Transient Inhibition of the JNK Pathway Promotes Human Hematopoietic Stem Cell Quiescence and Engraftment

Abstract The widespread clinical application of cord blood (CB) for hematopoietic stem cell (HSC) transplantation is limited mainly by the inadequate number of hematopoietic stem and progenitor cells (HSPCs) in single CB units, which results in unsuccessful or delayed engraftment in recipients. The identification of agents to promote CB HSPC engraftment has significant therapeutic value. Here, we found that transient inhibition of the JNK pathway increased the HSC frequency in CB CD34+ cells to 13.46-fold. Mechanistic studies showed that inhibition of the JNK pathway upregulated the expression of quiescence-associated and stemness genes in HSCs, preventing HSCs from entering the cell cycle, increasing glucose uptake and accumulating reactive oxygen species (ROS). Importantly, transient inhibition of the JNK pathway during CB CD34+ cell collection also enhanced long-term HSC (LT-HSC) recovery and engraftment efficiency. Collectively, these findings suggest that transient inhibition of the JNK pathway could promote a quiescent state in HSCs by preventing cell cycle entry and metabolic activation, thus enhancing the HSC number and engraftment potential. Together, these findings improve the understanding of the regulatory mechanisms governing HSC quiescence and stemness and have the potential to improve HSC collection and transplantation.


Introduction
The ability of HSCs to undergo self-renewal and differentiate into almost all blood cell lineages makes HSC transplantation the most commonly used cell therapy to treat malignant and non-malignant hematological disorders. 1 CB, which has the advantages of a less strict HLA-matching demand, low incidence of graft-versus-host-disease, and low disease relapse rate, is an attractive donor source for HSC transplantation. 2 Current estimates indicate that there have been more than 40 000 recipients of CB HSC transplantation worldwide 3 ; however, the insufficient number of HSCs in single CB units greatly limits the clinical application of CB HSCs. 4 Multiple efforts have been made to improve the engraftment efficacy of CB HSCs. 4,5 These approaches primarily include (1) increasing the HSC number by in vitro expansion with cytokines, small molecules 6,7 or artificial niches [8][9][10] and (2) boosting HSC homing with reagents that can enhance HSC homing receptor expression [11][12][13] or modify HSC niches in the recipient. 14 Although several methods 6,8,9 have significantly improved CB HSC engraftment efficacy, alternative strategies are needed to maximize the functional HSC harvest from single CB units.
The most primitive HSCs with long-term repopulation ability reside in a hypoxic environment and exhibit a quiescent status in their in vivo niche. 15,16 They barely divide and exhibit rather low metabolic activity with little ROS. 16 When exposed to cytokines or other ambient stimuli, quiescent HSCs can be activated and enter the cell cycle; however, they always lose their self-renewal ability. 17 The JNK pathway is heavily involved in the regulation of cell proliferation and metabolic processes. 18 Therefore, we hypothesized that manipulation of the JNK pathway could improve human HSC engraftment by regulating the transition of the cell cycle and metabolic status.

Mice and CB
All of the mice used for transplantation were NOD-Prkdcscid Il2rgtm1/Vst (NPG) mice (Stock No.: VS-AM-001) purchased from Beijing Vitalstar Biotechnology and ranged from 8

Transplantation and Engraftment Analysis
Cells were transplanted into mouse recipients via the tail vein. Repopulated human cells in NPG mouse peripheral blood (PB), bone marrow (BM), and spleen were monitored by flow cytometry at the indicated time points post-transplantation. Details are provided in the Supplementary Methods section of the Supplementary material.

Flow Cytometry Analysis
Fresh uncultured cells or cultured cells collected at 24 h were incubated with the indicated antibodies for 30 minutes at 4 °C in PBS containing 0.5% BSA (Sigma-Aldrich, Cat: A1470-100G). Then, the cells were washed with PBS and suspended in 0.2 mL PBS for analysis. Flow cytometry analysis was performed using a FACSCantoTM II (BD). Data were analyzed using FlowJo v10 (BD). Details are provided in the Supplementary Methods section of the Supplementary material.

Limiting Dilution Analysis
The HSC frequency was quantified by extreme limiting dilution analysis (http://bioinf.wehi.edu.au/software/elda/) with 95% CIs. The engraftment of more than 0.1% of human CD45 + cells in the BM was regarded as a positive response.

Quantification and Statistical Analysis
Statistical analysis was performed with GraphPad Prism software. Data are shown as the mean and SD. Pairwise comparisons between different groups were assessed using an unpaired t-test. For all analyses, P < .05 was considered statistically significant. The statistical significance and n value are reported in the figure legends. All the flow analysis data were processed with FlowJo v10 software. All the figures were prepared with Adobe Illustrator.

Results and Discussion
Transient Inhibition of the JNK Pathway was Identified to Increase the HSC Number in CB We screened a panel of JNK pathway-related small molecules to increase the CB Lin − CD34 + CD45RA − cell population, which is enriched for HSCs 19 (Fig. 1A). The results showed that the JNK inhibitors AEG3482, SP600125, and JNK-IN-8, could each significantly increase the frequency of Lin − CD34 + CD45RA − cells after 24 h of incubation compared with control treatment (cytokine-only or DMSO-treated cells) (Fig. 1B). The highest frequency of HSCs was obtained with 15 μM JNK-IN-8 ( Supplementary Fig. S1A), which significantly downregulated the expression of the target c-Jun ( Supplementary Fig. S1B), and there were significant increases in the frequency and number of phenotypic HSC subsets (Lin − CD34 + CD38 − and Lin − CD34 + CD38 − CD45RA − cells) in the JNK-IN-8-treated group compared with the uncultured and DMSO-treated groups ( Fig. 1C−1F). Furthermore, after transplantation, the JNK-IN-8-treated group exhibited better engraftment efficiency than the uncultured and DMSO-treated groups ( Fig. 1G; Supplementary Table S1). And a limiting dilution assay (LDA) showed that the JNK-IN-8-treated group possessed a higher transplantable HSC frequency, with 13.46-fold and 14.37-fold increases compared with the uncultured and DMSO-treated groups, respectively ( Fig. 1H; Supplementary Fig. S1C, S1D); the HSC number was approximately 800 for every 1 × 10 5 CD34 + cells in the JNK-IN-8-treated group, which was higher than the numbers in the uncultured and DMSO-treated groups (60 and 56, respectively) (Fig. 1I). Together, these results suggest that transient inhibition of the JNK pathway significantly enhances phenotypic and transplantable HSC numbers in CB CD34 + cells.
We next investigated whether transient inhibition of the JNK pathway during CD34 + cell isolation can increase LT-HSC harvest from single CB units by the LDA, in which 2000/500/200 conventionally isolated CD34 + cells (control group) and the same doses of JNK-IN-8-treated CD34 + cells were injected into separate immunodeficient NPG mice. We found that the JNK-IN-8-treated group exhibited better engraftment efficiency than the control group ( Fig Table S1. All data are shown as the mean value ± SD. Statistical significance was assessed using one-way ANOVA if not mentioned. ns, not significant; *P <.05; **P < .01; ***P < .001.  (Fig. 2G, 2H), and CD3 + T cells could be detected in the thymus (Fig. 2I).  Table S2C. All data are shown as the mean value ± SD. Statistical significance was assessed using one-way ANOVA if not mentioned. ns, not significant; *P < .05; **P < .01; ***P < .001; ****P < .0001.   better engraftment than the control group (Fig. 2J). The LDA showed that the 2° SCID-repopulating cells (SRCs) frequency of the JNK-IN-8-treated group was 14.92-fold higher than that of the control group (approximately 92 in the JNK-IN-8treated group versus 6 in the control group for every 1 × 10 7 primary recipient BM cells) (Fig. 2K, 2L  Moreover, gene set enrichment analysis (GSEA) showed that most of the previously reported 120 HSC-specific genes 21 were upregulated in the JNK-IN-8-treated group (Fig. 3D).
We next explored cell cycle-and metabolism-related gene expression. We found that DMSO-treated cells had highly upregulated cell cycle entry genes and active genes involved in glycolysis and oxidative phosphorylation compared with uncultured cells, while the expression levels of these genes were maintained or reduced in JNK-IN-8-treated cells (Fig.  3A, 3B). Notably, JNK-IN-8 treatment upregulated cell cycle inhibitor (CDKN1B, CDKN1C, and CCNG2) expression ( Fig. 3A; Supplementary Fig. S3E). Likewise, GSEA revealed that the expression of E2F and MYC targets was upregulated in the DMSO-treated group but downregulated in the JNK-IN-8-treated group ( Supplementary Fig. S3A, S3B). Similar results were also observed for glycolysis and oxidative phosphorylation gene sets ( Supplementary Fig. S3C-S3E).

Transient Inhibition of the JNK Pathway Prevented HSCs From Undergoing Cell Cycle Entry and Metabolic Activation
We further conducted cell cycle status analysis, glucose uptake activity measurement, and ROS level detection with DMSO-treated, uncultured, and JNK-IN-8-treated CB CD34 + cells. The DMSO-treated group showed a much higher percentage of cells in the S phase, while the JNK-IN-8-treated and uncultured groups consisted mostly of G0/G1 cells (Fig.  3E). DMSO-treated cells showed the highest glucose uptake activity and intracellular ROS level, while JNK-IN-8-treated and uncultured cells maintained similarly low levels ( Fig.  3F-3I). These results, together with the gene expression profile analysis (Fig. 3A-3D; Supplementary Fig. S3), indicate that JNK-IN-8 treatment promotes the quiescence status of HSCs by preventing HSCs from undergoing cell cycle entry and metabolic activation (Fig. 3J).
We have previously reported a supportive role for JNK inhibition in HSC self-renewal in vitro; however, the engraftment process was slow, and the detailed mechanism was largely unclear. 25 Here, by increasing the concentration of JNK inhibitors (15 μM) and shortening the incubation period (24 h), we obtained a greater increase in the HSC number (13.46-fold). We found, for the first time, a novel role for JNK pathway inhibition in the promotion of HSC quiescence and stemness.
In our study, transient JNK inhibition modestly increased the phenotypic HSC (Lin − CD34 + CD38 − CD45RA − ) number in CB CD34 + cells, but the transplantable HSC number was dramatically increased. There is no immunophenotype that can identify functional HSCs with 100% purity; therefore, engraftment efficiency is the gold standard to evaluate functional HSCs. In our context, CD34 + cells were a remarkably heterogeneous population, making the exact identity of the target cells affected by JNK inhibition unclear. Considering that the majority of cells were in the G0/G1 phase, we postulated that this notable increase in functional HSCs probably occurred through rapid enhancement of latent HSC activity in CB HSPCs rather than expansion. Transient JNK inhibition markedly enhanced the expression of key HSC-specific stemness genes, including the surface markers TEK, PROCR, and ITGA3 and the transcription factors MECOM, HOXA5, HLF, MYB, HOXB5, ETV6, MLLT3, MSI2, and HOPX; HSCs exhibited low metabolic activity and reduced ROS levels under JNK inhibition. Each of the characteristics above has been reported in HSCs showing the greatest engraftment potential. Therefore, we hypothesized that transient JNK inhibition enhances CB HSC engraftment ability by recruiting cells in the CD34 + compartment that would not otherwise engraft via upregulation of stemness genes and promotion of quiescence.

Conclusion
In summary, we observed a high increase in the HSC number (13.46-fold) in CB CD34 + cells after incubation with JNK-IN-8 (15 µM) for 24 h. We also found a novel role for JNK pathway inhibition in the promotion of HSC quiescence and stemness. More interestingly, we found that supplementation with JNK inhibitors during CD34 + cell collection could significantly increase the LT-HSC number in CB units, which could help maximize functional HSC harvest from CB in clinical settings.
Our study extends the understanding of HSC quiescence and stemness regulation. Further investigation of whether transient inhibition of the JNK pathway can complement previous endeavors to maximize functional HSC harvest from single CB units using a combinational approach is needed. through its roles in quiescence and stemness. All data are shown as the mean value ± SD. Statistical significance was assessed using one-way ANOVA if not mentioned. ns, not significant; **P <.01; ***P < .001; ****P < .0001.