Aging-induced pseudouridine synthase 10 impairs hematopoietic stem cells

Abstract

Aged hematopoietic stem cells (HSC) exhibit compromised reconstitution capacity and differentiation-bias towards myeloid lineage, however, the molecular mechanism behind it remains not fully understood. In this study, we observed that the expression of pseudouridine (Ψ) synthase 10 is increased in aged hematopoietic stem and progenitor cells (HSPC) and enforced protein of Ψ synthase 10 (PUS10) recapitulates the phenotype of aged HSC, which is not achieved by its Ψ synthase activity. Consistently, we observed no difference of transcribed RNA pseudouridylation profile between young and aged HSPC. No significant alteration of hematopoietic homeostasis and HSC function is observed in young Pus10-/- mice, while aged Pus10-/- mice exhibit mild alteration of hematopoietic homeostasis and HSC function. Moreover, we observed that PUS10 is ubiquitinated by E3 ubiquitin ligase CRL4DCAF1 complex and the increase of PUS10 in aged HSPC is due to aging-declined CRL4DCAF1- mediated ubiquitination degradation signaling. Taken together, this study for the first time evaluated the role of PUS10 in HSC aging and function, and provided a novel insight into HSC rejuvenation and its clinical application.

Introduction

Hematopoietic stem cells (HSC) generate all of the blood cells throughout their life-span.^1,2 During aging, the function of HSC declines, featured as compromised reconstitution capacity and differentiation skewing towards myeloid line-age.^3,4 Although previous studies have identified various molecular signaling pathways promoting HSC aging,^5-8 the exact molecular mechanism is still not fully understood.

It has been known for several decades that more than 170 different types of chemical modifications to RNA exist.⁹ Pseudouridine (Ψ), known as “the fifth nucleotide” in RNA, was first identified in 1951 and is the most abundant posttranscriptional RNA modification (with an estimated c/U ratio of 7–9%).^10-12 Ψ is generated from isomerization of uridine, which is catalyzed by Ψ synthases.^13-15 Ψ plays an important role in various aspects of RNA biology, and, therefore, participates in many biological process, including translational control,^16,17 RNA folding,^18-22 protein translation,^23-26 and clinical diseases.^27-32 A recent study revealed that dysfunction of PUS7 blocks the differentiation of HSC due to the lack of pseudouridylation of mTOG tRF.¹⁶ Moreover, the expression of PUS7 is decreased in hematopoietic stem and progenitor cells (HSPC) of patients with myelodysplastic syndrome and delivery of pseudouridylated mTOG to HSPC of myelodysplastic syndromes patients improves their colony formation capacity and differentiation potential.³³ In addition, DKC1 is required for accurate HSC differentiation and maintenance of HSC function.^34,35 The above-mentioned studies reveal the importance of Ψ in modulating HSC differentiation and malignancies, however, whether Ψ participates in HSC aging has never been investigated.

In this study, we observed that the protein of Ψ synthase 10 (PUS10) is increased in aged HSPC. By conducting in vivo functional assay, we observed that enforced PUS10 impairs the reconstitution capacity of HSPC, which is independent of their Ψ synthase activity. By profiling the Ψ landscape in HSPC, we observed no difference in Ψ between young and aged HSPC at detectable locations. Moreover, we observed that PUS10 interacts with E3 ubiquitin ligase CRL4^DCAF1 complex and is ubiquitinated by this complex. Aging-declined CRL4^DCAF1 results in the accumulation of PUS10 in HSPC. Taken together, this study for the first time elucidated the role of PUS10 in HSC aging and function, and provided novel insights into HSC rejuvenation and its clinical application.

Methods

Mice

C57BL/6 mice (CD45.2), C57BL/6-SJL (CD45.1) mice were obtained from the Jackson Laboratory. Pus10^-/- mice were kindly provided by Dr. Mo Li, Peking University Third Hospital, Beijing. Pus10^-/- mice on C57BL/6N background were generated by deleting the 2^nd exon using CRISPR-Cas9 system. The genomic RNA used to generate Pus10^-/-mice are listed in the Online Supplementary Table S2. The genotyping primers are listed in the Online Supplementary Table S3. The recipient mice (CD45.1/2) in the competitive transplantation assays were the first generation of C57BL/6 (CD45.2) and B6.SJL (CD45.1) mice. All mice were housed in specific pathogen-free conditions. All procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of Tsinghua University.

Small RNA demethylase–pseudouridine-sequencing

Small RNA (<200 nt) was extracted and purified using miRNeasy Mini Kit (Qiagen) and RNeasy MinElute Cleanup Kit (Qiagen). Purified small RNA fragments were demethylated by wild-type and mutant AlkB, purified by phenol/chloroform extraction and ethanol precipitation, labeled by CMC. The demethylation reaction and CMC labeling were performed as described.¹⁴ Briefly, 50 ng small RNA was denatured at 65°C for 5 minutes (min) and demethylated by wild-type and D135S mutant AlkB. The purified small RNA was denatured at 80°C for 5 min, added to BEU buffer with or without CMC, incubated at 37°C for 20 min, then purified by ethanol precipitation. The purified RNA was dissolved in sodium carbonate buffer and shaken at 37°C for 6 hours. The library was established as described.^14,36 Briefly, small RNA was dephosphorylated with CIP (NEB). The 3' adaptor ligation was added with T4 RNA ligase2, truncated KQ (NEB), followed by 5' Deadenylase (NEB) and RecJf (NEB) digestion. RNA was reverse transcribed by SuperScript III reverse transcriptase (Invitrogen), then digested by RNase H. The 5' adaptor ligation was added with T4 RNA ligase 1, high concentration (NEB). The ligated cDNA was amplified by Q5 High-Fidelity 2× Master Mix (NEB). The purified libraries were sequenced on Illumina NovaSeq 6000.

Identification of pseudouridine sites and levels on transcribed RNA

For transcribed RNA (tRNA) demethylase-pseudouridine sequencing (DM-Ψ-seq) data, the analysis was performed as described before.^14,36 Briefly, the adapter sequences of reads two were trimmed with Trim-galore v0.6.5 (parameters: -q 20--phred33--length 25-e 0.1--stringency 3). Polymerase chain recation (PCR) duplication was removed with Fastx_toolkit v0.0.14 before discarding the random barcode in the 5’ end. Processed reads were further mapped to the genomic tRNA sequences from GtRNA database (http://gtrnadb.ucsc.edu/genomes/eukaryota/Mmusc10/) with Bowtie2 v2.3.5 (parameters: bowtie-a--best--strata--chunkmbs 2000). In order to identify the pseudouridine sites of tRNA, the following criteria were considered: i) the Ψ sites appearing in all independent replicates; ii) stop rate <1% in the BEU sample; iii) CMC coverage >50; iv) stop reads number >5 in the CMC sample; v) stop rate (CMC-BEU) difference >4%; vi) Fisher test adjusted P value <0.05. Finally, the Ψ level change for tRNA between young and old HSPC was evaluated and visualized via R package ggplot2.

Statistical analysis

Data are shown as mean ± standard deviation (SD). Student’s t test (two-tailed unpaired) was used for comparisons between the groups using GraphPad Prism 6.0 software.

Results

PUS10 is increased in aged hematopoietic stem progenitor cells

Due to the scarcity of HSC, we purified HSPC by using the combination of CD48^-LSK(cKit⁺ Sca1⁺ Lin^-) according to previous reports.^{37-3 9} The protein of PUS10 between young and aged HSPC was examined by western blot and it showed that PUS10 is increased upon aging (Figure 1A, B). In order to further confirm this result, we investigated the expression of PUS10 in a database, wherein the researchers compared the proteomic profile between young and aged HSC (CD34^-CD150⁺Flt3^-LSK).⁴⁰ The result shows that PUS10 is indeed increased in aged HSC (Figure 1C).

Enforced PUS10 impairs the reconstitution capacity of hematopoietic stem and progenitor cells

In order to further investigate whether the increase of PUS10 plays a functional role on HSC, we cloned the cDNA of mouse Pus10 into a lentiviral vector,³ and it exhibited efficient overexpression of PUS10 (Online Supplementary Figure S1A, B). Freshly isolated WT LSK cells were infected by PUS10-carrying lentivirus; 72 hours later, 2×10⁴ green fluorescent protein-positive (GFP⁺) cells were purified and transplanted into lethally irradiated recipients together with 2×10⁵ competitor cells (Figure 1D). Chimera in peripheral blood was evaluated every 4 weeks until the 12^th week by using this gating strategy (Online Supplementary Figure S1C, D). The results showed that enforced PUS10 severely impairs the reconstitution capacity of HSPC (Figure 1E). Moreover, enforced PUS10 promotes HSPC differentiation bias towards the myeloid lineage (Figure 1F), which is a classical phenomenon of aged HSC. Consistently, enforced PUS10 significantly inhibits HSPC expansion in vitro (Figure 1G, H).

Figure 1.Aging-activated PUS10 impairs the reconstitution capacity of hematopoietic stem and progenitor cells independently on its enzymatic activity. (A) Representative western blot (WB) showing the expression of PUS10 in hematopoietic stem and progenitor cells (HSPC) between young (3-month-old) and aged (28-month-old) mice. (B) This histogram depicts the protein level of PUS10 in young and aged HSPC from quantitative WB data (N=2). (C) The protein expression of PUS10 in the proteomics dataset of young and old hematopoietic stem cells (HSC). (D) Experimental design of the transplantation assay. (E) These line plots depict the changes in peripheral blood (PB) chimerism in recipients transplanted with vector or PUS10-OE LSK (N=7 mice per group). (F) This histogram displays the lineage distribution of myeloid, T and B cells among donor-derived cells in the PB of vector and PUS10-OE recipients at the 12^th week (N=7 mice per group). (G, H) Freshly isolated wild-type (WT) LSK cells were infected by PUS10-carrying lentivirus for 72 hours, and 50 green fluorescent protein-positive (GFP⁺) HSPC (CD48^- LSK) were sorted into 96-well plate and cultured in SFEM medium for 7 days. Then, the clones from vector or PUS10-OE HSPC were photographed and the cell numbers of these clones were analyzed. (G) These images show the expansion of vector or PUS10-OE HSPC.(H) The scatter plots show the cell numbers of these clones. (I) The result of liquid chromatography-tandem mass spectrometry shows the Ψ/U ratio in WT and Pus10^-/- LK cells with overexpression of WT PUS10 or PUS10^D342A. (J) These line plots display the changes in PB chimerism in recipients transplanted with vector, PUS10-OE or PUS10^D342A-OE LSK (N=7-8 mice per group). All data are shown as mean ± standard deviation; *P<0.05, **P<0.01, ***P<0.001.

In order to investigate whether the inhibitory function of enforced PUS10 on HSPC depends on its Ψ catalytic activity, we mutated the key enzyme site of PUS10 to generate catalytic dead PUS10^D342A according to a previous study.¹⁴ We first measured the Ψ/U ratio in WT and Pus10^-/- lineage^-cKi^t+ (LK) cells using liquid chromatography-tandem mass spectrometry (LC-MS/MS). The result showed that targeted dysfunction of Pus10 leads to a significant decrease of the Ψ/U ratio. In order to test whether PUS10^D342A is an inactive Ψ synthase, we reintroduced WT PUS10 and PUS10^D342A into Pus10^-/- LK cells, and measured the Ψ/U ratio for them. The result showed that the decrease of the Ψ/U ratio upon Pus10 deletion is rescued by the reintroduction of WT PUS10, but not PUS10^D342A (Figure 1I). This result indicates that the D342 residue is the key enzyme site for its Ψ synthase activity.

In order to investigate whether the Ψ synthase activity of PUS10 modulates HSC aging, freshly isolated WT LSK cells were infected with either PUS10 or PUS10^D342A-over-expressing lentivirus for 72 hours, and 2×10⁴ GFP⁺ cells were transplanted into lethally irradiated recipients together with 2×10⁵ competitor cells. The chimera in peripheral blood was evaluated every 4 weeks until the 12^th week. The results revealed that both enforced PUS10 and PUS10^D342A significantly impair the reconstitution capacity of HSPC (Figure 1J), indicating that the destructive role of PUS10 on HSPC is independent on its enzymatic activity.

No difference of pseudouridylation profile between young and aged hematopoietic stem and progenitor cells

We then set out to investigate whether aged HSC exhibit aberrant Ψ profile compared to young ones. Due to the limited cell number of HSC and the large amount of cells required for Ψ sequencing, we performed DM-Ψ-seq by using freshly isolated lineage^- cells, which are hematopoietic stem and progenitor-enriched cells, from 3-month-old and 29-month-old mice according to an elegant approach^14,36 (Figure 2A). The result revealed no difference of the bulk Ψ profile between young and aged HSPC (Figure 2B). We then wondered whether the percentage of Ψ on certain sites of the tRNA exhibits a difference between them. In order to address this question, we evaluated all of the Ψ on detectable tRNA. The results revealed that the percentage of Ψ²⁸, Ψ³², Ψ⁵³ and Ψ⁵⁴ of tRNA^His-GTG is static in young and aged HSPC (Figure 2C, D), and identical to other detected tRNA (Online Supplementary Figure S2). Briefly, these results indicate that not only the bulk Ψ profile but also the percentage of each Ψ on tRNA exhibits no difference between young and aged HSPC, which is consistent with the data that the toxicity of enforced PUS10 on HSPC is independent on its enzymatic activity.

Aging-declined CRL4DCAF1-mediated ubiquitination degradation signaling leads to the increase of PUS10

Given that PUS10 is increased in aged HSPC and enforced PUS10 impairs the reconstitution capacity of HSPC, we then wondered how PUS10 is increased with aging. Firstly, we examined the mRNA level of Pus10 in young and aged HSPC (CD48^-LSK) by reverse transcription PCR (RT-PCR), and the result revealed no difference of Pus10 mRNA between them (Figure 3A). In order to confirm this observation, we examined the mRNA level of Pus10 between young and aged HSC by exploring published RNA-sequencing data, and the result revealed that the mRNA level of Pus10 remains static in young and aged HSC (Online Supplementary Figure S3). Then, it is conceivable that the increase of PUS10 might be modulated via post-transcriptional modification manner. In order to test this hypothesis, we purified proteins interacting with PUS10 via affinity purification and we observed that the CRL4^DCAF1 complex, including DDB1, DCAF1 and CUL4B, interacts with PUS10 (Figure 3B). CRL4^DCAF1 complex is E3 ubiquitin ligase targeting substrate for protein degradation.⁴¹ In order to confirm this observation, we performed co-immunoprecipitation (Co-IP) assay by infecting HEK293T cells with S-protein, FLAG, and streptavidin-binding peptide (SFB)-tagged PUS10 together with Myc-tagged DDB1, DCAF1 or CUL4B respectively. Cell lysates were incubated with S-protein beads and probed with anti-Flag, anti-Myc antibodies. The result showed that PUS10 exhibits strong interaction with DDB1, DCAF1 and CUL4B (Figure 3C).

Figure 2.No difference of pseudouridine modification profile between young and aged hematopoietic stem and progenitor cells. (A-D) Lineage^- cells were isolated from young (3-month-old) and aged (29-month-old) mice. Small RNA (<200 nt) was extracted and purified to perform demethylase-pseudouridine sequencing (DM–Ψ-seq). (A) Experimental design. (B) The scatter plot depicts Ψ levels in transcribed RNA (tRNA) between young and aged hematopoietic stem and progenitor cells (HSPC) (N=2). (C) Schematic of tRNA-His-GTG-2-1 shows Ψ sites Ψ²⁸, Ψ³², Ψ⁵³ and Ψ⁵⁴. (D) Ψ sites (red arrows) and levels of tRNA-His-GTG-2-1 are identified in young and aged HSPC. The x axis represents nucleotide position. The y axis represents Ψ levels.

Figure 3.Aging-declined CRL4^DCAF1-mediated ubiquitination degradation signaling leads to the increase of PUS10. (A) This histogram depicts the mRNA expression of Pus10 in young (2 months) and aged (31 months) hematopoietic stem and progenitor cells (HSPC). (B) Affinity purification of PUS10 protein from HEK293T cells stably expressing Flag-tagged PUS10. Proteins identified by mass spectrometry are listed. The bait protein is marked in bold letters. (C) HEK293T cells were co-transfected with plasmids encoding SFB-tagged PUS10 and Myc-tagged DDB1, DCAF1, CUL4B followed by co-immunoprecipitation using anti-Flag, anti-Myc antibody. Representative western blot (WB) shows that PUS10 interacts with DDB1, DCAF1 and CUL4B. (D) HEK293T cells were co-transfected with plasmids encoding SFB-tagged PUS10, Myc-tagged DDB1, DCAF1, CUL4B and HA-tagged Ub-WT or Ub-K48R followed by coimmunoprecipitation using anti-HA, anti-Flag, anti-Myc antibody. Representative WB shows that PUS10 is ubiquitinated by the CRL4^DCAF1 complex. (E) Representative WB shows the expressions of DDB1 and CUL4B in young (3 months) and aged (28 months) HSPC. (F) 32D cells were infected by lentivirus carrying Ddb1-shRNA. 72 hours later, green fluorescent protein-positive (GFP⁺) cells were sorted for WB to validate the expression of PUS10 and DDB1. Representative WB shows the expression of PUS10 and DDB1 in vector and DDB1-KD 32D cells. (G) This histogram depicts the protein level of PUS10 in vector and DDB1-KD 32D cells from quantitative WB data (N=2). (H) This histogram depicts the protein expression of DDB1 in vector and DDB1-KD 32D cells from quantitative WB data (N=3).

Previous study has shown that CRL4^DCAF1 complex participates in ubiquitin dependent degradation,⁴¹ we then set out to determine whether CRL4^DCAF1 regulates PUS10 ubiquitination. Plasmids encoding SFB-tagged PUS10, Myctagged DDB1, DCAF1, CUL4B and HA-tagged wild-type ubiquitin (Ub-WT), mutant ubiquitin (Ub-K48R) were cotransfected into HEK293T cells; 24 hours later, cell lysates were collected to detect the ubiquitination of PUS10. The result revealed that CRL4^DCAF1 vigorously promotes the ubiquitination of PUS10 in cells expressing WT ubiquitin (Figure 3D). Compared with WT ubiquitin, the ubiquitination of PUS10 was completely abolished in cells expressing K48R ubiquitin, indicating that CRL4^DCAF1 promotes polyubiquitination of PUS10 via the formation of the K48 linkage.

Next, we set out to investigate whether the increase of PUS10 is due to the alteration of CRL4^DCAF1 in aged HSPC. We first evaluated the expression of CRL4^DCAF1 complex in young and aged HSPC. The result showed that the expression of DDB1 and CUL4B is decreased in aged HSPC (the expression level of DCAF1 is too low to be detected) (Figure 3E), which is negatively correlated with the alteration of PUS10 between young and aged HSPC (Figure 1A). Given that the protein level of PUS10, but not mRNA level, is elevated in aged HSPC (Figures 1A and 3A), and that PUS10 is poly-ubiquitinated by CRL4^DCAF1 complex (Figure 3D), and that DDB1 and CUL4B are decreased in aged HSPC (Figure 3E), we then wondered whether aging-declined CRL4^DCAF1 complex leads to the increase of PUS10. In order to test this hypothesis, we generated two efficient small hairpin RNA (shRNA) against Ddb1 (Figure 3F, H; Online Supplementary Table S2), which is the key linker protein of CRL4^DCAF1 complex.⁴¹ 32D cells were infected by shDdb1 carrying lentivirus for 72 hours, and GFP⁺ cells were subjected to detect the protein level of PUS10 by western blot. The result showed that PUS10 is elevated upon the knockdown of Ddb1 (Figure 3F, G).

Taken together, these data suggest that aging-declined CRL4^DCAF1-mediated ubiquitination degradation signaling leads to the accumulation of PUS10.

Young Pus10-/- mice exhibit no influence on hematopoietic homeostasis and hematopoietic stem cell function

The above-mentioned results revealed the functional role of Pus10 in modulating HSC aging, we then wondered whether targeted dysfunction of Pus10 plays a role in regulating hematopoietic homeostasis and HSC function. In order to address this question, we generated Pus10 knockout (Pus10^-/-) mice on C57BL/6N background by deleting the 2^nd exon using CRISPR-Cas9 system (Figure 4A, see details in the Material and Methods) and we achieved efficient deletion of PUS10 in LSK cells (Figure 4B, C).

We then performed complete blood count assay for Pus10^-/- and age-matched control mice. The result revealed no difference of white blood cell (WBC), lymphocyte (LYM), neutrophil (NEUT), red blood cell (RBC) and platelet (PLT) count between Pus10^-/- and WT mice (Figure 4D). We then sought to investigate the lineage composition in peripheral blood (PB) and bone marrow (BM) of Pus10^-/- mice, including T cells, B cells and myeloid cells (Online Supplementary Figure S4A). The results revealed no difference of Pus10^-/- mice compared to WT in PB (Figure 4E) and BM (Figure 4F).

We next analyzed HSPC of Pus10^-/- mice, including common myeloid progenitors (CMP), granulocyte-macrophage progenitors (GMP), megakaryocyte-erythroid progenitors (MEP), common lymphoid progenitors (CLP), multipotent progenitor cells (MPP) and HSC (Online Supplementary Figure S4B). The results revealed no significant difference of the above-mentioned populations between Pus10^-/- and WT mice (Figure 4G-K).

In order to further investigate the reconstitution capacity of Pus10^-/- HSC, 20 freshly isolated Pus10^-/- and WT HSC were transplanted into lethally irradiated recipients together with 3×10⁵ competitor cells (Figure 5A). The chimera in the PB of recipients was evaluated every 4 weeks until the 16^th week (Online Supplementary Figure S4C, D). The results showed that the reconstitution capacity of Pus10^-/- HSC is comparable with WT ones (Figure 5B), while dysfunction of Pus10 promotes the differentiation bias towards lymphoid lineage (34.31% vs. 48.43%; Figure 5C). Donor-derived HSC of recipients revealed no significant difference between Pus10^-/- and WT mice (Figure 5D-F).

Aged Pus10-/- mice exhibit mild alteration of hematopoietic homeostasis and hematopoietic stem cell function

We then investigated the phenomenon of aged Pus10^-/-mice. We performed complete blood count assay for aged WT and Pus10^-/- mice (26-month-old). The result showed no significant difference of WBC, LYM, NEUT, RBC and PLT between aged WT and Pus10^-/- mice (Figure 6A). We then analyzed the frequency of T cells, B cells and myeloid cells in PB, BM, spleen and thymus of aged Pus10^-/- and WT mice. The results revealed no significant difference between them in the PB (Figure 6B) and thymus (Figure 6E). However, the percentage of T cells in the BM (Figure 6C) and the percentage of myeloid cells in the spleen (Figure 6D) of aged Pus10^-/- mice are increased.

Figure 4.Young Pus10^-/- mice exhibit no influence on hematopoietic homeostasis.(A) Schematic illustration of the Pus10 knockout (Pus10^-/-) mice. (B) Representative western blot (WB) shows the expression of PUS10 in wild-type (WT) and Pus10^-/- LSK cells. (C) This histogram depicts the protein expression of PUS10 in WT and Pus10^-/- LSK cells from quantitative WB data (N=2). (D) The scatter plots show the numbers of white blood cells (WBC), lymphocytes (LYM), neutrophils (NEUT), red blood cells (RBC) and platelets (PLT) between WT and Pus10^-/- mice. (E, F) The scatter plots depict the frequency of B cells, T cells and myeloid cells in peripheral blood (PB) (E) and bone marrow (BM) (F) of WT and Pus10^-/- mice. (G) The scatter plots show the cell numbers of BM in WT and Pus10^-/- femurs. (H-K) The scatter plots depict the cell numbers and frequency of common myeloid progenitors (CMP), granulocyte/macrophage progenitors (GMP), megakaryocytic/erythroid progenitors (MEP), common lymphoid progenitors (CLP), multipotent progenitor cells (MPP) and HSC in WT and Pus10^-/- femurs. All data are shown as mean ± standard deviation; *P<0.05, **P<0.01, ***P<0.001.

Figure 5.Young Pus10^-/- mice exhibit no influence on hematopoietic stem cell function. (A) Experimental design of the competitive transplantation strategy. (B) These line plots depict the changes in peripheral blood (PB) chimerism in recipients transplanted with wild-type (WT) or Pus10 knockout (Pus10^-/-) hematopoietic stem cells (HSC) (N=5 mice per group). (C) This histogram displays the lineage distribution of myeloid, T and B cells among donor-derived cells in PB of the recipients at the 16^th week (N=5 mice per group). (D) The gating strategies for the frequency of the test donor-derived HSC. (E, F) The scatter plots depict the cell numbers (E) and frequency (F) of donor-derived HSC in recipients transplanted with WT or Pus10^-/- HSC (N=5 mice per group). All data are shown as mean ± standard deviation; *P<0.05, **P<0.01, ***P<0.001.

Figure 6.Aged Pus10^-/- mice exhibit mild alteration of hematopoietic homeostasis and hematopoietic stem cell function. (A) The scatter plots show the numbers of white blood cells (WBC), lymphocytes (LYM), neutrophils (NEUT), red blood cells (RBC) and platelets (PLT) between aged wild-type (WT) and Pus10 knockout (Pus10^-/-) mice. (B-E) The scatter plots depict the frequency of B cells, T cells and myeloid cells in the peripheral blood (PB) (B), bone marrow (C), spleen (D) and thymus (E) of aged WT and Pus10^-/- mice. (F) The scatter plots show the cell numbers of bone marrow in WT and Pus10^-/- femurs. (G-J) The scatter plots indicate the cell numbers and frequency of common myeloid progenitors (CMP), granulocyte/macrophage progenitors (GMP), megakaryocytic/erythroid progenitors (MEP), common lymphoid progenitors (CLP), multipotent progenitor cells (MPP) and hematopoietic stem cells (HSC) in WT and Pus10^-/- femurs. (K) Experimental design of the transplantation assay. (L) These line plots depict the changes in PB chimerism in recipients transplanted with aged WT or Pus10^-/- HSC (N=8-10 mice per group). (M) This histogram displays the lineage distribution of myeloid, T and B cells among donor-derived cells in the PB of the recipients at the 12^th week (N=8-10 mice per group). All data are shown as mean ± standard deviation; *P<0.05, **P<0.01, ***P<0.001.

We next investigated HSPC of aged Pus10^-/- and WT mice. The results indicated no significant difference of the CMP, GMP, MEP, CLP and MPP, while the frequency of HSC is increased in aged Pus10^-/- mice compared to WT (Figure 6F-J). In order to further explore the reconstitution capacity of aged Pus10^-/- HSC, 150 freshly isolated aged Pus10^-/- and WT HSC were transplanted into lethally irradiated recipients together with 3×10⁵ competitor cells (Figure 6K). The chimera in the PB of recipients was evaluated every 4 weeks until the 12^th week. The results revealed that the reconstitution capacity of aged Pus10^-/- HSC is comparable with WT ones, while dysfunction of Pus10 promotes the differentiation bias towards B-cell lineage (Figure 6L, M). In brief, our study for the first time revealed that enforced PUS10 impairs the reconstitution capacity of HSPC. Howerer, the hematopoietic homeostasis and reconstitution capacity of young Pus10^-/- mice is comparable with control mice, while aged Pus10^-/- mice exhibit mild alteration of hematopoietic homeostasis and HSC function. In summary, these data suggest that aging-diminished CRL4^DCAF1-mediated ubiquitination degradation signaling leads to the accumulation of PUS10, which impairs HSPC (Figure 7).

Discussion

Our study for the first time evaluated the role of PUS10 in HSC aging and function. This study will expand our understanding of RNA modification on HSC function regulation.

Post-transcriptional regulation in aged hematopoietic stem cells

An elegant study reported a proteomics resource from mass spectrometry of mouse young and aged HSC, and identified a subset of genes with apparent post-transcriptional alteration during aging.⁴⁰ This indicates that transcriptomic levels may not reflect the functional change of aged HSC. The alteration of the protein level achieved either by RNA or protein modification in aged HSC might play an essential role in promoting HSC aging. Our unpublished data confirmed this observation by identifying a group of RNA modification genes which modulate HSC aging. In addition, our study also identified that CRL4^DCAF1-mediated ubiquitination participated in regulating HSC aging by degrading PUS10 and other important proteins (unpublished). Whether there are other proteins modified by ubiquitination which lead to their changes during aging, thereby regulating HSC aging and other cell aging is a question worthy of study. It is also intriguing to investigate the molecular mechanism by which ubiquitin ligases are altered during aging.

Therefore, exploring the functional role of post-transcriptional modification (PTM) in aging might strengthen the understanding of aging on HSC and its clinical relevance.

Figure 7.Aging-enforced PUS10 mediated by the decrease of ubiquitination degradation impairs hematopoietic stem and progenitor cells. The proposed model diagram illustrates the accumulation of PUS10 mediated by the decrease of ubiquitination degradation during aging impairs hematopoietic stem and progenitor cells (HSPC).

RNA epigenetics versus hematopoietic stem cell aging

In our study, we observed the Ψ profile is not changed in aged HSPC and the aging-increased PUS10 promotes HSC aging, which is not achieved by its Ψ synthase activity. Up to date, more than 170 RNA modifications have been identified and some of them play an essential role in various biological process and clinical diseases.^42,43 Whether other RNA modification profiles are altered in aged HSC and whether the corresponding enzymes are involved in regulating HSC aging is a question worthy of investigation. In this study, we performed Ψ profiling by using HSPC, but not pure HSC, due to the limitation of HSC number. Whether the Ψ profile of aged HSC is identical as we observed in HSPC is also a question worth investigating, which depends on the development of sequencing technology.

Footnotes

• Received October 4, 2022
• Accepted May 4, 2023

Correspondence

*YW and ZZ contributed equally as first authors

M.Q. Zhang michael.zhang@utdallas.edu

M. Shi shiml79@tsinghua.edu.cn

C. Yi chengqi.yi@pku.edu.cn

J. Wang jianweiwang@mail.tsinghua.edu.cn; wangjianwei@ihcams.ac.cn

Disclosures

No conflicts of interest to disclose.

Contributions

JW and CY developed the concept. JW and CY developed the methodology. YW, ZZ, HH, JS, YC, YC, YZ and XZ carried out the investigation. JW, CY, MS, MQZ, XZ and ML performed the formal analysis. JW, CY, MS and MQZ provided resources. JW and CY wrote the manuscript. JW and CY acquired funding. JW, CY, MS and MQZ supervised the study.

Data-sharing statement

The data are available on request. All sequencing raw data were deposited in the National Center for Biotechnology Information Gene Expression Omnibus. The accession code is GSE213422 with the enter token atstusiqpxmhvmj.

Funding

This work was supported by grant numbers Z200022, 82250002, 81870118 and 2018YFA0800200 to JWW, and 2019YFA0110900, 2019YFA0802200 to CQY, and 81890991 to MS from the National Key R&D Program of China or the Beijing Municipal Science & Technology Commission and the National Natural Science Foundation of China.

Acknowledgments

We thank the Tsinghua-Peking Center for Life Sciences for facility and financial support.