HSPA9/mortalin inhibition disrupts erythroid maturation through a TP53-dependent mechanism in human CD34+ hematopoietic progenitor cells

Myelodysplastic syndromes (MDS) are a heterogeneous group of clonal hematopoietic stem cell malignancies characterized by abnormal hematopoietic cell maturation, increased apoptosis of bone marrow cells, and anemia. They are the most common myeloid blood cancers in American adults. The full complement of gene mutations that contribute to the phenotypes or clinical symptoms in MDS is not fully understood. Around 10%–25% of MDS patients harbor an interstitial heterozygous deletion on the long arm of chromosome 5 [del(5q)], creating haploinsufficiency for a large set of genes, including HSPA9. The HSPA9 gene encodes for the protein mortalin, a highly conserved heat shock protein predominantly localized in mitochondria. Our prior study showed that knockdown of HSPA9 induces TP53-dependent apoptosis in human CD34+ hematopoietic progenitor cells. In this study, we explored the role of HSPA9 in regulating erythroid maturation using human CD34+ cells. We inhibited the expression of HSPA9 using gene knockdown and pharmacological inhibition and found that inhibition of HSPA9 disrupted erythroid maturation as well as increased expression of p53 in CD34+ cells. To test whether the molecular mechanism of HSPA9 regulating erythroid maturation is TP53-dependent, we knocked down HSPA9 and TP53 individually or in combination in human CD34+ cells. We found that the knockdown of TP53 partially rescued the erythroid maturation defect induced by HSPA9 knockdown, suggesting that the defect in cells with reduced HSPA9 expression is TP53-dependent. Collectively, these findings indicate that reduced levels of HSPA9 may contribute to the anemia observed in del(5q)-associated MDS patients due to the activation of TP53.


Introduction
Heat shock proteins (HSPs) constitute a group of proteins involved in assisting the folding and maturation of other proteins, and their expression is normally induced by heat shock or other stressors.Traditionally, HSPs are known as molecular chaperones due to their physiological and protective roles in cells.They facilitate protein folding and maintenance of natural structures and functions of other proteins when cells are exposed to homeostatic challenges such as extreme temperature, anoxia, hypoxia, heavy metals, drugs, or other chemical agents that may induce stress or protein denaturation. 1he human HSPs are generally classified according to their molecular weights with the majority belonging to the groups HSP27, HSP40, HSP60, HSP70, HSP90, and large HSPs (HSP110 and GRP170). 2 The human HSP70 family consists of 13 members encoded by the HSPA genes.HSP70 proteins have a highly conserved domain structure, including the ∼44 kDa N-terminal ATPase domain, ∼18 kDa substrate-binding domain, and ∼10 kDa C-terminal domain.2,3 The HSPA9 gene, encoding the protein mortalin, is located on human chromosome 5q31.The role of mortalin has been implicated in various cancer types. Fr example, overexpression of mortalin is detected in breast and liver cancers, and is associated with cell migration, invasiveness, epithelial-mesenchymal transition, and metastasis.[4][5][6] Elevated levels of mortalin are clinically associated with poor prognosis and survival in patients with gastric and colorectal cancers.7,8 In human colorectal adenocarcinoma cells, mortalin binds to and sequesters p53 (encoded by the TP53 gene) in the cytoplasm, thereby preventing the translocation of p53 into the nucleus, indicating its role in regulating cell cycle and apoptosis.9 Myelodysplastic syndromes (MDS) are a heterogeneous group of clonal hematopoietic stem cell malignancies characterized by abnormal hematopoietic cell maturation, increased apoptosis of bone marrow cells, and peripheral blood cytopenias.10 Patients with MDS present with symptoms of fatigue, shortness of breath, pallor, unusual bruising or bleeding, petechiae, and frequent infections.11,12 Even with intervention and treatment, a fraction of MDS patients will progress to secondary acute myeloid leukemia, characterized by a higher percentage of blasts in either the bone marrow or peripheral blood.6 Del(5q) is among the most common cytogenetic aberrations in MDS and defines a unique MDS subcategory, representing the first genomic alteration included in the World Health Organization classification of MDS. 17 Two distinct commonly deleted regions (CDR), distal and proximal CDR, have been identified on the del(5q) region in MDS patients.The HSPA9 gene is located in the proximal CDR which is associated with patients that have a higher risk of progressing from MDS to secondary acute myeloid leukemia compared to the distal CDR (Figure 1(a)).[18][19][20] In addition to its function as a molecular chaperone, the role of HSPA9/mortalin in hematopoiesis and erythroid maturation has been studied in nonhuman models.Zebrafish with a homozygous point mutation in Hspa9 present phenotypically with a variety of abnormalities including severe anemia, defects in erythroid differentiation, and elevated apoptosis.21 Tai-Nagara et al. 22 reported that inhibition of mortalin function increased reactive oxygen species and decreased the number of hematopoietic stem cells in mice, suggesting its essential role in maintaining hematological hemostasis by regulating oxidative stress.We reported that Hspa9 heterozygous deletion mice have normal basal hematopoiesis, but display altered B-cell lymphopoiesis.23 This is consistent with our later reports showing that haploinsufficiency of multiple del (5q) genes, including Hspa9, also induces B-cell abnormalities in mice.24 To further examine the relationship between HSPA9 and erythroid maturation in humans, we used human CD34+ hematopoietic progenitor cells as our experimental model in this study.We inhibited the expression of HSPA9 in these cells using gene knockdown or an allosteric inhibitor and assessed erythroid maturation.25,26 We have previously identified that HSPA9 knockdown induces apoptosis in human CD34+ cells, which is likely a TP53-dependent process, suggesting that reduced levels of HSPA9 may contribute to TP53 activation and increased apoptosis observed in del(5q)associated MDS.27 In this study, we measured the expression of TP53 after HSPA9 inhibition, in order to elucidate whether the mechanism of altered erythroid cell maturation induced by reduced HSPA9 expression is also TP53-dependent.

Knockdown of HSPA9 by siRNA increased p53 expression in human CD34+ hematopoietic progenitor cells
To study the role of HSPA9 in regulating erythroid maturation, we knocked down the expression of HSPA9 in human CD34+ hematopoietic progenitor cells grown in erythroid differentiation media to explore the mechanism of anemia observed in del(5q) MDS patients.
We first transfected small interfering RNA (siRNA) targeting HSPA9 into human CD34+ cells and studied the relationship between HSPA9 knockdown and p53 expression.We found that HSPA9 siRNA effectively inhibited the expression of its encoding protein mortalin to ∼50%, similar to levels occurring in heterozygous del (5q) deletions in patients, and increased the expression of p53 in these cells compared to control siRNA measured by western blotting (Figure 1(b)).Flow cytometry analysis also showed that HSPA9 knockdown resulted in increased p53 expression (Figure 1(c)) based on the mean fluorescent intensity (P < 0.01) (Figure 1(d)).

Knockdown of HSPA9 by siRNA inhibited cell growth, increased cell apoptosis, and inhibited erythroid maturation in human CD34 + hematopoietic progenitor cells
Del(5q) MDS patients present with anemia and cytopenias, and increased apoptosis in their erythroid cells.

Pharmacologic inhibition of HSPA9/mortalin increased p53 expression and inhibited erythroid maturation in human CD34 + hematopoietic progenitor cells
Next, we studied the effects of pharmacologic inhibition of HSPA9/mortalin on p53 expression and erythroid maturation in CD34+ cells using MKT-077, a small molecule inhibitor of HSP70 protein family members including HSPA9/mortalin. 28Human CD34+ cells were treated with two concentrations (0.5 and 2 μM) of MKT-077 in erythroid differentiation media for 5 days.Following 5 days of treatment, mortalin expression was reduced in a dose-dependent manner measured by western blotting (Figure 3(a)).Similar to HSPA9 siRNA, MKT-077 also increased the p53 expression levels measured by western blotting (Figure 3(a)) and flow cytometry (P < 0.05) (Figure 3(b)).In addition, MKT-077 treatment significantly inhibited the percentage of CD71+ cells (P < 0.01, Figure 3(c)), suggesting pharmacological inhibition of HSPA9/mortalin repressed erythroid maturation in CD34+ cells.

Knockdown of HSPA9 by shRNA inhibited erythroid maturation in human CD34+ hematopoietic progenitor cells
In order to further test the effects of HSPA9 knockdown in human CD34+ hematopoietic progenitor cells, we constructed two previously reported short hairpin RNAs (shRNAs) targeting the human HSPA9 gene [shHSPA9#1 (sh433 in prior study) and shHSPA9#2 (sh960 in prior study)]. 27The strategy and flowchart of lentiviral shRNA production and transduction are illustrated in Figure 4(a).The shRNAs targeting HSPA9 (shHSPA9#1 and shHSPA9#2) are resistant to the antibiotic puromycin.Cells were incubated in erythroid differentiation media with shRNAs for 7 days.We   measured the HSPA9/mortalin level by western blotting, and both shHSPA9#1 and shHSPA9#2 reduced the HSPA9 protein level to approximately 50% and 20%, respectively, compared to the control shRNA targeting green fluorescent protein (GFP) (Figure 4(b)).Consistent with the effects of inhibiting HSPA9 by siRNA and MKT-077, knockdown of HSPA9 by shRNA showed significant inhibition of erythroid maturation in human CD34+ cells cultured in erythroid differentiation media compared to control shRNA, indicated by the reduced percentage of CD71+ cells (Figures 4(c) and (d)).

TP53 inhibition reversed erythroid maturation disruption by HSPA9 inhibition in human CD34+ hematopoietic progenitor cells
In order to explore the molecular mechanism of HSPA9 regulating erythroid maturation, especially whether it is TP53-dependent or not, we constructed previously reported shRNAs targeting the human TP53 gene [shTP53#1 (#4 in prior study) and shTP53#2 (#3 in the prior study)]. 27The shRNAs targeting TP53 (shTP53) are resistant to antibiotic hygromycin.shTP53#1 and shTP53#2 were able to decrease p53 protein levels by ∼50% and 80%, respectively, compared to control short hairpin targeting GFP (shGFP) measured by western blotting (Figure 5(a)).Since we previously reported that knockdown of HSPA9 induces TP53dependent apoptosis in human CD34+ cells, next, we tested whether erythroid maturation inhibition induced by HSPA9 knockdown is also TP53-dependent or not.We cotransduced lentiviruses containing shHSPA9 and shTP53 into human CD34+ cells, followed by double antibiotic selection with puromycin and hygromycin, respectively.Cells were incubated in erythroid differentiation media with shRNAs for 7 days.The results of the experiment are shown in Figure 5, and additional statistical analyses are listed in Table 2.In cultures transduced with the shConthygrogmycin virus (i.e., shTP53 control virus), there was the expected reduction in CD71+ cells in shHSPA9#1 and shHSPA9#2 cultures compared to shCont-puromycin cultures (i.e., shHSPA9 control virus) (Figure 5(b), columns 4 and 7 compared to column 1, respectively).There was an increase in the percent of CD71+ cells in shHSPA9#1 and shHSPA9#2 cultures treated with shTP53#1 compared to shCont-hygromycin cultures (i.e., shTP53 control virus), but not in sh-Cont-puromycin cultures (i.e., shHSPA9 control virus) (Figure 5(b), columns 5 and 8 compared to column 2, respectively).There was a similar increase in the percent of CD71+ cells in shCont-puromycin, shHSPA9#1, and shHSPA9#2 cultures treated with shTP53#2 (Figure 5(b), columns 6 and 9 compared to column 3, respectively).These data suggest that erythroid maturation inhibition by HSPA9 knockdown is partially mediated through a TP53 mechanism.We also noticed that shTP53#2 affects control cells similar to shHSPA9 cells (Figure 5(b), column 3 compared to column 1).This is probably due to that TP53 is necessary for normal erythroid maturation at some level, and when its level is reduced to below 50% levels, it impacts normal erythroid maturation and actually induces it more.

Discussion
Mortalin, encoded by the HSPA9 gene, is a highly conserved heat shock chaperone belonging to the HSP70 family.It is predominantly presented in the mitochondria but is also found in other subcellular compartments including the plasma membrane, endoplasmic reticulum, and cytosol. 29HSPA9/mortalin is critical in regulating a variety of cell physiological functions such as response to cell stress, control of cell proliferation, and inhibition/prevention of apoptosis, 30 which may explain our observation that Hspa9 homozygous deletion mice are embryonic lethal. 23nemia, one of the significant clinical findings presented in MDS patients, is commonly treated with blood transfusions and erythropoietin. 31Bone marrow samples from del(5q) MDS patients display increased apoptosis associated with increased expression of TP53 and its target genes in erythroid cells. 32,33During the early stage of erythropoiesis, hematopoietic stem cells sequentially give rise to the common myeloid progenitor, megakaryocyte-erythrocyte progenitor, burstforming unit-erythroid, and colony-forming unit- The sequences of shRNAs used in the study are listed, including shRNAs targeting GFP (control), HSPA9, and TP53.

Table 2
Statistical analysis results of Figure 5.
Column A (column # in Figure 5) Column B (column # in Figure 5) Each group or column presented in Figure 5 was compared using one-way ANOVA with a Tukey post-test.

Statistics already labeled in
erythroid cells. 34Our present study addresses the effects of HSPA9 expression levels on erythroid maturation in human CD34+ hematopoietic progenitor cells.We observed that knockdown of the HSPA9 gene located on the proximal chromosome 5 CDR could inhibit erythroid maturation in CD34+ cells cultured in erythroid differentiation media, consistent with our prior results. 35In addition, our previous studies demonstrated that HSPA9 knockdown induces apoptosis in human CD34+ cells, through a TP53-dependent mechanism. 27n this report, we address whether inhibition of erythroid maturation by HSPA9 knockdown is TP53-dependent or not.We cotransduced lentiviruses containing shRNA targeting HSPA9 and TP53 genes into human CD34+ cells.We observed that HSPA9/ mortalin inhibition disrupted erythroid maturation by decreasing the percentage of CD71+ cells in a HSPA9 level-dependent manner.In addition, TP53 shRNAs could partially reverse the erythroid inhibition caused by HSPA9 shRNAs, indicating that erythroid maturation regulated by HSPA9 knockdown is also TP53-dependent.HSPA9/mortalin is reduced by ∼50% in del(5q) MDS cells, consistent with haploinsufficient levels. 20In this study we show that the proximal del(5q) candidate gene HSPA9 regulates erythroid maturation in human CD34+ cells, suggesting HSPA9/mortalin may be a potential target to treat anemia in del(5q) MDS patients by reactivating its expression from the residual wild-type nondeleted allele.Similarly, the ribosomal protein gene RPS14 is also a del (5q) candidate gene located on the distal chromosome 5 CDR.The Ebert lab reported that haploinsufficiency of RPS14 causes activation of p53 in human erythroid progenitor cells and a block in erythroid differentiation, as well as inducing apoptosis in a mouse model.They also reported that mice with Rps14 haploinsufficiency in hematopoietic cells developed a progressive anemia associated with the induction of a p53-dependent erythroid differentiation defect in late-stage erythroblasts. 32,36Huang et al. 37 reported that the deficiency of SF3B1, a core component of the splicing machinery, also impairs human erythropoiesis via activation of p53 in human CD34+ cells.Collectively, these and our studies confirm that insufficiency of specific MDSassociated genes, including HSPA9, could disrupt erythroid maturation via TP53 -dependent mechanisms, providing insights into ineffective erythropoiesis in MDS patients.
Studies have shown that p53 activation during ribosome biogenesis regulates normal erythroid differentiation using human CD34+ cells and mouse models. 38It has also been confirmed by multiple studies and in multiple cancer types that HSPA9/mortalin interacts with p53 and the regulation of apoptosis by HSPA9/mortalin is TP53-dependent. 27,39,40However, the precise mechanism of how HSPA9/mortalin regulates erythroid maturation, especially in del(5q) or 5q− syndrome, is unclear.As discussed above, some studies showed haploinsufficiency of certain del(5q) genes such as RPS14 led to p53 activation and underlies the anemia in the 5q− syndrome.For the HSPA9 gene, a clinical study 41 and a study using yeast 42 identified that HSPA9 mutations may contribute to congenital sideroblastic anemia, implicating a role of HSPA9 in erythroid maturation.In this study, we found that HSPA9/mortalin inhibition disrupts erythroid maturation dependent on TP53 using human CD34+ hematopoietic progenitor cells, consistent with the report by Caceres et al. 43 that TP53 suppression promotes erythropoiesis in del(5q) MDS.Collectively, our results suggest that the increased apoptosis and reduced erythroid maturation observed in del(5a)-associated MDS is TP53-dependent.Thus, HSPA9/mortalin may also be a potential target to treat anemia in del(5q) MDS patients, although the simultaneous loss of multiple genes on del(5q) likely contributes to the complex phenotypes observed in MDS.We believe that inhibition of erythroid maturation as indicated by decreasing the CD71 surface marker by HSPA9 knockdown is partially due to increased apoptosis through TP53, but there are additional mechanisms.In our study, when we measured the CD71 level by flow cytometry, we gated and measured in most live cells rather than apoptotic cells.This indicated that decreased CD71 is probably due to the impact of HSPA9 knockdown on erythroid differentiation.The mechanism of how HSPA9/mortalin regulates erythrogenesis has been explored.For example, Shan et al. 44 reported that HSPA9/ mortalin enhances the synthesis of mitochondrial iron-sulfur cluster in yeast, which is required for heme synthesis and erythroid differentiation.Yamamoto et al. 45 reported that mortalin cooperates with the inner mitochondrial translocase complex to facilitate the translocation of mitochondrial matrix proteins that are essential for mitochondrial function and cell viability in yeast.Therefore, along with the studies we have done, it is conceivable that depletion of HSPA9 could lead to increased mitochondrial dysfunction and activation of proapoptotic factors that induce hematopoietic progenitor cell death and maturation dysfunction.Intriguingly, we also showed that compared to progenitors of other lineages of hematopoiesis, a greater reduction of burst-forming unit-erythroid progenitors was observed when HSPA9 was knocked down in mice, suggesting that HSPA9 possibly plays an additional role(s) in maintaining the erythroid progenitor cell niche. 35However, it is unclear whether the decreased erythroid maturation marker CD71 caused by HSPA9 knockdown or chemical MKT-077 is due to higher levels of apoptosis in CD71 or not.First, from clinical aspects, del(5q) MDS patients are characterized by ineffective hematopoiesis (erythroid unable to mature) and peripheral blood cytopenia (due to increased apoptosis of bone marrow cells), both contributing to anemia in patients.Thus, it is hard to differentiate whether the clinical symptom of anemia is caused by apoptosis or erythroid maturation inhibition separately.Second, from basic science aspects, our studies have demonstrated that HSPA9 regulates both apoptosis 27 and erythroid maturation (this study) through a TP53-dependent mechanism.Thus, the precise relationship between apoptosis and erythroid maturation regulated by HSPA9 is well worth to be further investigated.

Cloning of lentiviral shRNA vector
shRNAs targeting the HSPA9 were cloned into pLKO.1 vector (Addgene Inc., Watertown, MA, USA) carrying the puromycin resistance gene and shRNAs targeting TP53 were cloned in the same vector carrying the hygromycin resistance gene following the manufacturer's protocol.shRNA oligonucleotides (oligos) were designed based on the information from the RNA interference (RNAi) Consortium of the Broad Institute and synthesized by Integrated DNA Technologies, Inc. (Coralville, IA, USA).shRNA sequences are listed in Table 1.Briefly, first, for oligo annealing, the oligos were resuspended in ddH2O to a concentration of 20 μM then mixed with 5 μL of forward oligo, 5 μL of reverse oligo, 5 μL of 10× New England Biolabs (NEB) buffer 2 (New England Biolabs, Inc., Ipswich, MA, USA), and 35 μL of ddH2O.The 40 μL mixture was incubated in an Eppendorf Mastercycler thermocycler (Hamburg, Germany) at 95 °C for 5 min, then 70 °C for 5 min, followed by lowering the temperature by 5 °C at every 5 min interval, until room temperature and held.Second, for pLKO.1 cloning vector digestion, the vector was digested with restriction enzymes (New England Biolabs, Inc.) AgeI and EcoRI follow the manufacturer's protocol.The digested vectors were purified using the QIAquick gel extraction kit (QIAGEN, Inc., Hilden, Germany).For vector ligating and transforming, a total of 20 μL final volume of mixture [including 2 μL of the annealed oligo, 20 ng of digested pLKO.1, 2 μL 10× NEB T4 DNA ligase buffer (New England Biolabs, Inc.), 1 μL NEB T4 DNA ligase (New England Biolabs, Inc.), and ddH2O] was incubated at 16 °C for 20 h.Next, 2 μL of ligation mix was transformed into 25 μL DH5-alpha E. coli competent cells (Invitrogen, Inc., Carlsbad, CA) following the manufacturer's protocol.The transformations were spread on ampicillin-selective plates and incubated overnight at 37 °C.Colonies were picked and cultured in Luria-Bertani (LB) medium containing 100 μg/mL ampicillin with shaking at 250 rpm overnight at 37 °C.On the next day, plasmid DNA was isolated by using the Invitrogen Pure-Link quick plasmid miniprep kit (Invitrogen, Inc.).The correct positive clones were confirmed by sequencing using the pLKO.1 sequencing primer (5′-CAA GGCTGT TAGAGAGATAATTGGA-3′) at the West Virginia University Genomics Core Facility.

Lentiviral shRNA production, transduction, and culture
The process of lentiviral shRNA production and transduction is illustrated in Figure 4(a).The pLKO.1 vector containing shRNA and lentiviral vectors were transfected into HEK293T cells using the calcium phosphate transfection method.The p8.9 and pMD.G vectors were used as packaging and envelop vectors, respectively.Vectors were transfected into HEK293T cells using the CalPhos mammalian transfection kit (Takara Bio USA, Inc., San Jose, CA, USA) following the manufacturer's protocol.The lentiviruses presented in the cell culture media were collected and titered prior to transduction into human CD34+ cells.The shRNAs targeting the human HSPA9 gene are resistant to the antibiotic puromycin, and shRNAs targeting TP53 (shTP53) are resistant to antibiotic hygromycin respectively.The transduction of lentiviral shRNA into human CD34+ cells and associated culture conditions have been described previously. 27,35Briefly, CD34+ progenitor cells were primed in X-VIVO 15 media with human cytokines overnight before lentiviral transduction.Cells were spinoculated in the presence of polybrene (5 μg/mL) and incubated overnight at 37 °C in 5% CO 2 .Cells were washed and incubated in erythroid differentiation media for 7 days.

Flow cytometry
Cell apoptosis and p53 levels were measured and analyzed using the BD Accuri C6 flow cytometry apparatus (BD Biosciences, Inc., San Jose, CA, USA).Cell apoptosis was measured using the PE Annexin V apoptosis detection kit (BD Biosciences, Inc.) following the manufacturer's protocol, and the percentage of apoptotic cells was detected and analyzed by flow cytometry, as described previously. 27To measure the level of intracellular p53, intracellular analysis by flow cytometry was used as described previously. 27,32Briefly, CD34+ cells were fixed with 2% paraformaldehyde for 15 min at 37 °C and permeabilized with ice-cold methanol for 30 min at −20 °C.Cells were then incubated for 1 h with the p53 antibody followed by analysis using flow cytometry.The following antibodies were used: CD71 (fluorescein isothiocyanate (FITC), 11-0719-42, eBioscience/Thermo Scientific, Inc., Waltham, MA, USA), glycophorin A (PE, 12-9987-80, eBioscience, Inc.), and p53 (Alexa647, 2533S, Cell Signaling Technology, Inc., Danvers, MA, USA).

Statistics
The statistical significance of the data between the two groups was analyzed by the Student t-test (GraphPad Prism 10).The statistical significance of the data with more than two groups was analyzed by one-way analysis of variance (ANOVA) with a Tukey post-test (GraphPad Prism 10).Significance levels were set at P < 0.05 (labeled *), P < 0.01 (labeled **), and P < 0.001 (labeled ***).

Significance statement
MDS is the most common adult myeloid blood cancer in the United States.Typical symptoms of MDS patients includes fatigue and pallor, which are associated with anemia.Some MDS patients harbor del(5q), with haploinsufficiency for a set of genes including HSPA9.We showed that inhibition of HSPA9 disrupts erythroid maturation in human CD34+ hematopoietic progenitor cells, through a TP53-dependent mechanism.Our findings not only indicate that reduced levels of HSPA9 may contribute to anemia observed in del(5q)-associated MDS patients, but also provide new insights into potential mechanisms of anemia.
Author contribution Christopher Butler: Performed research, analyzed and interpreted data; Morgan Dunmire: Performed research, manuscript writing; Jaebok Choi: Analyzed and interpreted data, manuscript writing; Gabor Szalai: Performed research, analyzed and interpreted data, manuscript writing; Anissa Johnson: Performed research, manuscript writing; Wei Lei: Analyzed and interpreted data, manuscript writing; Xin Chen: Analyzed and interpreted data, manuscript writing; Liang Liu: Analyzed and interpreted data; Wei Li: Analyzed and interpreted data, manuscript writing; Matthew J. Walter: Financial support, designed research, performed research, analyzed and interpreted data, manuscript writing; Tuoen Liu: Financial support, designed research, performed research, analyzed and interpreted data, manuscript writing.

Fig. 1
Fig. 1 Knockdown of HSPA9 by siRNA increases p53 expression in human CD34+ hematopoietic progenitor cells.(a) Genetic model of del(5q).Two CDRs are presented in del(5q) MDS patients.Proximal CDR contains HSPA9 and its deletion is associated with the risk of developing AML.(b) Expression of HSPA9/mortalin and p53 in human CD34+ cells after siRNA transfection for 72 h measured by western blotting.Beta-actin was used as the loading control.The blots shown are a presentation of two independent experiments.(c) Flow cytometry plot of p53 in human CD34+ cells after siRNA transfection for 72 h.(d) Representative histogram of intracellular p53 levels measured by flow cytometry in human CD34+ cells following HSPA9 knockdown by siRNA.(d) Mean fluorescence intensity of p53 following HSPA9 knockdown by siRNA.All error bars represent SD, N = 3 technical replicates, representative of two independent experiments.**P < 0.01.Abbreviations used: AML, acute myeloid leukemia; CDR, commonly deleted regions; MDS, myelodysplastic syndromes; SD, standard deviation; siRNA, small interfering RNA.

Fig. 2
Fig. 2 Knockdown of HSPA9 by siRNA increases apoptosis and inhibits erythroid maturation in human CD34+ hematopoietic progenitor cells.(a) The number of human CD34+ cells was measured after siRNA transfection for 5 days.The fold change in cell counts was calculated relative to the number of cells on day 1 (N = 3 for each siRNA).(b) Quantification of annexin V+ cells, which were considered as apoptotic cells, measured by flow cytometry following HSPA9 knockdown by siRNA in human CD34+ cells (N = 3 for each siRNA).(c) Representative plot of erythroid maturation following HSPA9 knockdown by siRNA measured by flow cytometry (N = 3 for each siRNA).(d) Quantification of CD71+ cells following HSPA9 knockdown by siRNA measured by flow cytometry (N = 3 for each siRNA).All error bars represent SD, N = 3 technical replicates, representative of two independent experiments.***P < 0.001.Abbreviations used: SD, standard deviation; siRNA, small interfering RNA.

Fig. 5
Fig. 5 TP53 inhibition partially reverses erythroid maturation disrupted by HSPA9 inhibition in human CD34+ hematopoietic progenitor cells.(a) Expression of p53 after knockdown by shRNAs in human CD34+ cells measured by western blotting.Beta-actin was used as the loading control.The blots shown are a presentation of two independent experiments.(b) CD34+ cells were cotransduced with lentiviral constructs carrying an shRNA targeting TP53 with a hygromycin resistance gene (shGFP/ control, shTP53#1, or shTP53#2) and an shRNA targeting HSPA9 with a puromycin resistance gene (shGFP/control, shHSPA9#1, or shHSPA9#2).Cells were grown in the presence of both hygromycin and puromycin and the fold change in the percentage of CD71+ cells was measured by flow cytometry (N = 3 technical replicates, representative of 2 independent experiments).*P < 0.05, **P < 0.01, ***P < 0.001.Abbreviation used: GFP, green fluorescent protein; shGFP, short hairpin targeting GFP; shRNA, short hairpin RNA.