spinal cord injury. Brain Res 1619:1–11 Hulselmans G et al (2017) SCENIC: single-cell regulatory network infer- 25. Hilla AM, Diekmann H, Fischer D (2017) Microglia are irrelevant for neu- ence and clustering. Nat Methods 14:1083–1086 ronal degeneration and axon regeneration after acute injury. J Neurosci 47. Van de Sande B, Flerin C, Davie K, De Waegeneer M, Hulselmans G, 37:6113–6124 Aibar S et al (2020) A scalable SCENIC workflow for single-cell gene 26. Hauk TG, Müller A, Lee J, Schwendener R, Fischer D (2008) Neuroprotec- regulatory network analysis. Nat Protoc 15:2247–2276 tive and axon growth promoting effects of intraocular inflammation 48. Shemer A, Scheyltjens I, Frumer GR, Kim JS, Grozovski J, Ayanaw S et al do not depend on oncomodulin or the presence of large numbers of (2020) Interleukin-10 prevents pathological microglia hyperactivation activated macrophages. Exp Neurol 209:469–482 following peripheral endotoxin challenge. Immunity 53:1033–1049 27. Leibinger M, Andreadaki A, Gobrecht P, Levin E, Diekmann H, Fischer 49. Pombo Antunes AR, Scheyltjens I, Lodi F, Messiaen J, Antoranz A, Duer- D (2016) Boosting central nervous system axon regeneration by inck J et al (2021) Single-cell profiling of myeloid cells in glioblastoma circumventing limitations of natural cytokine signaling. Mol Ther across species and disease stage reveals macrophage competition and 24:1712–1725 specialization. Nat Neurosci 24:595–610 28. Hauk TG, Leibinger M, Müller A, Andreadaki A, Knippschild U, Fischer 50. Tran NM, Shekhar K, Whitney IE, Jacobi A, Benhar I, Hong G et al (2019) D (2010) Stimulation of axon regeneration in the mature optic nerve Single-cell profiles of retinal ganglion cells differing in resilience to by intravitreal application of the toll-like receptor 2 agonist Pam3Cys. injury reveal neuroprotective genes. Neuron 104:1039–1055 Invest Ophthalmol Vis Sci 51:459–464 51. Browaeys R, Saelens W, Saeys Y (2020) NicheNet: modeling intercel- 29. Diekmann H, Leibinger M, Fischer D (2013) Do growth-stimulated lular communication by linking ligands to target genes. Nat Methods retinal ganglion cell axons find their central targets after optic nerve 17:159–162 injury? New insights by three-dimensional imaging of the visual path- 52. Jacobi A, Tran NM, Yan W, Benhar I, Tian F, Schaffer R et al (2022) way. Exp Neurol 248:254–257 Overlapping transcriptional programs promote survival and axonal 30. Andries L, Masin L, Salinas-Navarro M, Zaunz S, Claes M, Bergmans S regeneration of injured retinal ganglion cells. Neuron 110:2625–2645 et al (2021) Mmp2 modulates inflammatory response during axonal 53. Bray ER, Yungher BJ, Levay K, Ribeiro M, Dvoryanchikov G, Ayupe AC regeneration in the murine visual system. Cells 10:1672–1691 et al (2019) Thrombospondin-1 mediates axon regeneration in retinal 31. Benowitz LI, Popovich PG (2011) Inflammation and axon regeneration. ganglion cells. Neuron 103:642–657 Curr Opin Neurol 24:577–583 54. Böcker-Meffert S, Rosenstiel P, Röhl C, Warneke N, Held-Feindt J, Sievers 32. Cui Q, Yin Y, Benowitz LI (2009) The role of macrophages in optic nerve J et al (2002) Erythropoietin and VEGF promote neural outgrowth from regeneration. Neuroscience 158:1039–1048 retinal explants in postnatal rats. IOVS 43:2021–2026 33. Andries L, De Groef L, Moons L (2020) Neuroinflammation and optic 55. Rosenstein JM, Mani NM, Khaibullina A, Krum JM (2003) Neurotrophic nerve regeneration: where do we stand in elucidating underlying cel- effects of vascular endothelial growth factor ( VEGF) on organo - lular and molecular players? Curr Eye Res 45:397–409 typic cortical explants and primary cortical neurons. J Neurosci 34. Wong KA, Benowitz LI (2022) Retinal Ganglion cell survival and axon 23:11036–11044 regeneration after optic nerve injury: role of inflammation and other 56. Duan X, Qiao M, Bei F, Kim I-JJ, He Z, Sanes JRR (2015) Subtype-specific factors. Int J Mol Sci 23:10179–10194 regeneration of retinal ganglion cells following axotomy: effects of 35. Benhar I, Ding J, Yan W, Whitney IE, Jacobi A, Sud M et al (2023) Tem- osteopontin and mTOR signaling. Neuron 85:1244–1256 poral single cell atlas of non-neuronal retinal cells reveals dynamic, 57. Joshi Y, Sória MG, Quadrato G, Inak G, Zhou L, Hervera A et al (2015) The coordinated multicellular responses to central nervous system injury. MDM4/MDM2-p53-IGF1 axis controls axonal regeneration, sprouting Nat Immunol 24:700–713 and functional recovery after CNS injury. Brain 138:1843–1862 36. Bennett FC, Bennett ML, Yaqoob F, Mulinyawe SB, Grant GA, Hayden 58. Fricker FR, Lago N, Balarajah S, Tsantoulas C, Tanna S, Zhu N et al (2011) Gephart M et al (2018) A combination of ontogeny and CNS environ- Axonally derived neuregulin-1 is required for remyelination and regen- ment establishes microglial identity. Neuron 98:1170–1183 eration after nerve injury in adulthood. J Neurosci 31:3225–3233 37. Shemer A, Grozovski J, Tay TL, Tao J, Volaski A, Süß P et al (2018) 59. Murakami K, Tanaka T, Bando Y, Yoshida S (2015) Nerve injury induces Engrafted parenchymal brain macrophages differ from microglia in the expression of syndecan-1 heparan sulfate proteoglycan in primary transcriptome, chromatin landscape and response to challenge. Nat sensory neurons. Neuroscience 300:338–350 Commun 9:1–16 60. Edwards TJ, Hammarlund M (2014) Syndecan promotes axon regenera- 38. Cronk JC, Filiano AJ, Louveau A, Marin I, Marsh R, Ji E et al (2018) Periph- tion by stabilizing growth cone migration. Cell Rep 8:272–283 erally derived macrophages can engraft the brain independent of 61. Taggart CC, Cryan SA, Weldon S, Gibbons A, Greene CM, Kelly E et al irradiation and maintain an identity distinct from microglia. J Exp Med (2005) Secretory leucoprotease inhibitor binds to NF-κB binding sites in 215:1627–1647 monocytes and inhibits p65 binding. J Exp Med 202:1659–1668 39. Deczkowska A, Keren-Shaul H, Weiner A, Colonna M, Schwartz M, Amit 62. Mueller AM, Pedré X, Stempfl T, Kleiter I, Couillard-Despres S, Aigner L I (2018) Disease-associated microglia: a universal immune sensor of et al (2008) Novel role for SLPI in MOG-induced EAE revealed by spinal neurodegeneration. Cell 173:1073–1081 cord expression analysis. J Neuroinflamm 5:1–20 40. O’Koren EG, Yu C, Klingeborn M, Wong AYW, Prigge CL, Mathew R et al 63. Ghasemlou N, Bouhy D, Yang J, López-Vales R, Haber M, Thuraisingam T (2019) Microglial function is distinct in different anatomical locations et al (2010) Beneficial effects of secretory leukocyte protease inhibitor during retinal homeostasis and degeneration. Immunity 50:723–737 after spinal cord injury. Brain 133:126–138 41. Sala Frigerio C, Wolfs L, Fattorelli N, Thrupp N, Voytyuk I, Schmidt I et al 64. Wei K, Serpooshan V, Hurtado C, Diez-Cunado M, Zhao M, Maruyama S (2019) The major risk factors for Alzheimer’s disease: age, sex, and genes et al (2015) Epicardial FSTL1 reconstitution regenerates the adult mam- modulate the microglia response to Aβ plaques. Cell Rep 27:1293–1306 malian heart. Nature 525:479–485 42. Heindl S, Gesierich B, Benakis C, Llovera G, Duering M, Liesz A (2018) 65. Yin Y, Henzl MT, Lorber B, Nakazawa T, Thomas TT, Jiang F et al (2006) Automated morphological analysis of microglia after stroke. Front Cell Oncomodulin is a macrophage-derived signal for axon regeneration in Neurosci 12:1–11 retinal ganglion cells. Nat Neurosci 9:843–852 43. Guo C, Cho KS, Li Y, Tchedre K, Antolik C, Ma J et al (2018) IGFBPL1 66. Yin Y, Cui Q, Gilbert HY, Yang Y, Yang Z, Berlinicke C et al (2009) Onco- regulates axon growth through IGF-1-mediated signaling cascades. Sci modulin links inflammation to optic nerve regeneration. Proc Natl Acad Rep 8:1–13 Sci 106:19587–19592 44. Goldmann T, Wieghofer P, Müller PF, Wolf Y, Varol D, Yona S et al (2013) A 67. Kurimoto T, Yin Y, Habboub G, Gilbert H-YY, Li Y, Nakao S et al (2013) new type of microglia gene targeting shows TAK1 to be pivotal in CNS Neutrophils express oncomodulin and promote optic nerve regenera- autoimmune inflammation. Nat Neurosci 16:1618–1626 tion. J Neurosci 33:14816–14824 45. Boring L, Gosling J, Chensue SW, Kunkel SL, Farese RV, Broxmeyer HE 68. Singh B, Plemel JR (2014) Neutrophil contribution in facilitating optic et al (1997) Impaired monocyte migration and reduced type 1 ( Th1) nerve regeneration. J Neurosci 34:1081–1082 Andries et al. Acta Neuropathologica Communications (2023) 11:85 Page 20 of 20 69. Xie L, Cen LP, Li Y, Gilbert HY, Strelko O, Berlinicke C et al (2022) 90. Faust N, Varas F, Kelly LM, Heck S, Graf T (2000) Insertion of enhanced Monocyte-derived SDF1 supports optic nerve regeneration and alters green fluorescent protein into the lysozyme gene creates mice with retinal ganglion cells’ response to Pten deletion. Proc Natl Acad Sci U S green fluorescent granulocytes and macrophages. Blood 96:719–726 A 119:e2113751119 91. Yona S, Kim KW, Wolf Y, Mildner A, Varol D, Breker M et al (2013) Fate 70. Margeta MA, Yin Z, Madore C, Pitts KM, Letcher SM, Tang J et al mapping reveals origins and dynamics of monocytes and tissue mac- (2022) Apolipoprotein E4 impairs the response of neurodegenerative rophages under homeostasis. Immunity 38:79–91 retinal microglia and prevents neuronal loss in glaucoma. Immunity 92. Srinivas S, Watanabe T, Lin CS, William CM, Tanabe Y, Jessell TM et al 55:1627–1644 (2001) Cre reporter strains produced by targeted insertion of EYFP and 71. Krasemann S, Madore C, Cialic R, Baufeld C, Calcagno N, El Fatimy R et al ECFP into the ROSA26 locus. BMC Dev Biol 1:1–8 (2017) The TREM2-APOE pathway drives the transcriptional phenotype 93. De Groef L, Dekeyster E, Geeraerts E, Lefevere E, Stalmans I, Salinas of dysfunctional microglia in neurodegenerative diseases. Immunity Navarro M et al (2016) Differential visual system organization and 47:566–581 susceptibility to experimental models of optic neuropathies in three 72. Sousa C, Golebiewska A, Poovathingal SK, Kaoma T, Pires-Afonso commonly used mouse strains. Exp Eye Res 145:235–247 Y, Martina S et al (2018) Single-cell transcriptomics reveals distinct 94. Lefevere E, Salinas-Navarro M, Andries L, Noterdaeme L, Etienne I, Van inflammation-induced microglia signatures. EMBO Rep 19:e46171 WE et al (2020) Tightening the retinal glia limitans attenuates neuroin- 73. Lodder C, Scheyltjens I, Stancu IC, Botella Lucena P, Gutiérrez de Ravé flammation after optic nerve injury. Glia 68:2643–2660 M, Vanherle S et al (2021) CSF1R inhibition rescues tau pathology and 95. Devoldere J, Peynshaert K, Dewitte H, Vanhove C, De Groef L, Moons neurodegeneration in an A/T/N model with combined AD pathologies, L et al (2019) Non-viral delivery of chemically modified mRNA to the while preserving plaque associated microglia. Acta Neuropathol Com- retina: subretinal versus intravitreal administration. J Control Release mun 9:1–21 307:315–330 74. Beuker C, Schafflick D, Strecker JK, Heming M, Li X, Wolbert J et al (2022) 96. Pernet V, Schwab ME (2012) The role of Nogo-A in axonal plasticity, Stroke induces disease-specific myeloid cells in the brain parenchyma regrowth and repair. Cell Tissue Res 349:97–104 and pia. Nat Commun 13:1–14 97. Wu YE, Pan L, Zuo Y, Li X, Hong W (2017) Detecting activated cell popu- 75. Reyes NJ, Mathew R, Saban DR (2019) Fate mapping in vivo to lations using single-cell RNA-Seq. Neuron 96:313–329 distinguish bona fide microglia versus recruited monocyte-derived 98. Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D et al macrophages in retinal disease. Methods Mol Biol 1834:153–164 (2003) Cytoscape: a software environment for integrated models of 76. O’Koren EG, Mathew R, Saban DR (2016) Fate mapping reveals biomolecular interaction networks. Genome Res 13:2498–2504 that microglia and recruited monocyte-derived macrophages are 99. Davis BM, Salinas-Navarro M, Cordeiro MF, Moons L, De GL (2017) Char- definitively distinguishable by phenotype in the retina. Sci Rep acterizing microglia activation: a spatial statistics approach to maximize 6:20636–20648 information extraction. Sci Rep 7:1–12 77. Wieghofer P, Hagemeyer N, Sankowski R, Schlecht A, Staszewski O, 100. Schneider CA, Rasband WS, Eliceiri KW (2012) NIH image to ImageJ: 25 Amann L et al (2021) Mapping the origin and fate of myeloid cells years of image analysis. Nat Methods 9:671–675 in distinct compartments of the eye by single-cell profiling. EMBO J 101. Leon S, Yin Y, Nguyen J, Irwin N, Benowitz LI (2000) Lens injury 40:e105123 stimulates axon regeneration in the mature rat optic nerve. J Neurosci 78. Yin Y, Cui Q, Li Y, Irwin N, Fischer D, Harvey AR et al (2003) Macrophage- 20:4615–4626 derived factors stimulate optic nerve regeneration. J Neurosci 23:2284–2293 Publisher’s Note 79. Gensel JC, Nakamura S, Guan Z, Van RN, Ankeny DP, Popovich PG (2009) Springer Nature remains neutral with regard to jurisdictional claims in pub- Macrophages promote axon regeneration with concurrent neurotoxic- lished maps and institutional affiliations. ity. J Neurosci 29:3956–3968 80. Kigerl KA, Gensel JC, Ankeny DP, Alexander JK, Donnelly DJ, Popovich PG (2009) Identification of two distinct macrophage subsets with diver - gent effects causing either neurotoxicity or regeneration in the injured mouse spinal cord. J Neurosci 29:13435–13444 81. Rolls A, Shechter R, London A, Segev Y, Jacob-Hirsch J, Amariglio N et al (2008) Two faces of chondroitin sulfate proteoglycan in spinal cord repair: a role in microglia/macrophage activation. PLoS Med 5:1262–1277 82. David S, Greenhalgh AD, Kroner A (2015) Macrophage and microglial plasticity in the injured spinal cord. Neuroscience 307:311–318 83. Schwartz M, Ramos JMP, Ben-Yehuda H (2020) A 20-Year journey from axonal injury to neurodegenerative diseases and the prospect of immu- notherapy for combating Alzheimer’s disease. J Immunol 204:243–250 84. Sas AR, Carbajal KS, Jerome AD, Menon R, Yoon C, Kalinski AL et al (2020) A new neutrophil subset promotes CNS neuron survival and axon regeneration. Nat Immunol 21:1496–1505 85. Batlle M, Ferri L, Andrade C, Ortega F-J, Vidal-Taboada JM, Pugliese M Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : et al (2015) Astroglia-microglia cross talk during neurodegeneration in the rat hippocampus. Biomed Res Int 2015:102419–102434 fast, convenient online submission 86. Kirkley KS, Popichak KA, Afzali MF, Legare ME, Tjalkens RB (2017) thorough peer review by experienced researchers in your ﬁeld Microglia amplify inflammatory activation of astrocytes in manganese neurotoxicity. J Neuroinflamm 14:1–17 rapid publication on acceptance 87. Xu C, Su Z (2015) Identification of cell types from single-cell transcrip - support for research data, including large and complex data types tomes using a novel clustering method. Bioinformatics 31:1974–1980 • gold Open Access which fosters wider collaboration and increased citations 88. Van Dyck A, Bollaerts I, Beckers A, Vanhunsel S, Glorian N, Houcke J et al (2021) Müller glia–myeloid cell crosstalk accelerates optic nerve maximum visibility for your research: over 100M website views per year regeneration in the adult zebrafish. Glia 69:1444–1463 89. Gambarotta G, Fregnan F, Gnavi S, Perroteau I (2013) Neuregulin 1 At BMC, research is always in progress. role in Schwann cell regulation and potential applications to promote Learn more biomedcentral.com/submissions peripheral nerve regeneration. Int Rev Neurobiol 108:223–256

Journal

Acta Neuropathologica Communications – Springer Journals

Published: May 24, 2023

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Immune stimulation recruits a subset of pro-regenerative macrophages to the retina that promotes axonal regrowth of injured neurons

Immune stimulation recruits a subset of pro-regenerative macrophages to the retina that promotes axonal regrowth of injured neurons

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Immune stimulation recruits a subset of pro-regenerative macrophages to the retina that promotes axonal regrowth of injured neurons

Immune stimulation recruits a subset of pro-regenerative macrophages to the retina that promotes axonal regrowth of injured neurons

References (109)

Abstract

Journal

Recommended Articles

There are no references for this article.

Our policy towards the use of cookies