Comprehensive transcriptome analysis of erythroid differentiation potential of olive leaf in haematopoietic stem cells

Abstract Anaemia is one of the leading causes of disability in young adults and is associated with increased morbidity and mortality in elderly. With a global target to reduce the disease burden of anaemia, recent researches focus on novel compounds with the ability to induce erythropoiesis and regulate iron homeostasis. We aimed to explore the biological events and potential polypharmacological effects of water‐extracted olive leaf (WOL) on human bone marrow–derived haematopoietic stem cells (hHSCs) using a comprehensive gene expression analysis. HPLC analysis identifies six bioactive polyphenols in the WOL. Treatment with WOL for 12 days regulated gene expressions related to erythroid differentiation, oxygen homeostasis, iron homeostasis, haem metabolism and Hb biosynthesis in hHSCs. Functional clustering analysis reveals several major functions of WOL such as ribosomal biogenesis and mitochondrial translation machinery, glycolytic process, ATP biosynthesis and immune response. Additionally, the colonies of both primitive and mature erythroid progenitors, CFU‐E and BFU‐E, were significantly increased in WOL‐treated hHSCs. The expressions of erythroid markers, CD47, glycophorin A (GYPA), and transferrin receptor (TFRC) and adult Hb subunits‐HBA and HBB were also confirmed in immunofluorescent staining and flow cytometer analysis in WOL‐treated hHSCs. It is well known that induction of lineage‐specific differentiation, as well as the maturation of early haematopoietic precursors into fully mature erythrocytes, involves multiple simultaneous biological events and complex signalling networks. In this regard, our genome‐wide transcriptome profiling with microarray study on WOL‐treated hHSCs provides general insights into the multitarget prophylactic and/or therapeutic potential of WOL in anaemia and other haematological disorders.


| INTRODUC TI ON
Anaemia is a global public health problem affecting one-third population worldwide. According to the World Health Organization (WHO), children under 5 years old, pregnant women and people older than 60 years are particularly at risk of developing anaemia with a worldwide prevalence of 47•4%, 41•8% and 23•9%, respectively. Iron deficiency is estimated to cause half of all causes of anaemia. Other causes include physiological blood loss in women, acute and chronic infections, micronutrient deficiencies other than iron, and haemoglobinopathies. In adolescents and young adults (15-59 years of age), iron-deficiency anaemia is one of the leading causes of disability, causing extreme fatigue, decreased immunity, reduced work capacity, dizziness, headaches, hair loss, hearing deficit, anxiety and depression. In elderly, presence of anaemia is associated with worse prognosis of any disease, increased morbidity and mortality, decline in cognitive function, dementia and increased fracture risk. Current available treatment options for anaemia include iron or other nutrient supplementation, nutritional interventions in the form of food fortification, blood transfusion, erythropoietic agent infusion and anti-inflammatory therapies. [1][2][3][4][5] However, conventional treatments are not without adverse effects. Iron supplementation can cause abdominal cramping, nausea, constipation and dark stools.
Blood transfusion may cause allergic, immune and febrile reactions and may increase the risk of acquiring blood-borne diseases. Food fortification is not an accessible option for low-and middle-income countries. On the other hand, there is a growing demand for innovative plant-based nutritional alternatives for vegans and vegetarians who are more susceptible to anaemia. 6,7 In this context, medicinal plants with bioactive components have been receiving considerable attention from researchers as natural hematinic agents complemented with or alternatives to conventional treatment. 8 Haematopoietic stem cells (HSCs) are the primary multipotent stem cells that can self-renew as well as can further differentiate into a hierarchy of committed progenitors that ultimately give rise to mature blood cells through specific regulation of signal molecules. 9 Multicomponent plant extracts may target multiple biological events to alter and/or control cell lineage differentiation of HSCs. 10 Understanding the molecular functions and biological processes triggered by plant extracts will provide more general insights into their effects on stem cell regulation as well as facilitate their therapeutic use targeting HSCs.
In this regard, olive leaf extract (OLE) contains a wide variety of bioactive polyphenols that work synergistically and exert polypharmacological effects. 11 We have previously reported that OLE induces monocyte/macrophage differentiation 12 whereas its components apigetrin 13 and apigenin 14 induce erythroid differentiation in human chronic myelogenous leukaemia cell line K562. Additionally, phenolic compounds of OLE luteolin-7-glucoside and apigenin-7glucoside could direct HSC differentiation towards erythroid lineage. 15 However, the effects of multicomponent OLE on HSCs have not been evaluated yet.
In the present study, we have performed a comprehensive gene expression analysis of the regulatory networks in WOL-treated human HSCs (hHSCs) to develop a new understanding of WOL's mechanism of action. We have also evaluated the effect of WOL treatment on the proliferation and differentiation pattern of hHSCs using colony forming cell (CFC) assay. Additionally, protein expression of several erythroid markers and haemoglobin (Hb) subunits was investigated. Our study presents a detailed analysis of biological events in hHSCs induced by WOL and highlights its multilevel exploitable targets for anaemia therapy.

| MATERIAL S AND ME THODS
Detailed methodology is described in Appendix S1.

| Cell culture and sample treatment
The human chronic leukaemia cell line K562 was obtained from the Riken Cell Bank (RCB0027). Human bone marrow CD34+ progenitor cells were purchased from Lonza Inc (Lonza).

| RNA extraction
Isogen reagent (311-02501, Nippon Gene) was used to extract total RNA from K562 cells after treatment for 6 days and from hHSCs on day 9 (D0 before treatment) and day 21 (D12 after treatment), following the manufacturer's instructions. Concentrations of total RNA were quantified with NanoDrop 2000 spectrophotometer (Thermo Scientific).

| DNA microarray analysis
DNA microarray was performed on GeneAtlas™ System using GeneChip™ 3′ IVT PLUS Reagent Kit and GeneAtlas™ Hybridization, anaemia, DNA microarray, erythropoiesis, haematopoietic stem cell, iron homeostasis, olive leaf Wash, and Stain Kit for 3′ IVT Arrays (Applied Biosystems, Thermo Fisher Scientific Inc), following the manufacturer's instructions. Gene annotation and pathway analysis were conducted using an online data mining tool DAVID ver. 6.8 and the Molecular Signatures Database (MSigDB) ver. 7.1 of the Gene Set Enrichment Analysis (GSEA) software. 16 Heat maps were generated using visualization software Morpheus (https://softw are.broad insti tute.org/morpheus).

| Statistical analysis
Statistical analyses were performed with GraphPad Prism version 8.0 (GraphPad Software, Inc). Data were represented as the mean ± standard error of the mean (SEM). An unpaired two-tailed Student's t test was used to compare between two groups. A oneway ANOVA followed by Dunnett's post hoc test was performed to compare the treatment groups to a control group. A P-value <.05 was considered as significant.

| Bioactive compounds in WOL
To identify and quantify the components in WOL, HPLC analysis was carried out.

| Confirmation of differentiation-inducing effects of WOL-treated K562 cells
Firstly, we confirmed the differentiation-inducing potential of the WOL in K562 cells following our previous studies. 12,14 K562 is a leukaemic cell line but retains similar self-renewal and pluripotent properties of HSCs and is commonly used as a cell model for studies of HSC differentiation. 17 To evaluate the effect of WOL on K562 cell proliferation, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazo lium bromide (MTT) assay was performed. Figure S1A shows that cell proliferation was decreased to 66.1% at the concentration of 120 μg/mL WOL treatment and to 29.0% at 150 μg/mL in a dosedependent manner, compared to the untreated cells. To distinguish the cell death from the growth inhibition by WOL treatment, viable cells were counted using the Guava ViaCount flow cytometer. The cell viability was decreased to 71.0% at the concentration of 120 μg/ mL compared to the untreated cells. Excessive cell death was observed at the concentration of 150 μg/mL ( Figure S1B).  Figure 1A shows morphological changes in WOL-treated hHSCs.

| WOL induces erythroid marker GYPA expression in hHSC
The number and size of D12 control cells were increased compared to those of D0 control cells. On the other, the number of cells decreased, and the size of cells shrank by WOL treatment for 12 days.
Next, we examined the effect of WOL in the expression of glycophorin A (GYPA), the main marker of erythroid differentiation, to determine the optimum duration of WOL treatment for evaluation of biological events underlying its differentiation-inducing effects. Figure 1B

| WOL treatment significantly increased the colonies of primitive and mature erythroid progenitors
We performed CFC assay to evaluate the effect of WOL treatment on the proliferation and differentiation pattern of hHSCs. We found that WOL treatment significantly increased the numbers of both

| Differentially expressed genes (DEGs) in WOLtreated hHSCs
Next, we performed microarray analysis of WOL-treated and untreated control hHSCs to explore the biological events that took place during the differentiation period. We prepared three technical replications of each treatment groups-day 0 control (D0 control),

| Hallmark gene sets between D12 WOL vs D12 control hHSCs
We investigated significantly enriched hallmark gene sets between D12 WOL and D12 control hHSCs using the MSigDB of GSEA software. Hallmark gene sets represent specific well-defined biological processes (BPs) and reduce noise and redundancy. There is an initial collection of 50 hallmarks, which collect and condense information from over 4,000 original overlapping gene sets. 18 Figure 3 shows the

| Gene functional clustering of BP ontologies between D12 WOL vs D12 control hHSCs
We performed GO cluster analysis to classify highly related genes into functionally related groups using the Gene Functional Classification Tool of DAVID. A total of 29 clusters were identified, setting the classification stringency to medium (default), kappa similarity threshold = 0.35, and multiple linkage threshold to 50%.

| Significantly enriched canonical pathways between D12 WOL and D12 control hHSCs
To investigate the biological functions of WOL in a network context, we analysed the DEGs between D12 WOL and D12 control cells using the Pathway Viewer feature of the DAVID Functional Annotation Tool. Figure  significantly regulated in WOL-treated hHSCs ( Figure 4B).

| Gene expressions related to erythroid differentiation in WOL-treated hHSCs
We found that some specific GOs of erythroid differentiation were significantly regulated by the DEGs between D12 WOL vs D12 control  Table 2.
Next, we investigated individual gene expression profile in D0 and D12 controls and D12 WOL-treated cells. Heatmaps in Figure 5B   Next, we investigated the expression profile of stage-specific iron homeostasis-related genes ( Figure 6B). We found that genes

| D ISCUSS I ON
In the present study, we have documented a detailed gene expression profiling of erythropoiesis-inducing effects of OLE in hHSCs.
WOL at a concentration of 40 µg/mL, in the absence of any growth factor, induced erythroid differentiation in hHSCs. Similarly, WOL induced expressions of major erythroid markers and all Hb subunits in a human erythroleukaemic cell line K562 (Figure 9).

F I G U R E 6
Effect of WOL on expressions of genes involved in oxygen homeostasis/response to hypoxia or oxidative stress, iron homeostasis and haemoglobin synthesis in hHSCs. Heatmaps by microarray showing relative expression intensity of genes involved in (A) oxygen homeostasis/response to hypoxia or oxidative stress, (B) iron homeostasis and (C) haem metabolism in hHSCs of D0 control, D12 control and D12 WOL. Expressions of genes involved in (D) hypoxia-induced erythropoiesis and iron homeostasis, and (E) haemoglobin subunits in WOL-treated hHSCs were examined by real-time PCR. The mRNA expressions were normalized to HPRT1 as internal control and represented as the mean ± SE for n = 4 in a 96-well PCR plate. Statistically significant difference from the D12 control group (0 µg/mL WOL) at † P < .1, * P < .05 and ** P < .01 by Dunnett's multiple comparison test ficiency, which affects their differentiation ability. 22,23 Converging evidence establishes that ribosomal biogenesis, which is highly regulated in stem cells to maintain stem cell homeostasis, is also a crucial and constitutive molecular machinery that initiates gene-specific F I G U R E 7 Immunofluorescence staining for erythroid markers and haemoglobin subunits in WOL-treated hHSCs. The hHSCs were treated with 40 µg/mL WOL for 12 days and then stained using primary antibodies for GYPA, TFRC, CD47, HBA, HBB and HBG. Nuclei in the cells were stained using 1 µg/mL Hoechst 33342. Images were collected with an Olympus IX83 inverted microscope at a magnification of 400×. Scale bars represent 20 μm F I G U R E 8 Flow cytometric analysis for erythroid markers and haemoglobin subunits in WOL-treated hHSCs. The hHSCs were treated with 40 µg/mL WOL for 12 d and then stained using primary antibodies for GYPA, TFRC, CD47, HBA, HBB and HBG. Mean fluorescence intensity (MFI) of the stained cells in histogram were evaluated from acquired 5000 events using a Guava easyCyte™ 8HT Flow Cytometer. The data were represented as the mean ± SE for n = 4. ** P < .01 by unpaired two-tailed Student's t test to another major erythroid transcription factor GATA1 and can act synergistically. [35][36][37] We found that both KLF1 and GATA1 were significantly up-regulated in WOL-treated cells compared to controls.
Additionally, our protein expression data showed consistent results. Furthermore, we found that CASP3 was significantly increased in WOL-treated hHSCs. CASP3, a cysteine protease, plays essential roles in apoptosis but has also been implicated in non-apoptotic processes like cell differentiation and signalling. It is also involved in erythroid maturation through mitophagy and depletion of enucleation of erythroid cells. [44][45][46] CASP3 can arrest erythropoiesis via GATA1 cleavage, but heat shock protein 70 (HSP70) prevents this CASP3-mediated cleavage. 47 Our findings suggest that WOL treatment may modulate erythroid maturation through CASP3-mediated mitophagy and enucleation of erythroid cells and subsequently may decrease erythrocyte proliferation and promote cell death of mature cells.
Another important finding of the present study was the expression of HAMP, the gene encoding hepcidin, in WOL-treated hHSCs.
HAMP is an iron homeostasis regulator that limits iron export by binding to the iron exporter ferroportin. Disruption of hepcidin level is associated with haematological and renal disorders such as high hepcidin level is associated with chronic kidney disease, whereas decreased hepcidin level is found in iron-deficiency anaemia. 48 and SLC25A28, were up-regulated in WOL-treated hHSCs compared to untreated control. 54,55 Additionally, iron-sulphur cluster assembly nih.gov/geo/query/ acc.cgi?acc=GSE14 8775).