Monocyte-secreted Wnt reduces the efficiency of central nervous system remyelination

Abstract

The regeneration of myelin in the central nervous system (CNS) reinstates nerve health and function, yet its decreased efficiency with aging and progression of neurodegenerative disease contributes to axonal loss and/or degeneration. Although CNS myeloid cells have been implicated in regulating the efficiency of remyelination, the distinct contribution of blood monocytes versus that of resident microglia is unclear. Here, we reveal that monocytes have non-redundant functions compared to microglia in regulating remyelination. Using a transgenic mouse in which classical monocytes have reduced egress from bone marrow (Ccr2-/-), we demonstrate that monocytes drive the timely onset of oligodendrocyte differentiation and myelin protein expression, yet impede myelin production. Ribonucleic acid sequencing revealed a Wnt signature in wild-type mouse lesion monocytes, which was confirmed in monocytes from multiple sclerosis white matter lesions and blood. Genetic or pharmacological inhibition of Wnt release by monocytes increased remyelination. Our findings reveal monocytes to be critical regulators of remyelination and identify monocytic Wnt signaling as a promising therapeutic target to inhibit for increased efficiency of CNS regeneration.

Fig 1. Classical monocytes regulate central nervous system remyelination.

(A) Schematic of in vivo LPC-induced focal demyelinated lesions in corpus callosum of adult mice. (B) Schematic of use of lesioned Ccr2^RFP/+ reporter mice to track classical monocyte entry into brain lesions. (C) Mean numbers of Ccr2-RFP+ cells per lesion image at 10 and 21 days post-lesion (dpl) ± S.E.M. ***P = 0.0003; unpaired 2-tailed Student t test. n = 3 mice/group. (D) Representative immunofluorescent images of Ccr2-RFP+ cells in a corpus callosum lesion (left, outlined), *indicates needle tract. Non-lesion corpus callosum (right) stained with IBA1 (white) and counterstained with Hoechst (blue), **indicates choroid plexus monocyte-derived macrophages. Scale bar, 100 µm. (E) Flow cytometry plot demonstrating differential CD45 expression in CD11b⁺ Ly6G⁻ CD3⁻ cells, used to distinguish between lesion microglia (CD45^lo) and monocytes (CD45^hi) at 10 dpl. SSC: side scatter. Two mice pooled. (F) Mean numbers of lesion CD45^lo microglia and CD45^hi monocytes ± S.E.M. at 10 dpl; **P = 0.0080; unpaired 2-tailed Student t test. n = 6 mice/group, 2 mice pooled per replicate. (G) Schematic of use of Ccr2^−/− mice to impair the egress of classical monocytes from the bone marrow into the blood and subsequently into brain lesions. (H) Mean proportion of monocytes (CD11b+ cells which are CD45^hi) ± S.E.M. in sham PBS-injected corpus callosum, and lesions from wild type (WT) and Ccr2^−/− mice. n = 6 mice/group, 2 mice pooled per replicate. One-way analysis of variance (ANOVA) with Tukey’s multiple comparisons; *P = 0.0445 (sham vs. WT LPC), *P = 0.0356 (WT vs. Ccr2^−/− LPC), I) Mean number of monocytes (CD11b⁺ CD45^hi Ly6C^hi F4/80⁻ cells) ±  S.E.M. in sham PBS-injected corpus callosum, and lesions from WT and Ccr2^−/− mice. n = 6 mice/group, 2 mice pooled per replicate. One-way ANOVA with Tukey’s multiple comparisons; **P  = 0.0040 (WT LPC vs. Ccr2^−/− LPC), *P = 0.0439 (sham PBS vs. Ccr2^−/− LPC). (J) Representative immunofluorescent labeling of oligodendrocyte lineage cells (Olig2+, red) which are mature (CC1+, green) in WT and Ccr2^−/− mice at 10 dpl, counterstained with Hoechst. Scale bar, 20 μm. (K) Mean percentage of Olig2⁺ cells which are CC1⁺ per field ± S.E.M. in WT and Ccr2^−/− mice at 10 dpl; *P = 0.0156; unpaired 2-tailed Student t test. n = 3–4 mice/group. (L) Representative immunofluorescent labeling of myelin-associated glycoprotein (MAG, red) in WT and Ccr2^−/− mice at 10 dpl, counterstained with Hoechst. Scale bar, 20 μm. (M) Quantification of MAG intensity/mm² within lesioned corpus callosum at 10 dpl ± S.E.M. in WT and Ccr2^−/− mice, *P = 0.0135; unpaired 2-tailed Student t test. n = 4–5 mice/group. (N) Mean number of myelinated axons/mm² at 10 dpl ± S.E.M. in WT and Ccr2^−/− lesions quantified from electron micrographs. P = 0.6847; unpaired 2-tailed Student t test. n = 3–6 mice/group. (O) Quantification of MAG intensity/mm² within lesioned corpus callosum at 21 dpl ± S.E.M. in WT and Ccr2^−/− mice, *P = 0.0317; Mann–Whitney test. n = 4–5 mice/group. (P) Mean percentage of Olig2+ cells which are CC1+ per field ± S.E.M. in WT and Ccr2^−/− lesions at 21 dpl; P = 0.2879; unpaired 2-tailed Student t test. n = 5–7 mice/group. (Q) Mean number of myelinated axons in lesions at 21 dpl ± S.E.M. in WT and Ccr2^−/− lesions; **P = 0.0054; unpaired 2-tailed Student t test. n = 3–4 mice/group. (R) Average g-ratio at 21 dpl ± S.E.M. in WT and Ccr2^−/− lesions; P = 0.8466; unpaired 2-tailed Student t test. n = 3 mice/group. (S) Representative electron micrographs of lesioned corpus callosum at 21 dpl in WT and Ccr2^−/− mice. Scale bar, 2 µm. Mice were 8–12 weeks of age in each group. Source data may be found in S1 Data.

Fig 2. Monocytes have a Wnt signature during remyelination.

(A) Mean expression in Fragments per Kilobase of Transcript per Million mapped reads (FPKM) of microglia-enriched genes in lesion monocytes (red) and microglia (blue) at 10 dpl. **P = 0.0026; 2-tailed paired Student t test. n = 2–3 mice/group. (B) Venn diagram of top 10% of all expressed genes in lesion microglia and monocytes at 10 dpl. (C) Log2 fold change (FC) in top 50 enriched genes in microglia (left) and monocytes (right). Values were generated by comparing microglia transcriptomes to monocyte transcriptomes; the positive values indicate an upregulation (enrichment) in microglia, whereas negative values indicate a downregulation in microglia, therefore, an enrichment in monocytes. Red indicates highest enrichment and purple indicates the lowest enrichment. n = 2–3 mice/group. (D) Volcano plot showing significance as –Log10 transformed P values against Log2FC. Black indicates no significant change, red indicates genes upregulated in monocytes vs. microglia, and blue indicates genes upregulated in microglia vs. monocytes. P value cutoff 0.01. n = 2–3 mice per group. (E) Pathway analysis of genes upregulated in monocytes vs. microglia at 10 dpl. (F) Average FPKM of Wnt ligand genes in monocytes (red) and microglia (blue) in lesions at 10 dpl. ****P < 0.0001; one sample t test against microglia value of 0. (G) Average FPKM of Wnt signaling genes in monocytes (red) and microglia (blue) in lesions at 10 dpl. * P = 0.0152; one sample t test against microglia value of 0. (H) RFP+ lesion monocytes (red) expressing Axin2 (green) at 10 dpl, counterstained with myeloid cell nuclear marker PU.1 (purple). Inset, Axin2 isotype control. Scale bar, 100 µm. (I) Mean number of RFP+ cells expressing Axin2 ± S.E.M. at 10 and 21 dpl in Ccr2^RFP/+ reporter mouse lesions. *P = 0.00111; unpaired 2-tailed Student t test. n = 3 mice/time point. (J) Mean percentage of Olig2+ cells expressing Axin2 ±  S.E.M. at 21 dpl in WT and Ccr2^−/− mice; * P = 0.0162; unpaired 2-tailed Student t test. n = 3 mice/group. (K) Oligodendrocyte lineage cells (Olig2+; purple) expressing Axin2 (green) at 21 dpl. Scale bar, 50 µm. Mice were 8–12 weeks of age in each group. Source data may be found in S1 Data.

Fig 3. Wnt signature in MS monocytes.

(A) Expression of Wnt-related genes in monocyte cluster 1, sequenced in the rim of mixed active MS lesions, represented as percentage of total cluster genes (left) and Log2 fold change (FC) (right). (B) Expression of Wnt-related genes in monocyte cluster 2, sequenced in the rim of mixed active MS lesions, represented as percentage of total cluster genes (left) and Log2FC (right). (C) Mean expression of Wnt-related genes in blood monocytes from individuals with relapse-remitting MS (RRMS; EDSS ≤ 2; n = 5) and secondary progressive MS (SPMS; EDSS ≥ 6; n = 6), which showed a ≥ 2-fold increase over healthy control (n = 4), represented as fold change over control. Brown-Forsythe and Welch ANOVA with Dunnet’s T3 multiple comparisons test; P = 0.0202 (BCAT1), 0.006 (HIF1A), 0.0058 (HSPH1), 0.0097 (NAMPT), <0.0001 (FXR1), 0.0014(BCAS2), 0.003 (CLK1), 0.0002 (ERO1A), 0.0050 (SNAP23), 0.0012 (FMNL2), 0.0002 (P4HA1), 0.0002 (CACYBP), 0.0001 (TCP1), 0.0084 (ZNF331). Source data may be found in S1 Data.

Fig 4. Wnt signaling in monocytes regulates remyelination efficiency.

(A) Csf1r-iCre;Porc^fl/fl were lesioned to assess the impact on remyelination. (B) Representative images of myelin basic protein (MBP) expression within lesions (outlined) at 14 dpl in Csf1r-iCre;Porc^fl/fl compared to Porc^fl/fl controls. Scale bar, 25 µm. (C) Mean percentage area of MBP staining ± S.E.M. in corpus callosum of Csf1r-iCre;Porc^fl/fl compared to Porc^fl/fl controls at 14 dpl. *P = 0.0130, one-tailed t test against control value of 1. n = 3–5 mice/group. (D) Representative immunofluorescent labeling of oligodendrocyte lineage cells (Olig2⁺, green) which are mature (CC1⁺, magenta) in Csf1r-iCre;Porc^fl/fl compared to Porc^fl/fl control mice at 14 dpl, counterstained with Hoechst. Scale bar, 20 μm. (E) Mean percentage of Olig2⁺ cells which are CC1⁺ per field ± S.E.M. in Csf1r-iCre;Porc^fl/fl compared to Porc^fl/fl controls at 14 and 21 dpl; ^*P = 0.1210 (14 dpl), 0.1741 (21 dpl); unpaired 2-tailed Student t test. n = 3 mice/group. (F) Schematic of transplant of C59- or Vehicle-treated GFP⁺ monocytes into lesioned Ccr2^−/− mouse circulation. (G) GFP⁺ transplanted cells detected in lesions of Ccr2^−/− mice, following ex vivo treatment with vehicle or C59, and stained for Axin2 (red), at 10 dpl. Double positive cells indicated with arrows. Scale bar, 25 µm. (H) Representative images of MAG expression within lesions at 10 dpl in lesioned Ccr2^−/− mice transplanted with C59- or vehicle-treated monocytes. Scale bar, 25 µm. (I) Mean percentage area of MAG staining ± S.E.M. in C59-treated monocyte transplanted Ccr2^−/− mice at 10 dpl, normalized to vehicle-treated monocyte transplanted control. ^* P = 0.0131, one sample t test against value of 1. n = 4–6 mice/group. (J) Mean percentage of Olig2⁺ cells which are CC1⁺ per field ± S.E.M. in C59- or vehicle-treated monocyte transplanted Ccr2^−/− mice at 10 dpl. ^*P = 0.4217; unpaired 2-tailed Student t test. n = 3–5 mice/group. (K) Representative immunofluorescent labeling of oligodendrocyte lineage cells (Olig2+, green) which are mature (CC1+, magenta) in Vehicle-treated and C59-treated monocyte transplanted mice at 10 dpl, counterstained with Hoechst. Scale bar, 20 μm. (L) Graphical abstract of results: Classical monocytes support oligodendrocyte differentiation and myelin protein expression, yet impede myelin production during remyelnation in a Wnt-dependent manner. Csf1r-iCre; Porc^fl/fl were 8–14 weeks old. Ccr2^−/− in transplant experiments were 8–16 weeks old. Source data may be found in S1 Data.