Figure 5. CD44 mediates C1q-induced ERK signaling activation and chemotaxis in hNSC in vitro.

(A,B) CD44 expression profile in WT but not in CD44 KO hNSC was verified by (A) western blot and (B) Immunocytochemistry (scale bar 30 µm). (C,D) p-ERK western blot analysis of protein lysates from WT and CD44 KO hNSC shows that C1q-induced ERK activation is completely blocked in CD44 KO hNSC at C1q [0.1 nM] and [1 nM], and dramatically attenuated at C1q [200 nM]. p-ERK band optical intensities were normalized to β-actin within condition. Data show mean ± SEM (N = 3 biological replicates per condition) normalized to untreated controls for comparison between conditions (dashed line). Statistical analysis using one-sample t-test (NS, not significant; *p≤0.05) for comparison with control and Student’s t-test for comparison between WT and CD44 KO hNSC as indicated (*p≤0.05, **p≤0.01). (E) Transwell chemotaxis assay reveals that C1q-induced hNSC chemotaxis is completely blocked in CD44 KO hNSC at C1q [0.1 nM] and dramatically attenuated at C1q [200 nM]. Data show mean ± SEM (N = 3 biological replicates per condition) normalized to untreated controls (dashed line). Statistical analysis using one-sample t-test (NS, not significant; *p≤0.05) for comparison with control and Student’s t-test for comparison between WT and CD44 KO hNSC as indicated (*p≤0.05, **p≤0.01).

Figure 5—figure supplement 1. WT and CD44 KO hNSC generation.

The human CD44 loci (gene ID: 960) was targeted using the CRISPR/Cas9 system to create genetic knockout via non-homologous end joining (NHEJ). (A) CD44 Guide RNA sequences. For maximum disruption of CD44’s ORF, two guide RNAs (gRNAs) were simultaneously introduced into hNSC with chimeric guide RNA + hspCas9 co-expression vector. (B) hspCas9 co-expression vector. (C) Snap Gene Graphics for gRNA alignment. Both gRNAs (purple) were designed to target the second exon of CD44 to generate a 47 bp deletion within exon two resulting in a frame-shift that introduces a pre-mature stop codon within a targeted exon. (D) WT and CD44 KO FACS sorting after CRISPR. Two weeks after plasmid introduction, unsorted CRISPR-modified hNSC were separated based on CD44 expression using fluorescence activated cell sorting (FACS); CD44-negative knockout cells (29.7% of the total population) were isolated based on the gating of our unstained control; CD44-positive wild-type cells (39.3% of the total population) were isolated based on the gating of our positive control, unmodified CD44-PE-stained cells. FACS-isolated CD44 WT and CD44 KO hNSC retain expression identity after multiple passages. (E) Flow citometry analysis of CD44 WT and KO hNSC. After more than a month in culture, and having undergone four passages, we re-analyzed the expression of CD44 in WT and KO cells. FACS-isolated CD44 WT hNSC retained CD44 expression; 96.3% CD44-PE positive and 1.87% CD44-PE negative. FACS-isolated KO hNSC retained a CD44-negative expression phenotype; 99.7% CD44-PE negative and 0.24% CD44-PE positive. (F) Fluorescent images of CD44 WT immediately after flow cytometry. (G) Fluorescent images of CD44 KO immediately after flow cytometry. After CD44 -PE staining, hNSC were counterstained with Hoechst to label cell nuclei and then mounted on a slide using a Cytospin 4 centrifuge (Thermo Scientific). Scale bars 30 µm. Images were taken in an inverted fluorescent microscope at 20×.

Figure 5—figure supplement 2. CD44 WT and CD44 KO hNSC genotyping by sequencing.

(A) Human CD44 locus: Location of guide RNAs (gRNA) and genotyping primers. CD44-WT or CD44-KO hNSC generated by NHEJ with CRISPR/Cas9 system. Two primer sets (forward primer in green), (reverse primer in red) were used to amplify and/or sequence the region surrounding the targeted site. Graphics were produced using SnapGene software. The targeted site was then amplified using a high-fidelity polymerase and highly specific primer set (CD44_2 F/CD44_2R); this primer set was optimized to produce a single amplicon of 422 bp. This amplicon was then purified and submitted for Sanger sequencing to Retrogen Inc; each amplicon was sequenced with four different primers. The raw chromatograph data from the Sanger sequencing was then aligned with the targeted site using SnapGene software. (B) Alignment of CD44 WT sequencing results. FACS isolated CD44-positive hNSC showed no genetic changes within the targeted exon (no double peaks at any positions). (C) Alignment of CD44 KO sequencing results. FACS isolated CD44-negative hNSC contained a 47 bp deletion within exon 2 resulting in a frame-shift that introduced a premature stop codon within targeted exon; the same deletion was consistently detected in all traces. (D) Table containing all of the sequences used in producing the data of this figure.

Figure 5—figure supplement 3. Karyotype, stemness, proliferation capacity, and multipotency are unaltered after CRISPR Cas9 genetic deletion of CD44 in hNSC.

(A,B) CD44 WT and KO hNSC are karyotypically normal at p7. CD44 WT/KO hNSC were submitted to Cell Line Genetics, for G-banding karyotyping. (A) FACS-isolated CD44 wildtype (WT) G-banding karyotyping. (B) FACS-isolated CD44 knockout (KO) G-banding karyotyping. (C,D) CD44 WT and CD44 KO hNSC express similar levels of CD133+ cells. (C) CD44 WT and (D) CD44 KO hNSC were assessed for expression of the ‘stemness’ marker CD133. Relative to the unstained control, at passage 7, CD44 WT and CD44 KO hNSC express similar levels of CD133 as measured by flow cytometry; 96.1% and 93.5%, respectively. (E-G) CD44 WT and CD44 KO hNSC exhibit similar proliferation levels. Representative image of EdU (green) + nuclei and Hoechst counterstain (blue) of (E) CD44 WT and (F) CD44 KO hNSC. White arrows indicate examples of EdU+ nuclei, scale bars 30 µm. (G) EdU incorporation quantification at 2 d in vitro (DIV). Data shown as proportion of EdU+ cells for CD44 WT (blue column) and KO (red column), Mean + SEM. Student’s t-test, NS = not significant p=0.1126, N = 4 biological replicates. (H-S) Genetic deletion of CD44 in human NSC did not alter multipotency of hNSC but increased oligodendroglial differentiation in vitro. After 14DIV under differentiation conditions, cells were immunostained for an oligodendroglial marker (Olig2+, green). A second set of cells was double immunostained for astroglial (GFAP+, red) and neuronal (β tubulin III+, green) markers. Representative image of Olig2 immunostaining in (H) WT and (I) CD44 KO hNSC. White arrows indicate examples of Olig2+ nuclei, scale bars 30 µm. (J) Olig2+ nuclei quantification (%). Representative image of GFAP immunostaining in (K) CD44 WT and (L) CD44 KO hNSC. Pink arrows indicate examples of cytoplasmic GFAP+/β tubulin III- astroglial cells, scale bars 30 µm. (M) GFAP+/β-tubulinIII- astroglial cell quantification (%). Representative image of β-tubulinIII immunostaining in (N) CD44 WT and (O) CD44 KO hNSC. Green arrows point to examples of cytoplasmic GFAP-/β tubulinIII+ neuronal cells, scale bars 30 µm. (P) GFAP-/β tubulin III+ neuronal cell quantification (%). Representative image of GFAP and β tubulinIII double immunostaining in (Q) CD44 WT and (R) CD44 KO hNSC. White arrows point to examples of GFAP+/β tubulinIII+ undecided double positive cells, scale bars 30 µm. (S) GFAP+/β tubulin III+ double + undecided cell quantification (%). Data represent average percentage ± SEM of quantified positive cells, obtained from 10 random pictures/experiment, from N = 4 biological replicates (averaging 140 ± 60 cells/picture). Statistical analysis was performed using Student’s t-test between WT and CD44 KO hNSC; NS: not significant p≥0.2455, ****p≤0.0001.

Figure 5—figure supplement 4. CD44 WT and KO mNSC cell line generation.

Mouse NSC (mNSC) from CD44 WT and KO mice were derived from single embryo cortices at E11.5–12. DNA of parent mice and each littermate embryo was extracted from tail snips and direct PCR performed for genotyping. (A) Primers used for genotyping CD44 WT, KO, and heterozygous embryos. (B) Genotyping results of WT and CD44KO parents and first-generation heterozygous CD44 +/– colonies. (C) Genotyping results of WT and CD44 KO embryos that were used to generate CD44 WT and KO mNSC lines. (D) CD44 expression in CD44 WT and CD44 KO mNSC by immunostaining verifies positive staining only in WT cells. (E) mNSC express all C1q novel candidate receptors. mRNA RT-PCR analysis in mNSC reveals the positive expression of all C1q novel signaling candidates. (F) mNSC migrate toward PMN-CM and C1q similar to hNSC in vitro. Transwell chemotaxis assay shows that PMN-CM and C1q [200 nM] but not Mφ-CM induce mNSC chemotaxis. Data normalized to untreated control (dashed line; N = 3 biological replicates per condition with technical triplicates). Statistical analysis using one-sample t-test (NS, not significant; *p≤0.05, ***p≤0.001) for comparison with control. ANOVA analysis for comparison between groups, p=0.0018. (G) CD44 WT but not KO mNSC migrate toward C1q [200 nM] similar to hNSC in vitro. Transwell chemotaxis assay shows that C1q [200 nM] induces mNSC chemotaxis only in CD44 WT cells, while KO cells are repulsed by C1q. Both effects are blocked by temperature inactivation of C1q. Data normalized to untreated control levels (dashed line; N = 3 biological replicates per condition with technical triplicates). Statistical analysis using one-sample t-test (NS, not significant; *p≤0.05, ***p≤0.001) for comparison with control, ANOVA analysis for comparison between groups, ****p≤0.0001, and Student’s t-test for comparison between WT and CD44 KO mNSC as indicated (***p≤0.001).