RNA binding protein SYNCRIP maintains proteostasis and self-renewal of hematopoietic stem and progenitor cells

Abstract

Tissue homeostasis is maintained after stress by engaging and activating the hematopoietic stem and progenitor compartments in the blood. Hematopoietic stem cells (HSCs) are essential for long-term repopulation after secondary transplantation. Here, using a conditional knockout mouse model, we revealed that the RNA-binding protein SYNCRIP is required for maintenance of blood homeostasis especially after regenerative stress due to defects in HSCs and progenitors. Mechanistically, we find that SYNCRIP loss results in a failure to maintain proteome homeostasis that is essential for HSC maintenance. SYNCRIP depletion results in increased protein synthesis, a dysregulated epichaperome, an accumulation of misfolded proteins and induces endoplasmic reticulum stress. Additionally, we find that SYNCRIP is required for translation of CDC42 RHO-GTPase, and loss of SYNCRIP results in defects in polarity, asymmetric segregation, and dilution of unfolded proteins. Forced expression of CDC42 recovers polarity and in vitro replating activities of HSCs. Taken together, we uncovered a post-transcriptional regulatory program that safeguards HSC self-renewal capacity and blood homeostasis.

Fig. 1. SYNCRIP is dispensable for steady-state hematopoiesis but required for long-term HSC self-renewal.

A Targeting strategy to create Syncrip conditional knockout mouse (cKO). B Immunoblots showing SYNCRIP KO in bone marrow of Syncrip^Δ/Δ mice 3 weeks post pIpC injections. ACTIN as loading control. C Bone marrow (BM) cellularity Syncrip^f/f (n = 6) and Syncrip^Δ/Δ (n = 5) mice at 3- and 24 weeks post pIpC. D Representative flow analysis of stem/progenitor compartments Syncrip ^f/f and Syncrip^Δ/Δ mice 3 weeks post pIpC. LSK (Lin-Sca+ckit + ); HSC – hematopoietic stem cell (LSK CD48-CD150 + ); MPP-multipotent-progenitor; MPP1-LSK CD48-CD150-; MPP2-LSK CD48 + CD150 + ; MPP4-LSK CD48 + CD150-. E Frequencies of LSK cells Syncrip^f/f and Syncrip^Δ/Δ mice at 3- (p = 0.047), 8- (p = 0.000015) and 24 (p = 0.011), weeks post pIpC (n = 5/genotype). F Frequencies of HSCs in Syncrip^f/f and Syncrip^Δ/Δ mice at 3-, 8- and 24 weeks post pIpC (n = 5/genotype). G, H Chimerism of donor-derived BM cells from Syncrip^f/f or Syncrip^Δ/Δ (n = 2 donor, and n = 10 recipients/genotype). G Representative flow plots H Quantitative analysis at 8- (p = 0.035), 16- (p < 0.000001) and 24- (p < 0.0005) weeks post-transplantation. I Donor chimerism in BM mature populations at 16 weeks post-transplantation: total BM (p = 0.000002) Myeloid (Mac1 + Gr1 + ) (p = 0.00016), B (B220 + ) (p < 0.000001), CD4 (CD4 + ) (p < 0.000001); CD8 (CD8 + ) (p < 0.000016) cells (n = 8/ genotype). J Donor chimerism in stem/progenitor compartments at 16 weeks post-transplantation: LSK (p = 0.000146), MP (p < 0.000001), MPP1 (p = 0.000039), MPP2 (p = 0.000011), and MPP4 (p = 0.000219) (n = 8/ genotype). K Donor chimerism in BM cells (n = 2 donor, and n = 10 recipients/genotype). Left: representative flow plots. Right: Quantitative analysis at 8-(p = 0.000942), 16-(p = 0.000338) and 24-(p = 0.000194) weeks post-transplantation. L Donor chimerism in stem/progenitor compartments at 16 weeks post-transplantation. LSK p = 0.000718, MP (p = 0.000498), HSC (p = 0.001521), MPP1 (p = 0.002307), MPP2 (p = 0.001265), MPP4 (p = 0.002243). M Donor chimerism in BM of secondary recipient at 16 weeks post-transplantation (n = 2 donor, and n = 10 recipients/genotype). LSK (p < 0.000001), MP (p = 0.000211), HSC (p = 0.00001), MPP1 (p = 0.000145), MPP2 (p = 0.000002), MPP4 (p = 0.000004). All plots show Syncrip^f/f as black circles and Syncrip^Δ/Δ is represented as blue (box or symbol) squares. Source data are provided as Source Data File. All data represent mean±s.e.m. p values were calculated by two-tailed t test unless specified. *p < 0.05, **p < 0.01, ***p < 0.001 and ns: not significant.

Fig. 2. SYNCRIP plays a critical role in stress hematopoiesis.

A Immunoblots showing SYNCRIP KO in engrafted Syncrip ^Δ/Δ BM cells 3 weeks post pIpC injections. ACTIN as loading control. B Chimerism of donor-derived BM cells in recipient mice described in (A) at: pre-pIpC, 3-(p = 0.037), 8-(p = 0.0059), 16-(p = 0.0029) and 24-(p = 0.0041) weeks post pIpC (n = 3 donor, and n = 15 recipients/genotype at pre- and 3 weeks; n = 9 at 8-, 16- and 24- weeks). C Chimerism of donor-derived cells in stem/progenitor compartments of recipient mice described in (A) at 16 weeks post pIpC. LSK (p = 0.030), MP (p = 0.051), HSC (p = 0.0254), MPP1 (p = 0.0333), MPP2 (p = 0.0058), MPP4 (p = 0.141). D Quantitative summary of frequencies of HSC and MPPs cells within CD45.2+ LSK populations of recipient mice described in (A) at 16 weeks post pIpC. HSC (p = 0.95), MPP1 (p = 0.024), MPP2 (p = 0.037), MPP4 (p = 0.46). E Chimerism of donor-derived BM cells in secondary recipients at 8-, 16- and 24 weeks post-transplantation. (n = 3 donor, and n = 15 recipients/genotype). All p < 0.000001. F Chimerism of donor-derived cells in stem/progenitor compartments of secondary recipient mice (described in E) at 16 weeks post-transplantation (n = 3 donor, and n = 15 recipients/genotype). All p-values<0.000001. G Chimerism of donor-derived cells in total BM cells of secondary recipient mice at 1 week post-transplantation (n = 7 donor, n = 5 recipient/genotype). p-value=0.0256. H Kaplan–Meier analysis of survival of WT Syncrip^f/f and KO Syncrip^Δ/Δ mice following lethal irradiation (n = 5 each). p-value=0.0035. I, J Cell cycle analysis of (I) MPP1 (G0 p = 0.00089, G1 p = 0.15, S/G2/M p = 0.00095) and (J) HSCs Syncrip^f/f and Syncrip^Δ/Δ mice (n = 6 each genotype). Left: Representative flow plots. Right: Quantitative analysis. K Cumulative graphs tracking in vitro division of HSCs over the course of 60 h. L Representative flow plots showing gating strategy for CFSE stained HSC in recipient mice (described in G) at 1 week post-transplantation. M Quantitative summary of data shown in (L). Source data are provided as Source Data File. All data represent mean ± s.e.m. p values were calculated by two-tailed t test unless specified. *p < 0.05, **p < 0.01, ***p < 0.001 and ns not significant.

Fig. 3. Single-cell RNA sequencing (scRNA-seq) uncovered an activated unfolded protein response in Syncrip deficient HSC populations.

A Identification of hematopoietic cell populations within WT Syncrip^f/f (n = 3) and KO Syncrip^Δ/Δ (n = 3) Lin-ckit+ cells based on UMAP analysis of Single-cell RNA sequencing (scRNA-seq). B Gene expression heat map of the highly expressed genes in hematopoietic stem cell-cluster 1 (HSC-C1), hematopoietic stem cell-cluster 1 (HSC-C2), multipotent-progenitor 1 (MPP1) and 2 (MPP2) populations. C Reconstruction of the lineage branching among of four early hematopoietic stem/progenitor cells HSC-C1, HSC-C2, MPP1 and MPP2 populations using diffusion pseudotime (DPT) analysis. D GSEA analysis of genes differentially expressed between HSC-C1 vs. HSC-C2 in WT Syncrip^f/f mice for enrichment of gene signatures specific for low output and high output HSC. E UMAP displays of all hematopoietic clusters of WT Syncrip^f/f and KO Syncrip^Δ/Δ scRNA-seq as described in (A). Cluster HSC-C1 shows the most shift. HSC, Ba and MEP clusters were highlighted for comparison. F Quantitative summary of frequencies of different populations defined by scRNA-seq analysis in WT Syncrip^f/f (n = 3) and KO Syncrip^Δ/Δ (n = 3). Data represent mean ± s.e.m. p values were calculated by two-tailed t test unless specified. ns not significant. G, H Volcano plots showing genes differentially expressed between Syncrip^f/f vs. Syncrip^Δ/Δ within HSC-C1 and HSC-C2 clusters. The most differentially expressed genes are highlighted. I Enrichr analysis for GO biological processes and Reactome enrichment of significant (FDR < 0.05) downregulated and upregulated genes within the HSC-C1 population of Syncrip^Δ/Δ vs. Syncrip^f/f. X-axis: -log₁₀(p value). Enrichment of downregulated targets was depicted as negative log₁₀(p) and enrichment of upregulated targets was depicted as positive log₁₀(p). p-values were calculated by Fisher’s exact test. J Enrichr analysis for GO biological processes and Reactome enrichment of significant (FDR < 0.05) downregulated and upregulated genes within the HSC-C2 population of Syncrip^Δ/Δ vs. Syncrip^f/f. X-axis: -log₁₀(p value). p-values were calculated by Fisher’s exact test. Source data are provided as Source Data File.

Fig. 4. Deletion of SYNCRIP deregulates the proteostasis network in HSCs.

A Quantitative summary of relative tetraphenylethene maleimide (TMI) fluorescent signals in HSC, MPP1, MPP2 and MPP4. TMI signals were used to quantify the unfolded proteins in single hematopoietic cells in Syncrip^f/f (n = 4) vs. Syncrip^Δ/Δ (n = 5). B OP-Puro incorporation in HSC (p = 0.00017) and MPP (p = 0.0017) populations isolated from Syncrip^f/f (n = 5) vs. Syncrip^Δ/Δ (n = 5) 1st transplant recipient mice. OP-Puro incorporation was used to quantify the level of global protein synthesis in single hematopoietic cells. C Representative histograms of PU-FITC flow analysis of HSCs from Syncrip^f/f vs. Syncrip^Δ/Δ. D Quantitative summary of PU-FITC fluorescent signals normalized to FITC control in HSC p = 0.0123, MPP1 p = 0.32, MPP2 p = 0.27 and MPP4 p = 0.12. PU-FITC signals were used to quantify the epichaperome signal of hematopoietic cells in Syncrip^f/f (n = 10) vs. Syncrip^Δ/Δ (n = 10). E Immunofluorescence (IF) staining of HSP70 and SYNCRIP in Syncrip^f/f and Syncrip^Δ/Δ HSCs (shown in brightfield). Scale bar 5 μm. F, G Quantitative summary of normalized immunofluorescence reflecting SYNCRIP (p = 0.0041) and HSP70 (p < 0.0001) protein expression in Syncrip ^f/f (n = 324) and Syncrip^Δ/Δ (n = 654)HSCs (n = 5 each genotype). H Transmission electron microscopy (TEM) images of HSCs isolated from Syncrip^f/f vs. Syncrip^Δ/Δ mice. Red arrow: endoplasmic reticulum (ER); Blue arrow: nuclear envelope (NE). I, J Quantitative summary of diameter average length (nm) of ER (p < 0.0001) (I) and NE (p = 0.0027) (J) (n = 7 each genotype; total cells Syncrip^f/f n = 20 and Syncrip^Δ/Δ n = 20). K, L Quantitative summary of normalized immunofluorescence reflecting ATF4 and CHOP protein expression in Syncrip^f/f and Syncrip^Δ/Δ HSCs (n = 5 each genotype; total cells Syncrip^f/f n = 548 and Syncrip^Δ/Δ n = 477; Syncrip^f/f n = 342 and Syncrip^Δ/Δ n = 266) All data p < 0.0001. Source data are provided as Source Data File. All data represent mean ± s.e.m. p values were calculated by two-tailed t-test unless specified. *p < 0.05, **p < 0.01, ***p < 0.001 and ns: not significant.

Fig. 5. Multi-omics analysis identifies SYNCRIP functional targets in HSCs.

A Number of SYNCRIP-HyperTRIBE significant edit sites (FDR < 0.05 and differential editing frequency (SYNCRIP-ADAR vs. control) ≥ 0.1) and their genic locations in HSC and MPP cells. B Number of target genes with edited sites in (A) in hematopoietic stem cell (HSC) and multipotent progenitor (MPP). C Venn diagram of SYNCRIP target mRNAs identified in HSC and MPP: 534 shared targets, 262 HSC unique targets and 71 MPP unique targets. D De novo motif search identifies SYNCRIP-specific binding motifs enriched in mRNA targets with edited sites in both coding regions (CDS and 3’-UTR). E Probability density function (PDF) plots showing the distance from edits sites to the nearest SYNCRIP motifs as depicted. F Enrichr analysis for GO biological processes, GO molecular function, Reactome and KEGG enrichment of HSC and MPP shared SYNCRIP target genes. X-axis: -log₁₀(p value). G Enrichr analysis for GO biological processes and Reactome enrichment of significant (FDR < 0.05) downregulated and upregulated genes in Syncrip deficient HSCs. X-axis: Enrichment of downregulated targets negative log₁₀(p) and enrichment of upregulated targets positive log₁₀(p. H Venn diagram of SYNCRIP target mRNAs with genes significantly (FDR < 0.05, FC≥2) upregulated (499 genes) and downregulated (n = 561) in Syncrip deficient HSCs. I, J GSEA analysis showing negative enrichment of proteins significantly downregulated for gene expression signatures upregulated in HSC vs. MPP, Cellular response to heat stress. NES-normalized enrichment score. K Venn diagram of SYNCRIP direct target mRNAs (n = 534) with mRNAs significantly downregulated ((FDR < 0.05, FC≤−2: n = 561) in Syncrip deficient HSCs and genes of which protein levels are significantly downregulated (FDR < 0.1, FC≤−2: n = 803). L Heat map depicting relative protein levels of 12 and 51 overlapping genes highlighted in (K) in Syncrip^Δ/Δ vs. Syncrip^f/f LSKs. Red boxes highlight GTPase/cytoskeleton associated targets. M Enrichr analysis for GO biological processes, molecular function, Reactome and KEGG enrichment of SYNCRIP target genes whose proteins are downregulated in Syncrip deficient LSKs. X-axis: log₁₀(p value). Source data are provided as Source Data File. For all enricher analysis, p-values were calculated by Fisher’s exact test.

Fig. 6. SYNCRIP controls HSC polarity and symmetric vs. asymmetric division.

A Representative images of immunofluorescence (IF) staining of SYNCRIP, CDC42 and TUBULIN in hematopoietic stem cells (HSCs). B–D Quantification of normalized IF intensity SYNCRIP (Syncrip^f/f n = 379, Syncrip^Δ/Δ n = 300) (B) and CDC42 (Syncrip^f/f n = 133, Syncrip^Δ/Δ n = 129) (C) and TUBULIN (Syncrip^f/f n = 2482, Syncrip^Δ/Δ n = 1703) (D). All p < 0.0001. E Percentage of polarized vs. unpolarized HSCs based on TUBULIN (n = 4-5/genotype; Syncrip^f/f n = 99 and Syncrip^Δ/Δ n = 175). F Quantification of normalized IF NUMB intensity in HSCs. (Syncrip^f/f n = 1319 and Syncrip^Δ/Δ n = 929, p < 0.0001). G Representative images of paired NUMB daughter assay of HSCs. H Percentage of doublet cells in each type of cell division (n = 4-5/genotype; Syncrip^f/f n = 108 and Syncrip^Δ/Δ n = 52). I Representative images of IF staining of TUBLIN, LAMP1 and DAPI in HSCs. J Quantification of normalized IF LAMP1 intensity in HSCs. (Syncrip^f/f n = 566 and Syncrip^Δ/Δ n = 566, p < 0.0001). K Representative images of paired LAMP1 daughter assay of HSCs. L Percentages of doublet cells in each type of cell division (n = 4-5 each genotype; Syncrip^f/f n = 110 and Syncrip^Δ/Δ n = 165). M Representative images of paired TMI daughter assay of HSCs. N Percentages of doublet cells in each type of cell division (n = 4-5 each genotype; Syncrip^f/f n = 71 and Syncrip^Δ/Δ n = 38). O Representative images of paired TMI and LAMP1 daughter assay of HSCs. P Percentages of asymmetric divided cells. (n = 4-5 each genotype; Syncrip^f/f n = 10 and Syncrip^Δ/Δ n = 6). Q, R Quantification of normalized IF intensity of TMI in doublet cells symmetric high LAMP1 Syncrip^f/f n = 40 and Syncrip^Δ/Δ n = 8, p = 0.018 (Q) and symmetric low LAMP1 Syncrip^f/f n = 56 and Syncrip^Δ/Δ n = 48, p = 0.022 (R). S Normalized colony numbers 1st, 2nd and 3rd plating of Syncrip^f/f and Syncrip^Δ/Δ LSK cells transduced with empty vector (EV) or expressing CDC42 (CDC42-OV) (n = 7/condition). T Percentages of polarized vs. unpolarized cells based on TUBULIN (Syncrip^f/f-EV n = 20; Syncrip^Δ/Δ-EV n = 29; Syncrip ^f/f CDC42-OV n = 22; and Syncrip^Δ/Δ CDC42-OV n = 19). Scale bars 5μm. Source data are provided as Source Data File. All data represent mean ± s.e.m. p values were calculated by two-tailed t-test. *p < 0.05,^**p < 0.01, ^***p < 0.001 and ns: not significant.

Fig. 7. Graphical abstract illustrating SYNCRIP controls proteome quality and CDC42-mediated cell polarity and division to maintain self-renewal of reserve HSCs.

Model showing mechanism for how SYNCRIP regulates HSPC properties. Left panel shows normal conditions with SYNCRIP and right panel shows HSPC conditions without SYNCRIP. Figure created with BioRender.com.