Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Jan 1.
Published in final edited form as: Am J Hematol. 2023 Nov 6;99(1):99–112. doi: 10.1002/ajh.27145

Phenotypic and Proteomic Characterization of the Human Erythroid Progenitor Continuum Reveal Dynamic Changes in Cell Cycle and in Metabolic Pathways

Julien Papoin 1,2,*, Hongxia Yan 3,*, Marjorie Leduc 4,*, Morgane Le Gall 4, Anupama Narla 5, James Palis 6, Laurie A Steiner 6, Patrick G Gallagher 7,8, Christopher D Hillyer 3, Emilie-Fleur Gautier 4,#, Narla Mohandas 3,#, Lionel Blanc 1,9,10,#
PMCID: PMC10877306  NIHMSID: NIHMS1940525  PMID: 37929634

Abstract

Human erythropoiesis is a complex process leading to the production of 2.5 million red blood cells per second. Following commitment of hematopoietic stem cells to the erythroid lineage, this process can be divided into three distinct stages: erythroid progenitor differentiation, terminal erythropoiesis, and reticulocyte maturation. We recently resolved the heterogeneity of erythroid progenitors into four different subpopulations termed EP1 to EP4. Here, we characterized the growth factor(s) responsiveness of these four progenitor populations in terms of proliferation and differentiation. Using mass spectrometry-based proteomics on sorted erythroid progenitors, we quantified the absolute expression of ~5,500 proteins from EP1 to EP4. Further functional analyses highlighted dynamic changes in cell cycle in these populations with an acceleration of the cell cycle during erythroid progenitor differentiation. The finding that E2F4 expression was increased from EP1 to EP4 is consistent with the noted changes in cell cycle. Finally, our proteomic data suggest that the protein machinery necessary for both oxidative phosphorylation and glycolysis is present in these progenitor cells. Together, our data provide comprehensive insights into growth factor-dependence of erythroid progenitor proliferation and the proteome of four distinct populations of human erythroid progenitors which will be as useful framework for the study of erythroid disorders.

Keywords: erythropoiesis, erythroid progenitors, cell cycle, proteomics

Graphical Abstract

graphic file with name nihms-1940525-f0007.jpg

Introduction

Erythropoiesis involves a series of differentiation steps beginning with self-renewal of hematopoietic stem and progenitor cells (HSPCs). HSPCs give rise to more committed hematopoietic progenitors including myeloid progenitors, erythroid progenitors and megakaryocytes1. Erythroid progenitors are classically divided into erythroid burst-forming unit (BFU-E) and erythroid colony-forming units (CFU-E). During terminal erythroid differentiation CFU-E give rise to proerythroblasts, the first morphologically recognizable erythroid precursors in the bone marrow, which subsequently undergo five cell divisions to generate orthochromatic erythroblasts. Enucleation of orthochromatic erythroblasts leads to generation of nascent reticulocytes2 that mature into erythrocytes3,4.

While significant progress has been made in our understanding of terminal erythropoiesis, detailed cellular and molecular characterization of the erythroid progenitors is lacking. BFU-E and CFU-E were first identified based on their ability to form erythroid colonies in semi-solid medium5–8 and more recently, by immuno-phenotyping using cell surface markers and flow cytometry9. Using different combinations of cell surface markers, we were able to identify a population of immature CFU-E, which exists in the adult (in contrast to the neonate) in the continuum of erythroid progenitors10,11. Furthermore, we defined 4 distinct subpopulations of erythroid progenitors (EP1 to EP4) in the bone marrow of healthy individuals12.

A characteristic feature of BFU-E and CFU-E is their differential response to cytokines. While BFU-E need both stem cell factor (SCF) and erythropoietin (EPO) to proliferate, CFU-E need only EPO5,6,13. These distinct characteristics have been used for the development of in vitro culture systems to study human erythropoiesis in health and disease.

In the present study, we characterized the growth and differentiation potential of EP1 to EP4 in response to increasing concentrations of SCF and EPO. The four progenitor populations exhibited differential dose-dependent responses to SCF and EPO, while their growth and differentiation was IL-3 independent. Using mass spectrometry-based proteomics on sorted erythroid progenitors, we quantified the expression levels of ~5,500 proteins from EP1 to EP4. Further functional analyses highlighted dynamic changes in cell cycle in these populations with an acceleration of the cell cycle during erythroid progenitor differentiation. The finding that expression levels of E2F4 protein was increased from EP1 to EP4 correlates with the noted changes in cell cycle. From an energetic perspective, we observed that the protein machinery necessary for both oxidative phosphorylation and glycolysis is present in these fast-cycling cells. Taken together, our findings provide a comprehensive characterization of human erythropoiesis at the progenitor level and resources to improve our understanding of the contribution of erythroid progenitors to normal and disordered erythropoiesis.

Methods

Human biological samples

Non-mobilized peripheral blood CD34+ cells were isolated from Leukopaks obtained from the New York Blood Center. Bone marrow CD34+ cells were isolated from hip surgery samples collected through the Tissue Donation Program at Northwell Health and hip aspirates from the Hospital for Special Surgery. All studies were conducted in accordance with the declaration of Helsinki and under institutional review board (IRB) approval of Northwell Health, the New York Blood Center and the Hospital for Special Surgery.

Cell culture

CD34+ cells were cultured in-vitro and differentiated towards erythrocytes using a modified version of a previously described three-phase culture system11,12,14. Briefly, CD34+ cells were cultured in IMDM supplemented with 3% (vol/vol) AB+ human serum, 2% (vol/vol) human plasma, 10μg/mL insulin, 3U/mL Heparin, 200μg/mL transferrin, 1ng/mL interleukin-3 (IL-3), 3IU/mL erythropoietin (EPO), 10ng/mL Stem cell factor (SCF) and 100U/mL Penicillin/Streptomycin (P/S). After 4–5 days, cells were collected and stained with a cocktail of antibodies for FACS-based cell sorting. To monitor their responses to different growth factors, purified erythroid progenitor subpopulations were further cultured in vitro for 2 additional days using the three-phase culture system, with different growth factors concentrations (Supplemental Table 1).

Flow cytometry and FACS cell sorting

For FACS-based cell sorting and detection of erythroid progenitor differentiation, cells were fluorescently labeled with a cocktail of antibodies containing an anti-glycophorin A (GPA) BV421–conjugated, anti-CD105 PECF594–conjugated, anti-CD71 APC-H7–conjugated, anti-IL-3R PE-Cy7–conjugated, anti-CD45RA Alexa Fluor 700-conjugated, anti-CD41a BV510-conjuated, anti-CD117 PE-conjugated, anti-CD34 APC-conjugated. 7-AAD dye was used for exclusion of dead cells. All flow cytometric analysis was performed using a BD LSRFortessa™ Cell Analyzer, and cell sorting was done by a BD FACSAria™ II Cell sorter.

Cell proliferation assay

CellTrace™ Violet Cell Proliferation Kit for flow cytometry (Invitrogen, Catalog number C34557) was used to study proliferation of isolated erythroid progenitor subpopulations, EP1 to EP4. Sorted populations were labelled with 2.5uM staining solution at 37 °C for 20 minutes. Extra free dye was removed by diluting with five times excess volume of dye-free medium (containing >1% protein) and incubated for 15 minutes at 37 °C. The cells were then spun down and resuspended in fresh complete medium. After 2 days of culture, cell proliferation was analyzed for fluorescent dye dilution by flow cytometry.

Cell cycle analyses

CD34+ cells were cultured for 4–5 days in the phase I medium of the three-phase culture system described above. EdU incorporation in erythroid progenitor cells was measured using Click-iT™ PLUS EdU flow cytometry assay kit (Invitrogen, catalog number C10632). Briefly, cells were incubated with 10 uM EdU for 2 hours and were subsequently fixed and permeabilized for EdU detection. Cells were then analyzed by flow cytometry following additional staining with cell surface markers and 7-AAD staining for DNA content.

Mitochondria staining

To assess mitochondrial biomass and membrane potential during erythroid progenitor differentiation, cells were stained with MitoTracker probes. 500,000 cells were incubated with 100nM of Mitotracker Green FM and Red CMXRos at 37°C for 20 minutes. The labelled cells were subsequently stained for cell surface markers, washed twice with PBS/2% FBS and analyzed by flow cytometry.

Sample preparation for proteomic analysis

CD34+ cells were cultured for 5 days and 250,000 to 500,000 of each EP population were isolated by FACS in fetal bovine serum (FBS) to ensure cell viability. Sorted cells were then washed twice with cold PBS and boiled in lysis buffer containing Tris HCL 100mM pH8.5 and 2% SDS at 95°C for 5 minutes. A total of 50μg of proteins per sample were reduced and alkylated using TCEP 20mM and chloroacetamide 50mM and digested by filter-assisted sample preparation (FASP) using trypsin. Resulting peptides were fractionated on strong cation exchange chromatography (SCX) StageTips into five fractions.

nLC-MS/MS proteomic analysis

Peptides from each of these five fractions were solubilized in 0.1% trifluoroacetic acid (TFA) and 10% acetonitrile (ACN) and loaded, concentrated, and washed for 3 min on a C18 reverse phase precolumn (3μm particle size, 100 Å pore size, 75 μm inner diameter, 2 cm length, Thermo Fisher Scientific) with loading solvent containing 0.1% TFA and 2% ACN. Peptides were then separated by C18 rapid separation liquid chromatography (RSLC) Dionex U3000 on an Aurora C18 reverse phase column (1.6 μm particle size, 100 Angström pore size, 75μm inner diameter, 25cm length) with a 3 h binary gradient from 99% solution A (0.1% formic acid in H2O) to 55% solution B (80% ACN, 0.085% formic acid) before injection into an Orbitrap Fusion mass spectrometer (Thermo Scientific).

The Orbitrap Fusion mass spectrometer acquired data throughout the elution process and operated in a data-dependent scheme with a cycle time of 3 s with full mass spectrometry (MS) scans acquired with the orbitrap detector, followed by HCD fragmentation and Ion trap fragment detection of the most abundant ions detected in the MS scan. Mass spectrometer settings for full scan MS were: 1.0E6 AGC, 60,000 target resolution, 350–1500 m/z range, maximum injection time of 60 ms. HCD MS/MS fragmentation was allowed for 2–7+ precursor ions reaching more than 5.0E4 minimum intensity. Quadrupole-filtered precursors within 1.6 m/z isolation window were fragmented with a Normalized Collision Energy of 30. The limiting ions accumulation values were 1.0E5 AGC Target and 60 ms maximum injection time. The Ion-trap detector was used for its fast and sensitive detection capabilities. A 30 s dynamic exclusion time was set.

nLC-MS/MS proteomic data analysis

Maxquant software version 1.6.17.015,16 was used for the analysis of raw data from mass spectrometer. The database used was a concatenation of human sequences from the Uniprot-Swissprot database (release 2020–03) and the list of contaminant sequences adapted from Maxquant. The enzyme specificity was trypsin. Carbamidomethylation of cysteins was set as constant modification and acetylation of protein N-terminus and oxidation of methionines were set as variable modification. Second peptide search was allowed and minimal length of peptides was set at 7 amino acids. False discovery rate (FDR) was kept below 1% on both peptides and proteins. Label-free protein quantification (LFQ) was done using both unique and razor peptides. At least 2 such peptides were required for LFQ ratio. The “match between runs” (MBR) option was allowed with a match time window of 0.7 min and an alignment time window of 20 min.

For statistical analysis, Perseus and R softwares were used. Absolute protein quantification was performed by standardization using histones MS signal with Perseus proteomic ruler17. Only proteins with at least 3 out of 4 replicates valid LFQ intensity value per condition were selected for statistics. Paired t-tests comparing EP1 to EP2, EP2 to EP3, and EP3 to EP4 were carried out on these proteins and those showing Pvalue<0.05 were considered as significant.

The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the Pride partner repository with the dataset identifiers PXD042615.

In silico functional analysis

Functional analyses were generated using IPA (QIAGEN Inc., https://www.qiagenbioinformatics.com/products/ingenuitypathway-analysis) version 73620684 for each list of differential proteins. Significantly over-represented biological terms (canonical pathways and functions) were identified with a right-tailed Fisher’s Exact Test that calculates an overlap p-value determining the probability that each term associated with our lists of differential proteins was not due to chance alone. The activation status was evaluated by calculating the z-score which is a statistical measure of correlation between relationship direction and experimental protein expression. Its calculation assessed the activation (positive z-score) or repression (negative one) of each term. To be considered significant the z-score has to be greater than 2 in absolute value.

Statistical analyses

Results represent the average ± standard error of the mean (SEM). All statistical analyses comparing the different groups were performed using GraphPad Prism 10 software for two tailed Student’s t-test or ANOVA. A P value of less than 0.05 was considered statistically significant.

Results

The proliferation and differentiation responses to growth factors of EP1 to EP4 are stage-dependent.

It is well-established that erythropoiesis relies on both SCF and EPO for growth, survival, and differentiation5,6. Two- or three-phase in vitro culture systems from hematopoietic stem and progenitor cells (CD34+ cells, HSPCs) have used a combination of these two cytokines with the addition of IL-3 in the first phase of culture, to improve the expansion of CD34+ cells. We previously demonstrated that erythroid progenitors could be divided into four subpopulations, from EP1 to EP4 as they mature towards the pro-erythroblast stage. However, their individual response to these cytokines remains unclear. To address this issue, we isolated CD34+ cells from the peripheral blood of healthy adult individuals and cultured them for 5 days in a 3-phase culture media11,14. The four erythroid progenitor populations (EP1 to EP4) were then sorted based on our previously described flow strategy12 and further cultured for 2 days (Figure 1A) in the presence of different concentrations of IL-3, SCF or EPO (Supplemental Table 1). Increasing concentrations of IL-3 did not affect the proliferation of any of the EP population (Figure 1B). Even in the absence of IL-3, EPs proliferated normally, suggesting that their growth is independent of IL-3. As assessed by the percentage of GPA+ cells and the CD105/GPA pattern, differentiation of the sorted EP populations over two days was not affected by IL-3 (Figure 1C, D).

Figure 1. Dose-dependence responses of erythroid progenitor subpopulations to growth factors.

Figure 1.

Open in a new tab

Peripheral blood derived CD34+ cells were differentiated for 5 days in modified 3-phase medium as described in the methods and four erythroid progenitor populations including EP1, EP2, EP3 and EP4 were sorted and cultured for an additional 2 days according to the gating strategy presented in (A) for erythroid maturation in the presence of different doses of IL3, SCF and EPO. (B) No significant effects of IL3 on EPs proliferation were observed, with presence of 100 ng/mL SCF and 3 IU/mL EPO in all groups. (C) Representative flow cytometric plots of CD105 and GPA showing differentiation of sorted EPs with or without presence of IL3, after 2 days of culture. (D) No significant effects of IL3 on EPs differentiation were observed, as demonstrated by GPA+ %. (E) Effects of SCF on EPs proliferation, with presence of 3 IU/mL EPO in all groups. (F) Representative flow cytometric plots of CD105 and GPA showing differentiation progression of sorted EPs with presence of different concentration of SCF, after 2 days of culture. (G) Effects of SCF on EPs differentiation, as shown by GPA+ % after 2 days of differentiation. (H) Effects of EPO on EPs proliferation, with presence of 100 ng/mL SCF. (I) Representative analysis of EPs’ apoptosis with or without presence of different concentration of EPO, after 4 days of culture. (J) Representative flow cytometric plots of CD105 and GPA showing differentiation progression of sorted EPs with presence of different concentration of EPO, after 2 days of culture. (K) Effects of EPO on EPs differentiation, as shown by GPA+ % after 2 days of differentiation. Statistical significance was determined by two-tailed Student’s T-test (*p<0.05; **p<0.01; ***p<0.001). Means ± SEM are presented (n=3–4).

In contrast, SCF and EPO had a pronounced effect on the proliferation and differentiation of EPs. In the absence of SCF, proliferation of EP1 was inhibited, while EP2 to EP4 increased 100-fold in cell number in culture. Increasing the concentration of SCF led to an increase in the cell number, with EP1 and EP2 being the most responsive to 10 or even 100ng/mL SCF (Figure 1E). With regards to the differentiation kinetics, we observed that each of these populations was dependent on SCF, although there was no dose-dependence effect between 10 and 100ng/mL in terms of differentiation (Figure 1F, G). Indeed, the presence of SCF delayed the differentiation of erythroid progenitors towards more mature stages. These results confirm previous studies indicating that SCF enhances proliferation and restrains differentiation of erythroid progenitors18. Further, it appears that while the proliferation of erythroid progenitors is dependent on SCF in a dose-dependent manner, SCF inhibits differentiation similarly at the two doses tested, at least in vitro.

In the absence of EPO, the proliferation of all erythroid progenitor populations was inhibited (Figure 1H). There was a progressive increase in the amount of cell death from EP1 to EP4, as measured by lactadherin staining, a marker for phosphatidylserine exposure (Figure 1I). These data confirm a role for EPO as survival factor at all stages in the erythroid progenitor continuum19. However, it should be noted that EP1 survival was less dependent on EPO. As little as 0.2U/mL of EPO was sufficient to maintain survival of all EP populations (Figure 1J). Proliferation of each developmental stage was dependent on the dose of EPO, although EP4 had reached saturation at 0.2IU/mL (Figure 1H). In terms of differentiation, flow cytometry analyses based on CD105 and GPA showed that while EP3 and EP4 were responsive to increasing concentrations of EPO (Figure 1J,K), EP2 was less dependent of EPO concentrations. These results indicate that while EPO contributes to the proliferation of EP1 to EP4, it is required for the survival of EP2 to EP4 and the proliferation of EP3 and EP4.

Taken together, these data demonstrate different and specific growth factor requirements for survival and proliferation of differentiating erythroid progenitors.

Analyses of sorted erythroid progenitor populations reveal discrete changes at the proteome level.

To further characterize the subpopulations of EPs, we used an unbiased proteomic approach using mass spectrometry on sorted EP1 to EP4. This led to the identification of 5,554 unique proteins (Supplemental Table 2). The mean number of expressed proteins did not appear to change dramatically from EP1 to EP4 (Figure 2A), and 5,122 proteins (92.2%) were shared among the four populations (Figure 2B). As quality control, Pearson correlations between samples of the same population are particularly high for different biological replicates (>0.95, Supplemental Figure 1A), reflecting high reproducibility among replicates. Moreover, within the different ribosomal proteins, the mean copy numbers for EP1 to EP4 were close together, around one log2, which is expected in this type of analysis (Supplemental Figure 1B). To further assess the quality of our dataset, we performed hierarchical clustering based on the log2 of copy numbers highlighting two distinct groups, EP1/EP2 and EP3/EP4 (Figure 2C), with EP3 and EP4 being more similar than EP1 and EP2. For Principal component analysis, we used the protein expression profile (Figure 2D). In contrast to erythroid precursors that separate on the x-axis20, erythroid progenitors separate on the y-axis, suggesting that while most of the protein content remains constant through EP1 to EP4, specific changes occur strictly depending on the extent of erythroid progenitor differentiation. Although the different biological replicates for each progenitor population separated on the x-axis, correlations between the two most distant replicates for each EP stage remain high (between 0.96 and 0.98) (Figure 2E).

Figure 2. Overview of proteomic data.

Figure 2.

Open in a new tab

(A) Mean of quantified proteins for the different EPs. (B) Venn diagram showing shared proteins identified between EPs. (C) Heatmap of EPs mean log2(Copy number), the hierarchical clustering distance is Euclidean. (D) Principal Component Analysis done using z-score (Copy number) for proteins having at least 3 quantification values in a least one condition and after imputation of missing values. (E) Scatter plot representations and Pearson correlation factors for the most distant replicates at each EP stage.

Dynamics of protein expression during erythroid progenitor differentiation

The total protein content remained constant throughout differentiation from EP1 to EP4. However, the synthesis of globin chains and proteins involved in heme synthesis and iron metabolism increased from EP1 to EP4 (Supplemental Figure 2). While transcription factors central to erythropoiesis (e.g. GATA1, KLF1) showed increased expression during differentiation, others such as GATA2 could not be quantified, further validating the purity of our populations (Figure 3A). Interestingly, RUNX1, a transcription factor essential at the megakaryocyte/erythrocyte progenitor (MEP) stage and for megakaryopoiesis21, decreased from EP1 to EP4 but continued to be expressed in EP4. The expression levels of several proteins involved in nutrient transport increased from EP1 to EP4 except for SLC2A9 and SLC12A2, two ion channels whose expression levels decreased from EP1 to EP4 (Figure 3B). Among amino acid transporters, we noted an increase in SLC7A1 between EP1 and EP4, a cationic transporter responsible for arginine uptake which was recently shown to control human erythroid specification through eIF5A activation22. Mitochondrial membrane proteins were also quantified, and we noted that most of these proteins were expressed at a constant level during erythroid progenitor differentiation except for SLC25A21, which transports C5-C7 oxodicarboxylates across the inner mitochondrial membrane, whose expression increased during progenitor maturation. (Figure 3C). Finally, the absolute amount of the different ribosomal proteins, such as RPS19 and RPL5, important for erythropoiesis, did not change from EP1 to EP4, except for RPL37A, RPS27L, and MRPL54 (Figure 3D). Of note, while the quantity of individual mitochondrial ribosomal proteins is generally lower than their cytosolic counterparts, we noted high level of expression for MRPL27. Conversely, we detected lower quantities of RPS4Y1 and RPS27L than the rest of the cytosolic ribosomal proteins.

Figure 3. Evolution of specific proteins and dynamics of the proteome during erythroid progenitors’ differentiation.

Figure 3.

Open in a new tab

Quantitative expression in copy number per cell of (A) erythroid-specific transcription factors, (B) transporters for cell nutrients and amino acids, (C) mitochondrial membrane transporters, and (D) ribosomal proteins. (E) Heatmap representation of z-score of proteins differentially expressed (Anova test with q-value < 5%). (F) TOP 20 of enriched pathways between each adjacent stage of progenitor differentiation selected on their p-value.

Having shown that the absolute protein content remains constant from EP1 to EP4 (Figure 2), we generated a heatmap based on the expression profile of proteins differentially expressed between the four EP populations (ANOVA Q-Value<0.05, Figure 3E). Expression profiles showed important changes from EP1 to EP3 while EP3 and EP4 populations were more similar. We also noted 2 major clusters of proteins: one with increasing levels of expression and the other with decreasing levels of expression from EP1 to EP4. While the increase in expression occurred progressively from EP1 to EP4, the decrease occurs specifically between EP2 and EP3.

Over-representation analyzes demonstrated a significant enrichment in proteins involved in different biological pathways. Based on their Fisher-exact test p-value, we noted that the most dramatic changes in terms of pathways occurred from EP1 to EP3, with more subtle changes from EP3 to EP4 (Figure 3F). Based on their z-score, most of the cellular processes were activated in EP2 compared to EP1 (Figure 3F top panel) notably those related to the cell cycle and its regulation. In striking contrast, among the top 20 canonical pathways between EP2 and EP3, only the RHO signaling pathway was activated in EP3 compared to EP2; all other cellular processes were inhibited (Figure 3F middle panel). No major functional change in term of activation status was detected between EP3 and EP4 (Figure 3F bottom panel), due to reduced number of differentially expressed proteins: 77 compared to the other adjacent stages (372 for EP2 vs. EP1 and 269 for EP3 vs. EP2).

Expression levels of cell cycle regulators from EP1 to EP4.

EPs contribute to the erythroid output through their ability to proliferate. Thus, we chose to further characterize changes in the cell cycle pathway and its regulation during transition from EP1 and EP2. Our proteomics data led to development of a predictive model (Figure 4A). Among the differentially expressed proteins between EP1 and EP2, IPA identified an over-representation of proteins involved in Cyclin & Cell Cycle regulation pathway (p-value=2,96E-02). The model based on experimental expression of the seven involved proteins and relationships between molecules, IPA predicted an activation of the downstream S phase event reflecting an increase in erythroid progenitor proliferation between EP1 and EP2. This mitotic effect is supported by the predicted activation of Mitotic Roles of Polo-Like Kinase pathway (z-score=2,646) between EP1 and EP2. Based on this model, the expression levels of the cyclins and cyclin-dependent kinases were quantified from EP1 to EP4 (Figure 4B–C). The expression pattern of these proteins remained essentially constant during EP differentiation, except for CCNB1 (encoding Cyclin B1), whose copy number nearly doubled from EP1 to EP4.

Figure 4. Cell cycle analyses and identification of E2F4.

Figure 4.

Open in a new tab

(A) Cyclin & Cell Cycle regulation pathway is the sixth most relevant pathway between the adjacent stages EP1 and EP2 according to IPA (p-value=2,96E-02). The 7 proteins involved in this pathway that are upregulated in EP2 compared to EP1 are colored in red. Some of them are grouped in the same shape. CCNA2 is annotated “Cyclin A” as CCNB1 & CCNB2 are merged in “Cyclin B”. ABL1 & E2F4 are members of “E2F-RB1”and PPP2R2A & PPP1R5D are grouped into “PP2A” node. Based on the seven experimental expression values and relationships between pathway’s nodes, IPA predicts the activation status of other involved proteins: blue ones are predicted to be inhibited and orange ones activated. As for cell cycle stages, IPA predicts that S phase event is promoted downstream of the cascade. (B) Cyclins and (C) Cyclin-dependent kinases expression during progenitors’ differentiation. (D) E2F4 protein expression in each progenitor stages. (E) Western blot analyses on sorted populations for the proteins indicated on the left. Statistical significance was determined by two-tailed Student’s T-test (*p<0.05; **p<0.01; ***p<0.001).

The cell cycle is controlled at multiple levels; one of them being the G1 to S transition. Members of the E2F family are involved in this transition and we observed that E2F4 protein copy numbers were increased 3–4-fold from EP1 to EP4 (Figure 4D).

To validate our proteomics findings, we measured the expression levels of E2F4 in different populations of EPs obtained from the bone marrow of normal human donors. Western blot analyses on primary, uncultured sorted populations12 showed that the total levels of E2F4 increased dramatically from EP1 to EP4 (Figure 4E), validating our proteomic findings.

Human erythroid progenitors cycle faster as they differentiate.

To build on our findings with E2F4 expression levels, we investigated the dynamics of the cell cycle from EP1 to EP4. To do so, we first measured the proliferation potential of sorted EPs over two weeks of culture, under optimal concentrations of SCF (100ng/mL) and EPO (3U/mL). Consistent with the degree of immaturity, EP1 proliferated the most, followed by EP2 and EP3 and EP4 (Figure 5A). While the overall kinetics appeared similar, we noticed a difference within the first few days of culture (Figure 5A, red circle). Indeed, during the first two days of culture, sorted EP3 and EP4 cells presented a higher rate of proliferation than EP1 and EP2, as evidenced by the fold change in total cell number (Figure 5B). We then assessed the number of cell divisions over two days using CellTrace Violet (Figure 5C) and noted that, in line with the higher numbers of cells produced from sorted EP3 and EP4, these cells had divided more frequently than sorted EP1 or EP2 cells (Figure 5D). Together, these results suggest that as EPs differentiate, they cycle faster.

Figure 5. Proliferation kinetics of EPs are coupled with distinct cell cycle patterns.

Figure 5.

Open in a new tab

(A) Growth curve of sorted EP1, EP2, EP3 and EP4 under optimal concentration of SCF (100 ng/mL) and EPO (3 IU/mL). The red circle marks different proliferation kinetics of EPs. (B) Quantitative analysis of EPs proliferation after 2 days of in vitro culture. (C) Representative flow cytometric analysis of EPs proliferation in 2 days using CellTrace™ Violet dye. (D) Quantitative analysis of Violet dye dilution using fold change of Median Fluorescence Intensity. (E) Representative flow cytometric plots showing stage-wise analysis of cell cycle kinetics during early erythropoiesis by EdU (5-ethynyl-2´-deoxyuridine) incorporation assay, using primary human bone marrow cells. (F) Quantitative analysis of EPs at the different phases of the cell cycle in the primary human bone marrow. (G) Normalized mean fluorescence intensity (MFI) of the S-phase for each EP and ProEB stage in primary bone marrow. (H) Representative flow cytometric plots showing stage-wise analysis of cell cycle kinetics during early erythropoiesis, using an in vitro model of human erythropoiesis. (I) Quantitative analysis of EPs at the different phases of the cell cycle in an in vitro model of human erythropoiesis. (J) Normalized mean fluorescence intensity (MFI) of the S-phase for each EP and ProEB stage in the in vitro system. Statistical significance was determined by two-tailed Student’s T-test except for (G and J) where ANOVA was used (*p<0.05; **p<0.01; ***p<0.001, #p<0.0001). Means ± SEM are presented (n=3–5).

Recent studies demonstrated that the dual use of EdU and DNA content as markers can not only resolve the different phases of the cell cycle, but can also provide insights into the S-phase speed by measurement of the mean fluorescence intensity (MFI) of the S phase23–25. We thus measured the different phases of the cell cycle from EP1 to EP4 on sorted populations from primary bone marrow (Figure 5E) and observed a decrease in the percentage of cells in the G1 phase of the cell cycle and a concomitant increase in cells in S phase as EPs differentiated from EP1 to EP4 (Figure 5F). While the proportion of cells in G2/M progressively increased from EP1 to EP3, we observed a significant decrease from EP3 to EP4. Finally, in terms of S phase speed, we noticed an increase in the MFI from EP1 to Proerythroblasts (Figure 5G), suggesting an acceleration of the S phase as differentiation proceed, contributing to faster cycles for EP3 and EP4, when compared to EP1 and EP2.

Interestingly, the relative proportions of G1/S/G2/M were different in EPs isolated from in vitro cultures of peripheral blood-derived CD34+ and in these cells the S phase speed was not changed, suggesting that culture conditions are affecting cell cycle dynamics (Figure 5H–J).

Expression of proteins contributing to oxidative phosphorylation and glycolysis

Glycolysis is the essential energy source in mature erythrocytes26. Indeed, mitochondria are lost during terminal erythropoiesis27. While numerous studies have focused on the metabolic pathways during terminal erythroid differentiation, much less is known during the differentiation of erythroid progenitors.

We first evaluated the mitochondrial biomass and mitochondrial potential during erythropoiesis using Mitotracker Green and Red respectively. The mitochondrial biomass increased from EP1 to EP4 (Figure 6A). We observed the same pattern for the mitochondrial potential. These data suggest that oxidative phosphorylation activity is high and increases from EP1 to EP4.

Figure 6. Metabolic regulation of erythroid progenitors.

Figure 6.

Open in a new tab

(A) The MFI of the mitotracker green and mitotracker red was obtained for each EP population to determine the mitochondrial biomass and mitochondrial potential respectively. (B) Heatmap representation of mitochondrial proteins expression in z-score (Copy number). (C) Oxidative phosphorylation pathway is the eighth most activated pathway between the adjacent stages EP2 and EP1 according to IPA (z-score = 2.236). The 5 differentially involved proteins upregulated in EP2 compared to EP1 are colored in red. This trend is expected in the context of oxphos activated pathway. NDUFB9 & MT-ND5 are part of complex I, MT-CO2 (annotated “COX2”) in the complex IV and ATP5A1 in the complex V. As for CYCS, it belongs to “CYT C” group. Based on the five experimental expression values and relationships between pathway’s nodes, IPA predicts the upregulation of other involved proteins and downstream metabolites (orange nodes). (D) Expression of glycolytic enzymes. (E) Schematic representation of glycolysis along with the expression of the different proteins involved. (F) Differential expression of proteins between each adjacent stage of progenitors. Statistical significance was determined by two-tailed Student’s T-test (*p<0.05; **p<0.01; ***p<0.001). Means ± SEM are presented (n=5). ns: not significant.

Based on their copy number, 110 out of the 765 mitochondrial proteins (according to Uniprot Keyword KW-0496) identified in our proteomic analyses showed a change in expression, with most of them showing a tendency to increase from EP1 to EP4 (with mean EP1<EP2<EP3<EP4 expression level) (Figure 6B). Among these, we focused on proteins involved in the respiratory chain between EP1 and EP2. Considering the differentially expressed proteins between EP1 and EP2, IPA highlighted a significant activation of Oxidative phosphorylation pathway (z-score = 2.236) in EP2 compared to EP1. Using prediction based on experimental expression of the five involved proteins and relationships between molecules, IPA predicted an activation of the downstream production of NAD+ and ATP required for energy production for processes such as mitosis (Figure 6C).

Finally, we investigated the contribution of glycolysis to the metabolism of EPs. We observed that all proteins involved in glycolysis are expressed at the progenitor stages, with expression levels not changing in large part from EP1 to EP4 (Figure 6D). However, the expression of PKLR, encoding the red cell form of pyruvate kinase and essential for glycolysis, progressively increased from EP1 to EP4, suggesting that glycolytic activities in EPs increase as they differentiate towards EP4 (Figure 6E). Differential analyses revealed that PKLR is the most significantly upregulated protein from EP1 to EP2 (Figure 6F). Taken together, these data suggest that the proteins necessary for both oxidative phosphorylation and glycolysis are expressed in erythroid progenitor cells.

Discussion

In vitro culture systems have been the mainstay for the study of human erythropoiesis due, in part, to their capacity to produce large numbers of erythroid precursors in response to cytokines. These culture systems have been developed and adapted over time to meet the needs of specific studies12,28–30. These studies made significant contributions to our understanding of human terminal erythropoiesis; however, they have not provided detailed knowledge of erythroid progenitors, which are heterogeneous and classically characterized by their ability to form colony in semi-solid media5. We recently resolved the heterogeneity of erythroid progenitors into 4 distinct subpopulations from EP1 to EP4 based on the expression of specific surface markers. Using this classification, we demonstrated that sorted EPs have different requirements with regards to growth factors. We observed an inverse relationship in terms of proliferation and differentiation of EPs with regards to SCF. While EP1 proliferation is dependent on SCF, the subsequent stages are much less dependent, if EPO is maintained in the media. Further, SCF delayed differentiation of erythroid progenitors.

Based on our previous studies10–12, focused on the immunophenotyping of erythroid progenitors and the findings from the present study on the individual responses of EP1 to EP4 to growth factors, we propose that EP1 are BFU-Es while EP2-EP4 represent different maturational stages of CFU-Es.

Previous studies have contributed to the proteomic characterization of murine and human erythropoiesis20,31. However, these studies primarily focused on terminal erythroid differentiation. We provide the first in depth proteomics study of human erythroid progenitor cells and document specific changes, mainly in the cell cycle and metabolic pathways. This dataset provides a resource regarding protein expression in highly purified and characterized erythroid progenitors. Nevertheless, it should be noted that this analysis in not exhaustive due to limitations in the currently available proteomic platforms. This is reflected for example in our inability to quantitate EPO-R in our progenitor populations.

Cell cycle and its regulation are important determinants for erythropoiesis. Indeed, during terminal differentiation, each cell division generates two cells that are distinctly different from the mother cell. The situation is more complex in erythroid progenitors, which can self-renew and/or lead to a cell that is different. We demonstrate that only a few proteins involved in cell cycle regulation show a different patter of expression from EP1 to EP4. Among those, E2F4 expression progressively increased. E2F4 is conventionally reported as a repressor of the G1/S transition. However, a pro-proliferative role was recently shown for E2F4 in fast-cycling cells32, and a role for E2F4 was previously reported in murine erythropoiesis33,34. E2F4 was first described as a repressor E2F, silencing the expression of cell cycle genes in conjunction with the retinoblastoma (RB) protein family35. However, studies in specific, fast cycling cell types (such as erythroid cells), suggest that E2F4 can activate genes regulating proliferation of cells33,34,36. Of note, E2f4−/− erythroblasts proliferate less and result in macrocytic reticulocytes compared to their wild-type counterpart36. Future studies are warranted to explore the effects of knocking out E2F4 using CRISPR/Cas9 technologies.

Adding to this complexity, recent studies demonstrate that the length of the cell cycle is different depending on the erythroid developmental stage. Differentiation programs have evolved to interact with the cell cycle25. For example, in the human Megakaryocyte-Erythrocyte Progenitor (MEP), the length of the cell cycle promotes cell fate decision, with fast cycling promotes erythroid fate and slow cycling promotes megakaryocyte fate37–39. We observed that, in primary bone marrow, the percentage of cells in S phase increases from EP1 to EP4 along with an increase in cell-cycle speed. Together, these findings suggest that cells divide faster as they proceed towards terminal differentiation.

Our study has implications for the study of erythropoietic disorders such as Diamond Blackfan anemia (DBA). DBA is a rare inherited bone marrow failure syndrome caused in most cases by a mutation in a ribosomal protein and characterized by failure of erythropoiesis at the progenitor stages40. Patients with DBA present with macrocytic, reticulocytopenic anemia. Defects in the cell cycle have often been inferred in the etiology of DBA to explain the change in red cell volume. However, whether differentiating cells skip a division or alter the length of the cell cycle remains to be defined. Recent studies have established a link between cell cycle duration and regulation of red cell volume during terminal erythroid differentiation in the mouse23. We propose the use of combined immunophenotyping with cell cycle analyses in isolated bone marrow from patients with DBA could provide new insights into volume regulation.

With regards to the metabolic regulation of human EPs, our proteomics data revealed that proteins involved in both oxidative phosphorylation and glycolysis are expressed in EPs. Interestingly, we observed a progressive increase of the mitochondrial biomass from EP1 to EP3 along with increase in the mitochondrial potential. These results suggest that oxidative phosphorylation activities increase during EP differentiation, peaking at the EP3 stage. However, functional assays are needed to confirm this hypothesis.

Finally, several studies have also identified a link between heme biosynthesis, oxidative stress and DBA. In this context, it would be of interest to measure the expression levels of PKLR as well as those of other enzymes in erythroid progenitors from patients with DBA.

Nevertheless, one of the main limitations of our study pertains to both the cell cycle and metabolic aspects. We used quantitative mass spectrometry, allowing us for a robust quantitation of most of the different proteins expressed at the individual progenitors’ stages. However, most of the proteins involved in cell cycle regulation and metabolism need to be activated or inhibited to exert their function. Thus, more work is needed to understand how these different actors are functionally involved in the EPs’ regulation.

Despite these limitations, our data provide a comprehensive resource for the study of entire spectrum of human normal and disordered erythropoiesis.

Supplementary Material

SUP FIG 1

Supplemental Figure 1. Proteomic quality control. (A) Table of Pearson correlation of Log2(Copy number) of each sample of EP. (B) Boxplot showing expression of ribosomal complexes, known to have a 1:1 ratio for all their proteins. These data show copy number measurement precision.

SUP TAB 1

Supplemental Table 1. Concentrations used for the different cytokines. M denotes media.

SUP TAB 2

Supplemental Table 2. List of proteins identified by mass spectrometry.

SUP FIG 2

Supplemental Figure 2. Protein expression of globins, proteins involved in iron metabolism and heme synthesis.

Acknowledgements

This work was supported in part by NIH grants DK32094 (to PGG and NM) and HL144436 (to LB).

The glycolysis pathway was pictured using BioRender (BioRender.com). The Orbitrap Fusion mass spectrometer was acquired with funds from the FEDER through the « Operational Programme for Competitiveness Factors and employment 2007-2013 », and from the « Cancéropôle Ile-de-France ».

References

  • 1.Orkin SH, Zon LI. Hematopoiesis: an evolving paradigm for stem cell biology. Cell. 2008;132(4):631–644. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Koury MJ, Haase VH. Anaemia in kidney disease: harnessing hypoxia responses for therapy. Nat Rev Nephrol. 2015;11(7):394–410. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Blanc L, Vidal M. Reticulocyte membrane remodeling: contribution of the exosome pathway. Curr Opin Hematol. 2010;17(3):177–183. [DOI] [PubMed] [Google Scholar]
  • 4.Ney PA. Normal and disordered reticulocyte maturation. Curr Opin Hematol. 2011;18(3):152–157. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Gregory CJ, Eaves AC. Human marrow cells capable of erythropoietic differentiation in vitro: definition of three erythroid colony responses. Blood. 1977;49(6):855–864. [PubMed] [Google Scholar]
  • 6.Gregory CJ, Eaves AC. Three stages of erythropoietic progenitor cell differentiation distinguished by a number of physical and biologic properties. Blood. 1978;51(3):527–537. [PubMed] [Google Scholar]
  • 7.McLeod DL, Shreeve MM, Axelrad AA. Improved plasma culture system for production of erythrocytic colonies in vitro: quantitative assay method for CFU-E. Blood. 1974;44(4):517–534. [PubMed] [Google Scholar]
  • 8.Stephenson JR, Axelrad AA, McLeod DL, Shreeve MM. Induction of colonies of hemoglobin-synthesizing cells by erythropoietin in vitro. Proc Natl Acad Sci U S A. 1971;68(7):1542–1546. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Li J, Hale J, Bhagia P, et al. Isolation and transcriptome analyses of human erythroid progenitors: BFU-E and CFU-E. Blood. 2014;124(24):3636–3645. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Ashley RJ, Yan H, Wang N, et al. Steroid resistance in Diamond Blackfan anemia associates with p57Kip2 dysregulation in erythroid progenitors. J Clin Invest. 2020;130(4):2097–2110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Yan H, Hale J, Jaffray J, et al. Developmental Differences Between Neonatal and Adult Human Erythropoiesis. Am J Hematol. 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Yan H, Ali A, Blanc L, et al. Comprehensive phenotyping of erythropoiesis in human bone marrow: Evaluation of normal and ineffective erythropoiesis. Am J Hematol. 2021;96(9):1064–1076. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Udupa KB, Reissmann KR. In vivo erythropoietin requirements of regenerating erythroid progenitors (BFU-e, CFU-e) in bone marrow of mice. Blood. 1979;53(6):1164–1171. [PubMed] [Google Scholar]
  • 14.Dulmovits BM, Tang Y, Papoin J, et al. HMGB1-mediated restriction of EPO signaling contributes to anemia of inflammation. Blood. 2022;139(21):3181–3193. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Cox J, Hein MY, Luber CA, Paron I, Nagaraj N, Mann M. Accurate proteome-wide label-free quantification by delayed normalization and maximal peptide ratio extraction, termed MaxLFQ. Mol Cell Proteomics. 2014;13(9):2513–2526. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Cox J, Matic I, Hilger M, et al. A practical guide to the MaxQuant computational platform for SILAC-based quantitative proteomics. Nat Protoc. 2009;4(5):698–705. [DOI] [PubMed] [Google Scholar]
  • 17.Wisniewski JR, Hein MY, Cox J, Mann M. A “proteomic ruler” for protein copy number and concentration estimation without spike-in standards. Mol Cell Proteomics. 2014;13(12):3497–3506. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Muta K, Krantz SB, Bondurant MC, Dai CH. Stem cell factor retards differentiation of normal human erythroid progenitor cells while stimulating proliferation. Blood. 1995;86(2):572–580. [PubMed] [Google Scholar]
  • 19.Koury MJ, Bondurant MC. Erythropoietin retards DNA breakdown and prevents programmed death in erythroid progenitor cells. Science. 1990;248(4953):378–381. [DOI] [PubMed] [Google Scholar]
  • 20.Gautier EF, Ducamp S, Leduc M, et al. Comprehensive Proteomic Analysis of Human Erythropoiesis. Cell Rep. 2016;16(5):1470–1484. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Kuvardina ON, Herglotz J, Kolodziej S, et al. RUNX1 represses the erythroid gene expression program during megakaryocytic differentiation. Blood. 2015;125(23):3570–3579. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Gonzalez-Menendez P, Phadke I, Olive ME, et al. Arginine metabolism regulates human erythroid differentiation through hypusination of eIF5A. Blood. 2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Hidalgo D, Bejder J, Pop R, et al. EpoR stimulates rapid cycling and larger red cells during mouse and human erythropoiesis. Nat Commun. 2021;12(1):7334. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Hwang Y, Futran M, Hidalgo D, et al. Global increase in replication fork speed during a p57KIP2-regulated erythroid cell fate switch. Sci Adv. 2017;3(5):e1700298. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Hwang Y, Hidalgo D, Socolovsky M. The shifting shape and functional specializations of the cell cycle during lineage development. WIREs Mech Dis. 2021;13(2):e1504. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.van Wijk R, van Solinge WW. The energy-less red blood cell is lost: erythrocyte enzyme abnormalities of glycolysis. Blood. 2005;106(13):4034–4042. [DOI] [PubMed] [Google Scholar]
  • 27.Moras M, Lefevre SD, Ostuni MA. From Erythroblasts to Mature Red Blood Cells: Organelle Clearance in Mammals. Front Physiol. 2017;8:1076. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Dulmovits BM, Appiah-Kubi AO, Papoin J, et al. Pomalidomide reverses gamma-globin silencing through the transcriptional reprogramming of adult hematopoietic progenitors. Blood. 2016;127(11):1481–1492. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Fibach E, Manor D, Oppenheim A, Rachmilewitz EA. Proliferation and maturation of human erythroid progenitors in liquid culture. Blood. 1989;73(1):100–103. [PubMed] [Google Scholar]
  • 30.Giarratana MC, Rouard H, Dumont A, et al. Proof of principle for transfusion of in vitro-generated red blood cells. Blood. 2011;118(19):5071–5079. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Gautier EF, Leduc M, Ladli M, et al. Comprehensive proteomic analysis of murine terminal erythroid differentiation. Blood Adv. 2020;4(7):1464–1477. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Hsu J, Arand J, Chaikovsky A, et al. E2F4 regulates transcriptional activation in mouse embryonic stem cells independently of the RB family. Nat Commun. 2019;10(1):2939. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Humbert PO, Rogers C, Ganiatsas S, et al. E2F4 is essential for normal erythrocyte maturation and neonatal viability. Mol Cell. 2000;6(2):281–291. [DOI] [PubMed] [Google Scholar]
  • 34.Zhang J, Lee EY, Liu Y, Berman SD, Lodish HF, Lees JA. pRB and E2F4 play distinct cell-intrinsic roles in fetal erythropoiesis. Cell Cycle. 2010;9(2):371–376. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Hsu J, Sage J. Novel functions for the transcription factor E2F4 in development and disease. Cell Cycle. 2016;15(23):3183–3190. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Kinross KM, Clark AJ, Iazzolino RM, Humbert PO. E2f4 regulates fetal erythropoiesis through the promotion of cellular proliferation. Blood. 2006;108(3):886–895. [DOI] [PubMed] [Google Scholar]
  • 37.Kwon N, Thompson EN, Mayday MY, Scanlon V, Lu YC, Krause DS. Current understanding of human megakaryocytic-erythroid progenitors and their fate determinants. Curr Opin Hematol. 2021;28(1):28–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Lu YC, Sanada C, Xavier-Ferrucio J, et al. The Molecular Signature of Megakaryocyte-Erythroid Progenitors Reveals a Role for the Cell Cycle in Fate Specification. Cell Rep. 2018;25(8):2083–2093 e2084. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Sanada C, Xavier-Ferrucio J, Lu YC, et al. Adult human megakaryocyte-erythroid progenitors are in the CD34+CD38mid fraction. Blood. 2016;128(7):923–933. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Da Costa L, Leblanc T, Mohandas N. Diamond-Blackfan anemia. Blood. 2020;136(11):1262–1273. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

SUP FIG 1

Supplemental Figure 1. Proteomic quality control. (A) Table of Pearson correlation of Log2(Copy number) of each sample of EP. (B) Boxplot showing expression of ribosomal complexes, known to have a 1:1 ratio for all their proteins. These data show copy number measurement precision.

NIHMS1940525-supplement-SUP_FIG_1.tif (1,013.9KB, tif)
SUP TAB 1

Supplemental Table 1. Concentrations used for the different cytokines. M denotes media.

NIHMS1940525-supplement-SUP_TAB_1.docx (26.3KB, docx)
SUP TAB 2

Supplemental Table 2. List of proteins identified by mass spectrometry.

NIHMS1940525-supplement-SUP_TAB_2.xlsx (1.8MB, xlsx)
SUP FIG 2

Supplemental Figure 2. Protein expression of globins, proteins involved in iron metabolism and heme synthesis.

NIHMS1940525-supplement-SUP_FIG_2.tif (1.4MB, tif)

RESOURCES