Fig 2: Sulf2 deletion results in improved oligodendrocyte recruitment and differentiation and is upregulated in chronic active MS lesions.Tamoxifen-dependent OPC-specific ablation of either Sulf1, Sulf2 alone, or both Sulf1/2 in NG2+ OPCs was initiated prior to demyelination. Control animals (lacking cre) were treated in an identical manner. Widefield immunofluorescence for Olig2 (green) and CC1 (red) at 7 d.p.l. (a–d). The density (cells/mm2) of Olig2+ oligodendrocytes lineage cells (e), Olig2+CC1+ postmitotic oligodendrocytes (f) and percentage of CC1+ oligodendrocytes (g) was quantified (n = 5 for each group except n = 6 for Sulf1/2 cKO). Mean ± SEM is shown. Two-way ANOVA for Sulf1 and Sulf2 genotypes. Holm–Sidak post-test vs. wild-type control *, **, ***, and *** indicates p < 0.05, 0.01, 0.001, and 0.0001, respectively. h–j Human primary OPCs infected with lentiviral SULF2 knockdown (KD) or scrambled control cultured in mitogen-free conditions for 4 days. Oligodendrocyte differentiation assessed by O4 immunocytochemistry (green) and DAPI (blue) (h, i). j SULF2 KD accelerated O4+ oligodendrocyte differentiation (****p < 0.0001 paired t test, n = 6 fetal human samples). k PLP1 immunohistochemistry of chronic active human MS lesion and surrounding normal white matter (NAWM). l RNAscope for PLP1 mRNA in NAWM (pink box in k). m–p Confocal-based RNAscope in situ hybridization at lesion border (black box in k) for SULF2 (green), PDGFRA (red, n), and PLP1 (red, p) mRNA. Arrow represents colocalization of SULF2 with PDGFRA and PLP1 mRNAs. Scale: 20 µm (a–d, m–p), and 50 µm (i, j).

Fig 3: Sulfatase promotes inhibitory WNT and BMP signaling on oligodendrocyte differentiation following demyelination.Pharmacological manipulation of WNT and BMP signaling was performed in the context of Sulf1/2 cKO. WNT pathway activator (CHIR-99021, 3 µM), WNT antagonist (XAV939, 100 nM), BMP pathway activator (A01, 100 nM), or receptor antagonist (LDN-193189, 100 nM) were co-injected with lysolecithin. a–j Oligodendrocyte lineage cell density was assessed by Olig2 (green) and CC1 (red) widefield immunofluorescence. Density of Olig2+ oligodendrocyte lineage cells (k), Olig2+CC1+ postmitotic oligodendrocyte (l), and percentage of CC1+ oligodendrocytes (m) quantified (k–m, n = 8, 5, 4, 6, 8, 5, 4, 7, 5, and 3 mice from left to right, for ctrl, CHIR, XAV, A01, and LDN treatment groups, in WT and Sulf1/2 cKO mice respectively). Two-way ANOVA was performed (see Source data for full details). Holm–Sidak multiple comparisons tests vs. ctrl wild-type (blue *) or ctrl Sulf1/2 cKO (red *), or as indicated. *, **, ***, and **** indicated p < 0.05, <0.01, <0.001, and <0.0001 respectively. n hOPCs were co-infected with lentiviral SULF2 shRNAi or scrambled control and lentiviral TCF/LEF reporter virus and exposed to WNT3A (50 ng/mL) or vehicle control. Two-way RM ANOVA followed by Holm–Sidak post-test (n = 4 fetal human samples, mean ± SEM normalized to control). o hOPCs infected with lentiviral BMP reporter and SULF2 KD virus. SULF2 KD reduced BMP activity in hOPCs following BMP7 treatment (50 ng/mL) (n = 5, p < 0.05 two-way RM ANOVA Holm–Sidak post-test). Two-way ANOVA revealed significant main effects for ligand treatment, SULF2 KD, and the interaction for both WNT and BMP. p quantification of O4+ oligodendrocyte differentiation following SULF2 KD and BMP7 treatment (5–50 ng/mL BMP7) (n = 5 at 5 ng/mL BMP7 and n = 6 at 0 and 50 ng/mL BMP7 individual fetal human samples). SULF2 knockdown attenuated the effects of WNT/BMP signaling and significantly increased differentiation of hOPCs in the presence of BMP (mixed-effects model; Sidak’s post-test, p < 0.05). * and ** indicate p < 0.05 and < 0.01, respectively. Mean ± SEM is shown. Scale: 20 µm (a–j). n.s. Not significant.

Fig 4: Conditional genetic ablation of OPC-expressed Sulf1/2 accelerates oligodendrocyte recruitment and lesion colonization following demyelination.a–j Dual RNAscope in situ hybridization (ISH) of mouse spinal cord (a–e) revealed increased Sulf1 and Sulf2 mRNA expression following lysolecithin-induced demyelination (f–j; 5 d.p.l.). Sulf1 (gray), Sulf2 (green), or Pdgfra mRNA (red) and DAPI (blue). Yellow arrows denote Pdgfra+ OPCs that co-expresses Sulf1/2. Blue arrows denote Sulf2+Pdgfra+ OPCs or Sulf1+Pdgfra+ OPCs. RNAscope images were imaged by confocal microscopy, others using widefield epifluorescence. Inducible NG2creER-mediated OPC-specific ablation of Sulf1/2 (Sulf1/2 cKO) in adult mice followed by lysolecithin-induced demyelination (k–r). Analysis of OPC dynamics at 5 days post lesion (d.p.l.) (o–p) (n = 6 and 3 mice, for WT and Sulf1/2 cKO respectively) and 7 d.p.l. (q–r) (n = 5 and 6 mice, for WT and Sulf1/2 cKO, respectively). At 5 d.p.l., Olig2+ cell density was significantly increased in Sulf1/2 cKO (o) (**p = 0.0016, two-sided t test). No difference in the proportion of proliferating EdU+Olig2+ cells compared to wild type (p). At 7 d.p.l., Sulf1/2 cKO exhibited increased overall Olig2+ and postmitotic Olig2+CC1+ oligodendrocyte density (m, n) (cells/mm2 quantified in q, **p = 0.0027 and 0.00071, two-sided t test). The percentage of CC1+Olig2+ oligodendrocytes among Olig2+ cells was increased in Sulf1/2 cKO animals compared to controls (r, ***p = 0.0009, two-sided t test). Mean ± SEM shown. Scale: 20 µm (a, f, c, h), 40 µm (b, g, d–j), and 10 µm (k–n).