Evaluation of Reliable Reference Genes for In Vitro Erythrocyte Generation from Cord Blood CD34 + Cells

Abstract

In vitro generation of red blood cells has the potential to circumvent shortfalls in the global demand for blood for transfusion applications. However, cell differentiation and proliferation are often regulated by precise changes in gene expression, but the underlying mechanisms and molecular changes remain unclear. Quantitative reverse transcription–polymerase chain reaction (qRT-PCR) can be used to evaluate multiple target genes. To make the results more reliable, suitable reference genes should be used to calibrate the error associated with qRT-PCR. In this study, we utilized bioinformatics to screen 3 novel candidate reference genes (calcium and integrin binding family member 2 [CIB2], olfactory receptor family 8 subfamily B member 8 [OR8B8], and zinc finger protein 425 [ZNF425]) along with eight traditional reference genes (glyceraldehyde-3-phosphate dehydrogenase [GAPDH], β-actin [ACTB], 18S RNA, β2-microglobulin [β2-MG], peptidylprolyl isomerase A [PPIA], TATA box-binding protein [TBP], hydroxymethylbilane synthase [HMBS], and hypoxanthine phosphoribosyltransferase 1 [HPRT1]). Two software algorithms (geNorm and NormFinder) were used to evaluate the stability of expression of the 11 genes at different stages of erythrocyte development. Comprehensive analysis showed that expression of GAPDH and TBP was the most stable, whereas ZNF425 and OR8B8 were the least suitable candidate genes. These results suggest that appropriate reference genes should be selected before performing gene expression analysis during erythroid differentiation and that GAPDH and TBP are suitable reference genes for gene expression studies on erythropoiesis.

Introduction

The transfusion of red blood cells (RBCs) has become an indispensable therapeutic intervention for patients experiencing traumatic injury, undergoing surgical procedures, or suffering from chronic conditions such as chronic anemia (Lee et al., 2018; Zamani et al., 2021). However, the shortage of blood for transfusions is a global problem. Research on stem cells, specifically hematopoietic stem cells (HSCs), holds promise for large-scale generation of mature RBCs through induction of their expansion and differentiation (Rallapalli et al., 2019; Liu et al., 2021; Zamani et al., 2021).

Cord blood (CB) is an excellent source of HSCs and often shows more efficiency and plasticity in multilineage differentiation than adult bone marrow or peripheral blood (Lee et al., 2018; Zamani et al., 2021). Differentiation of stem cells involves various morphological variations as well as changes in the expression levels of several genes (Vossaert et al., 2013; Ayanoglu et al., 2020). Hence, it is important to select a reference gene that is suitable for all cell types during erythroid differentiation.

Quantitative reverse transcription–polymerase chain reaction (qRT-PCR) is a highly sensitive method for measuring alterations in gene expression (Balaji and Vanniarajan, 2020; Ho and Patrizi, 2021). Normalizing the expression level of target genes against that of a reference gene is essential to obtain accurate results (Panina et al., 2020). However, the use of commonly used reference genes such as glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and β-actin (ACTB) has not been confirmed in different cell types or experimental conditions (Radonic et al., 2004; Li et al., 2015; Ayanoglu et al., 2020). Therefore, stable reference genes should be performed based on the cell type and differentiation protocol.

In the present gene expression stability analysis, we selected novel reference genes in CD34⁺ HSCs based on the Gene Expression Omnibus (GEO) microarray dataset (GSE66260). geNorm and NormFinder were used to analyze the expression stability of eight traditional candidate reference genes (GAPDH, ACTB, 18S RNA, β2-microglobulin [β2-MG], peptidylprolyl isomerase A [PPIA], TATA box-binding protein [TBP], hydroxymethylbilane synthase [HMBS], and hypoxanthine phosphoribosyltransferase 1 [HPRT1]) and three novel candidate reference genes (calcium and integrin binding family member 2 [CIB2], olfactory receptor family 8 subfamily B member 8 [OR8B8], and zinc finger protein 425 [ZNF425]) during erythroid differentiation using qRT-PCR.

Materials and Methods

CD34⁺ cell isolation and culture and erythropoietic differentiation

Human CB samples were collected from healthy pregnant women after delivery (n = 3). This study was approved by the Institutional Review Board of the Beijing Yuhe Chinese and Western Medicine Integrative Rehabilitation Hospital (ZXYEC-KT-2017-04-P01). To obtain CB CD34⁺ cells, we first separated low-density mononuclear cells (MNCs) from CB by Ficoll-Hypaque (1.077 g/L; TBD Science, Tianjin, China) density gradient centrifugation and then isolated CD34⁺ cells from MNCs through positive selection using a CD34 MicroBead Kit (Miltenyi Biotec, Bergisch Gladbach, Germany) according to the manufacturer's instructions. More than 90% of the isolated cells were CD34⁺, as confirmed by flow cytometry (FCM).

In vitro generation of erythrocytes comprised three steps. In the first step (days 0–7), CB CD34⁺ cells were seeded at a density of 2 × 10⁵/mL in expansion medium comprising a serum-free expansion medium (SFEM; StemCell Technologies, Vancouver, Canada) supplemented with 100 ng/mL human Flt-3 ligand (Flt3L; PeproTech, Rocky Hill, NJ), 100 ng/mL human stem cell factor (SCF; PeproTech), 20 ng/mL interleukin (IL)-3 (PeproTech), 10 ng/mL human thrombopoietin (PeproTech), 10 ng/mL IL-6 (PeproTech), and 1 μM StemRegenin 1 (StemCell Technologies). In the second step (days 7–19), the expansion medium was replaced with differentiation medium (DM; SFEM II [StemCell Technologies] containing 100 ng/mL SCF, 40 ng/mL IGF-1, 10 U/mL erythropoietin [PeproTech], 200 μg/mL transferrin [Sigma, St. Louis, MO], 1 μM dexamethasone [Dex; Sigma], 2 mM GlutaMAX [Gibco, Grand Island, NY], and 40 μg/mL lipid [Gibco]), which was changed every 3 days. In the third step (days 19–21), the cells were cultured in DM without Dex for improved maturation. The cultures were maintained at 37°C in a humidified atmosphere containing 5% CO₂.

Flow cytometry

The cultured cells were harvested on days 0, 7, 10, 14, 19, and 21 and their immunophenotypes were determined as follows. The cells were washed and resuspended in phosphate buffer saline supplemented with 1% bovine serum albumin. Antibodies against CD34-APC (BD Bioscience, San Diego, CA) or CD235a-PE (BD Bioscience) and CD71-APC (BD Bioscience) with the appropriate isotype controls were used according to the manufacturer's instructions. The gating strategy was based on the isotype controls, and analyses were performed using FlowJo software (Version 10; TreeStar, Ashland, OR).

Assessment of cell morphology

A total of 5 × 10⁴ sorted cells on slides were spun for 3 min at 1000 rpm (CYTOPRO 7622; Wescor, Logan, UT) and air-dried before being subjected to Wright–Giemsa staining (BASO, Zhuhai, Guangdong, China) according to the manufacturer's instructions. Images of the stained cells were captured using a digital camera (Nikon Corporation, Tokyo, Japan) at 200 × magnification.

RNA samples and cDNA synthesis

Total RNA was extracted using the TRIzol reagent (Life Technologies, Carlsbad, CA). The concentration and purity of the RNA samples were determined using a NanoDrop spectrophotometer (Thermo Scientific, Waltham, MA). cDNA was synthesized from ∼0.8 mg of total RNA using the ReverTra Ace qPCR RT Master Mix with gDNA Remover (Toyobo, Osaka, Japan), following the manufacturer's instructions.

Candidate genes and primers for qRT-PCR

Novel candidate reference genes were selected from a published erythropoiesis transcriptome array dataset (GEO accession number: GSE66260) (Merryweather-Clarke et al., 2016). Briefly, GEO2R analysis was performed using data from the Bioconductor project to determine the log fold change (logFC) in gene expression during erythropoiesis (P1–P3). Coding genes with logFC > −0.025 and <0.025 in each of these differentiation stages were selected as putative novel reference genes. Another eight candidate reference genes (whose primers were already available) commonly used in qRT-PCR analyses in HSC research were selected as traditional reference genes.

Real-time quantitative reverse transcription-PCR

qRT-PCR was performed using the THUNDERBIRD SYBR qPCR Mix (Toyobo) on Bio-Rad CFX Connect™ (Bio-Rad, Hercules, CA). Briefly, a 20-μL reaction mixture was subjected to qPCR under the following conditions: an initial denaturation at 95°C for 3 min, 40 cycles of denaturation at 94°C for 20 s, annealing at 58°C for 20 s, and extension at 72°C for 30 s. After amplification, a melting curve analysis was performed. The primers used for detection are listed in Supplementary Table S1.

Analysis of the expression stability of candidate genes

geNorm software written in Visual Basic for Applications in Microsoft Excel was used to analyze gene expression stability (Vandesompele et al., 2002). For each reference gene, we determined the pairwise variation with all the other reference genes as the standard deviation of logarithmically transformed expression ratios and defined the internal reference gene stability measure M as the average pairwise variation of a particular reference gene with all the other reference genes. An M value of less than 0.15 indicated that the number of reference genes met the requirements and that no housekeeping gene was needed. Genes with the lowest M values had the most stable expression among different culture environments or cell types. NormFinder software can be used to directly evaluate the stability of each tested reference gene according to intragroup and intergroup variances. Generally, the smaller the stability value, the better the expression stability of the corresponding gene (Wang et al., 2020). Finally, the most stable reference gene was selected according to the ranking of the candidate reference genes in geNorm and NormFinder. RefFinder, a comprehensive web-based tool, was used to determine the most stable reference gene for the final ranking (Xie et al., 2012).

Results

Flow cytometric analysis of erythrocytes at different developmental stages

The cultured cells expressed CD71 on day 7 and gained CD235a expression as they continued to mature. Finally, CD71 expression was gradually lost (Fig. 1). Alterations in marker expression indicate the gradual establishment of cells committed to the erythroid lineage and the maturation of these erythroid cells.

FIG. 1.

Flow cytometric analyses of in vitro-generated red blood cells from CB CD34⁺ cells. Cells from the indicated days were double-labeled with APC-conjugated anti-CD71 mAb and PE-conjugated anti-CD235a mAb. The percentages of cultured cells expressing CD71 and/or CD235a on days 0, 7, 10, 14, 19, and 21 are given in the representative dot plots. Marker expression indicates the gradual establishment of cells committed to the erythroid lineage. CB, cord blood; mAb, monoclonal antibody.

To accurately evaluate erythrocytes at different stages, CD34⁺, CD71⁺CD235a⁻, CD71⁺CD235a⁺, and CD71⁻CD235a⁺ cells were sorted (Fig. 2A, B). Following erythropoietic differentiation, the cell size, nuclear condensation, and nuclear extrusion were detected by Wright–Giemsa staining. The morphological features of the CD34⁺ population predominantly corresponded to those of HSCs (Fig. 2C). Proerythroblasts and early basophilic erythroblasts composed the CD71⁺CD235a⁻ population (P1); basophilic, polychromatic, and orthochromatic erythroblasts composed the CD71⁺CD235a⁺ population (P2); and late orthochromatic erythroblasts and reticulocytes composed the CD71⁻ CD235a⁺ population (P3) (Wojda et al., 2002).

FIG. 2.

Fluorescence-activated cell sorting strategy and cell morphology in different developmental stages. (A) Cells isolated by magnetic-activated cell sorting were stained with APC-conjugated anti-CD34 mAb. (B) P1 to P3 erythrocytes were defined by characteristic antibody staining patterns comprising CD71⁺CD235a⁻, CD71⁺CD235a⁺, and CD71⁻CD235a⁺ cells, respectively. (C) Wright–Giemsa staining of sorted CD34⁺, CD71⁺CD235a⁻, CD71⁺CD235a⁺, and CD71⁻CD235a⁺ cells (magnification × 200, bar = 50 μm).

Selection of novel candidate reference genes from published microarray data

A gene expression profiling study (GSE66260) using the GEO database (http://www.ncbi.nlm.nih.gov/geo/) was performed to compare the different stages of erythropoiesis in CB CD34⁺ cells. Magnetic beads were used to isolate CD34⁺ progenitors, and erythroblasts (CD71⁺CD235a⁻, CD71⁺CD235a⁺, and CD71⁻CD235a⁺) were sorted by fluorescence-activated cell sorting (FACS). We hypothesized that any gene stably expressed in this array could be used as a candidate reference gene.

To identify these genes, GEO2R, an online interactive web tool for gene expression analysis, was used to process this dataset (Davis and Meltzer, 2007). First, gene expression analysis was performed to compare CD34⁺ cells and CD71⁺CD235a⁻ cells. We selected genes with a logFC between −0.025 and 0.025, which reflects stable expression of genes between the two cell populations. We then performed a similar analysis with CD34⁺ cells and CD71⁺CD235a⁺ cells and CD34⁺ cells and CD71⁻CD235a⁺ cells. Of note, 4057, 3842, and 2097 genes showed a stable expression in CD34⁺/CD71⁺CD235a⁻, CD34⁺/CD71⁺CD235a⁺, and CD34⁺/CD71⁻CD235a⁺ cells, respectively. The overlap resulted in a list of only 19 genes that were stably expressed among all the tested stages (Fig. 3); of these, only three were coding genes and were thus selected as the final set of novel reference genes (CIB2, OR8B8, and ZNF425).

FIG. 3.

Identification of novel candidate reference genes. Venn diagram showing the overlap of genes with a logFC between −0.025 and 0.025 identified from CD71⁺CD235a⁻, CD71⁺CD235a⁺, and CD71⁻CD235a⁺ cells versus CD34⁺ cells. Only 19 genes were shared between all three comparisons. logFC, log fold change.

Identification of the specificity of primers

For each traditional and novel reference gene, primer specificity was determined by a single peak using the melting curve analysis, which indicated a single PCR product (Fig. 4).

FIG. 4.

Identification of the specificity of qRT-PCR amplicons. Dissociation curves with single peaks were generated from all amplicons (A–K). The specificity of qRT-PCR peaks showing the amplification of a single product without dimer formation for the 11 selected candidate reference genes is presented. ACTB, β-actin; CIB2, calcium and integrin binding family member 2; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; β2-MG, β2-microglobulin; HMBS, hydroxymethylbilane synthase; HPRT1, hypoxanthine phosphoribosyltransferase 1; OR8B8, olfactory receptor family 8 subfamily B member 8; PPIA, peptidylprolyl isomerase A; qRT-PCR, quantitative reverse transcription–polymerase chain reaction; TBP, TATA box-binding protein; ZNF425, zinc finger protein 425.

Variation in expression levels of the candidate reference genes

The value of cyclic quantification (Cq) is inversely proportional to the transcription level of a gene; that is, a higher Cq value corresponds to a lower expression level. Variation in the Cq value of the same gene in different cell types or conditions reflects the change in expression levels. In this study, the Cq values of the eight traditional reference genes and three novel reference genes changed during erythroid differentiation (Fig. 5).

FIG. 5.

Variation in the Cq values of 11 candidate reference genes in all the samples. The box diagram shows median values (horizontal lines inside the boxes); one-quarter (Q1, 25th percentile) and three-quarter (Q3, 75th percentile) values; and whisker caps (minimum and maximum Cq values). Cq, cyclic quantification.

Among the 11 candidate reference genes, ACTB and OR8B8 had the lowest and highest Cq values, at 18.79 and 31.44, respectively, indicating a wide range of expression levels of these genes. The Cq values of GAPDH, ACTB, 18S RNA, β2-MG, PPIA, TBP, HMBS, HPRT1, CIB2, OR8B8, and ZNF425 ranged from 17.58 to 22.87, 15.73 to 21.92, 18.62 to 24.09, 17.49 to 25.13, 16.62 to 22.27, 23.55 to 28.65, 21.55 to 27.39, 22.29 to 27.75, 28.31 to 35.12, 27.71 to 33.85, and 27.80 to 34.63, respectively (Supplementary Table S2). Among the 11 genes, GAPDH and TBP had the smallest range of variation and the highest abundance in expression; the novel candidate gene ZNF425 showed the most variable expression. Our preliminary results indicated that GAPDH, TBP, HPRT1, and PPIA expression was relatively stable.

geNorm analysis for candidate reference gene stability

For accurate gene expression measurements, the expression of an ideal reference gene should be stable within the experimental parameters. The underlying principle of the geNorm algorithm is that ideal reference genes show stable expression patterns independent of experimental parameters (van de Moosdijk and van Amerongen, 2016). Furthermore, geNorm can score not only each candidate reference gene but also expression changes in each gene compared with the average expression of candidate genes.

The M values of TBP and GAPDH were the lowest and those of OR8B8 and HMBS were the highest, indicating that the expression of TBP and GAPDH was the most stable, whereas that of OR8B8 and HMBS was the least stable (Fig. 6). Regarding the expression stability of genes in CD34⁺, CD71⁺CD235a⁻, CD71⁺CD235a⁺, and CD71⁻CD235a⁺ cells, the two most stable genes were TBP and GAPDH, HPRT1 and GAPDH, HPRT1 and TBP, and OR8B8 and β2-MG, respectively.

FIG. 6.

Average expression stability values (M) of the 11 candidate reference genes assessed based on the geNorm algorithm ranking. Graphs show candidate reference gene ranking according to the M value for gene expression stability analysis in CD34⁺ cells (A), CD71⁺CD235a⁻ cells (B), CD71⁺CD235a⁺ cells (C), CD71⁻CD235a⁺ cells (D), and all the samples (E). The most stable genes are on the right, and the least stable genes are on the left.

NormFinder analysis of candidate reference gene stability

Generally, the more stable the expression of the candidate reference gene, the smaller the stability value. In all samples, TBP and GAPDH had the lowest stability values, at 0.626 and 0.638, respectively, indicating that their expression was the most stable (Table 1). OR8B8 had the highest stability value of 2.486, indicating that it had the most variable expression. These results concurred with those of the geNorm analysis.

Table 1.

Stability of Candidate Reference Genes Ranked by NormFinder

Ranking order	CD34⁺		CD71⁺CD235a⁻		CD71⁺CD235a⁺		CD71 ⁻ CD235a⁺		Total
Ranking order	Gene	Stability	Gene	Stability	Gene	Stability	Gene	Stability	Gene	Stability
1	TBP	0.257	HPRT1	0.069	TBP	0.079	β2-MG	0.023	TBP	0.626
2	GAPDH	0.271	GAPDH	0.080	HPRT1	0.192	OR8B8	0.023	GAPDH	0.638
3	HPRT1	0.389	TBP	0.122	ACTB	0.279	TBP	0.062	HPRT1	0.825
4	β2-MG	0.412	ACTB	0.242	PPIA	0.338	ACTB	0.068	18S RNA	0.912
5	ACTB	0.437	PPIA	0.255	CIB2	0.425	HMBS	0.598	PPIA	0.914
6	18S RNA	0.456	18S RNA	0.338	GAPDH	0.710	18S RNA	0.702	ACTB	0.928
7	HMBS	0.466	HMBS	0.421	β2-MG	0.788	GAPDH	0.713	β2-MG	0.987
8	PPIA	0.579	OR8B8	0.487	ZNF425	1.123	CIB2	0.762	CIB2	1.530
9	CIB2	0.581	β2-MG	0.639	HMBS	1.370	HPRT1	0.970	HMBS	1.549
10	ZNF425	0.747	CIB2	0.685	18S RNA	1.812	PPIA	1.468	ZNF425	1.650
11	OR8B8	1.318	ZNF425	1.179	OR8B8	2.655	ZNF425	3.216	OR8B8	2.486

ACTB, β-actin; CIB2, calcium and integrin binding family member 2; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; β2-MG, β2-microglobulin; HMBS, hydroxymethylbilane synthase; HPRT1, hypoxanthine phosphoribosyltransferase 1; OR8B8, olfactory receptor family 8 subfamily B member 8; PPIA, peptidylprolyl isomerase A; TBP, TATA box-binding protein; ZNF425, zinc finger protein 425.

RefFinder analysis of candidate reference gene stability

For a more comprehensive overview of reference gene stability, the web-based tool, RefFinder (http://blooge.cn/RefFinder/) (Xie et al., 2012), was used to rank the candidate reference genes according to two additional algorithms: BestKeeper (Pfaffl et al., 2004) and DeltaCt (Silver et al., 2006). Across all stages, GAPDH and TBP were the most stable reference genes for normalization of target gene expression levels (Table 2).

Table 2.

Stability of Housekeeping Genes Ranked by RefFinder

Ranking order	CD34⁺	CD71⁺CD235⁻	CD71⁺CD235⁺	CD71 ⁻ CD235⁺	Total
1	TBP	HPRT1	HPRT1	OR8B8	GAPDH
2	β2-MG	GAPDH	TBP	β2-MG	TBP
3	ACTB	ACTB	PPIA	TBP	HPRT1
4	HPRT1	PPIA	ACTB	HMBS	PPIA
5	GAPDH	TBP	β2-MG	ACTB	HMBS
6	CIB2	18S RNA	CIB2	GAPDH	ACTB
7	PPIA	ZNF425	GAPDH	HPRT1	18S RNA
8	18S RNA	HMBS	HMBS	18S RNA	β2-MG
9	HMBS	OR8B8	ZNF425	CIB2	CIB2
10	ZNF425	CIB2	18S RNA	PPIA	OR8B8
11	OR8B8	β2-MG	OR8B8	ZNF425	ZNF425

Discussion

Umbilical CB-derived CD34⁺ hematopoietic stem/progenitor cells (CB-HSPC-derived RBCs) are a major cell type for the study of erythropoiesis, owing to their easy availability and strong self-renewal ability (Lee et al., 2018; Zamani et al., 2021). Therefore, it is of significance to utilize umbilical CB-derived CD34⁺ cells to study changes in gene expression during erythroid differentiation and the underlying molecular mechanisms. Numerous gene expression changes occur during erythroid maturation (Zhang et al., 2017; Royer et al., 2018). As the expression levels of candidate reference genes can vary with cell differentiation (Farrokhi et al., 2012; He et al., 2015), treatment methods (Yan et al., 2020), growth factors (Kumar et al., 2018), and organ development stage (Xia et al., 2021), the expression stability of the reference gene should be assessed before the target gene. However, to the best of our knowledge, there has been no study to identify suitable reference genes for erythroid differentiation.

Most studies on gene expression during erythroid differentiation directly use traditional reference genes, such as 18S RNA (Afreen et al., 2020) and GAPDH (Ma et al., 2010; Liu et al., 2021). However, the stability of these reference genes in different developmental stages has not been assessed, and this could have a serious effect on gene expression analysis. For instance, actin is downregulated during the terminal differentiation of erythroid cells because it plays an important role in cell shrinkage and enucleation (Chen et al., 2009). Particularly in reticulocytes, the RNA transcription level of ACTB is too low to be quantified accurately and therefore it is not appropriate for use as a reference gene (Silver et al., 2006). Hence, we selected eight traditional internal reference genes and evaluated their expression changes at different stages of cell differentiation.

To identify suitable reference genes, we also screened three new reference genes (CIB2, OR8B8, and ZNF425) using bioinformatic tools based on the analysis of the GEO gene expression profile of erythroid cells. By setting the fold change in expression within a certain range (−0.025 to 0.025) (van de Moosdijk and van Amerongen, 2016), genes with a relatively constant expression level in different cell development stages can be identified and used as new candidate reference genes. In this study, 19 genes were expressed stably, as determined by the analysis of cell expression profiles, at different developmental stages, of which 3 were coding and 16 were noncoding genes (Fig. 3). To screen out more novel reference genes, the range of fold change in expression should be set high to investigate more new candidate genes.

In this study, two commonly used gene stability evaluation tools, NormFinder and geNorm, were utilized to assess the expression stability of 11 candidate genes screened during RBC development. The most stably expressed reference genes varied among the differentiation stages, and the appropriate reference genes were selected by analyzing the entire differentiation process. During erythropoiesis, the NormFinder analysis indicated that TBP (0.626) and GAPDH (0.638) were the most stable, whereas OR8B8 (2.486) and ZNF425 (1.650) were the least stable reference genes, among the eight traditional genes and the three novel genes. Simultaneously, the geNorm analysis revealed that TBP and GAPDH were the most suitable reference genes. Despite the expression stability, determined using the database analysis, the three new reference genes did not show a stable expression level at different stages of erythroid differentiation, probably due to different induction protocols. Meanwhile, the expression levels of the novel candidate genes, CIB2, OR8B8, and ZNF425, were found to be relatively low (with Cq values of 28.31–35.12, 27.71–33.85, and 27.80–34.63, respectively); therefore, it would be inappropriate to compare them against a relatively highly expressed gene. An expanded candidate gene range might help to identify better expression references.

Furthermore, RefFinder was used to determine the best reference gene ranked by different evaluation methods (Chen et al., 2021). According to the results, GAPDH and TBP fulfilled most of the criteria for a suitable reference gene, that is, they were highly abundant and displayed minimal fluctuation. Therefore, they were the most suitable reference genes for analysis during erythrocyte generation from CB CD34⁺ cells.

Of note, these optimal reference genes are specific to our experimental conditions, and researchers should comprehensively investigate and screen the reference genes before gene expression studies.

Conclusions

In conclusion, taking advantage of qRT-PCR and algorithms, two stably expressed genes, GAPDH and TBP, were screened out as internal reference genes for the study of erythroid differentiation and maturation of CB CD34⁺ cells. These reliable reference genes can be used in the mechanism study of erythroid development and the application of CB-HSPC-derived RBCs. In the study of erythrocyte or other lineage differentiation, reference genes with a stable expression level should be screened for better analysis of target genes.

Footnotes

Author Contribution Statement

L.X. performed the experiments, collected data, and participated in original draft and figure preparation. Z.G. and Z.Y. performed the experiments and collected data. H.L. and Z.F. performed the experiments and analyzed the data. L.C., Y.L., and M.Q. provided training and consultation. X.X. participated in the structured outline preparation and article revision. W.Y. and C.L. participated in the article revision. X.P. participated in conception of the review and final revision of the article.

Disclosure Statement

No competing financial interests exist.

Funding Information

This work was supported by the National Key Research and Development Program of China (Nos. 2017YFA0103100, 2017YFA0103103, and 2017YFA0103104), the Beijing Natural Science Foundation (No. 7194303), and the Guangzhou Health Care and Cooperative Innovation Major Project (No. 201803040005).

Supplementary Material

Supplementary Table S1

Supplementary Table S2

References

Afreen

, Bohler

, Muller

, Demmerath

E.M.

, Weiss

J.M.

, Jutzi

J.S.

, et al. (2020). BCL-XL expression is essential for human erythropoiesis and engraftment of hematopoietic stem cells. Cell Death Dis 11, 8.

Ayanoglu

F.B.

, Elcin

A.E.

, and Elcin

Y.M.

(2020). Evaluation of the stability of standard reference genes of adipose-derived mesenchymal stem cells during in vitro proliferation and differentiation. Mol Biol Rep, 47, 2109–2122.

Balaji

, and Vanniarajan

(2020). Implication of pseudo reference genes in normalization of data from reverse transcription-quantitative PCR. Gene 757, 144948.

Chen

, Liu

, Heck

, Chasis

J.A.

, An

, and Mohandas

(2009). Resolving the distinct stages in erythroid differentiation based on dynamic changes in membrane protein expression during erythropoiesis. Proc Natl Acad Sci U S A, 106, 17413–17418.

Chen

M.D.

, Wang

, Li

Y.P.

, Zeng

M.J.

, Liu

J.T.

, Ye

X.R.

, et al. (2021). Reference gene selection for qRT-PCR analyses of luffa (Luffa cylindrica) plants under abiotic stress conditions. Sci Rep 11, 3161.

Davis

, and Meltzer

P.S.

(2007). GEOquery: a bridge between the Gene Expression Omnibus (GEO) and BioConductor. Bioinformatics, 23, 1846–1847.

Farrokhi

, Eslaminejad

M.B.

, Nazarian

, Moradmand

, Samadian

, and Akhlaghi

(2012). Appropriate reference gene selection for real-time PCR data normalization during rat mesenchymal stem cell differentiation. Cell Mol Biol, 58 Suppl, OL1660–OL1670.

Y.X.

, Zhang

, Yang

, Wang

, and Su

(2015). Selection of suitable reference genes for reverse transcription-quantitative polymerase chain reaction analysis of neuronal cells differentiated from bone mesenchymal stem cells. Mol Med Rep, 12, 2291–2300.

K.H.

, and Patrizi

(2021). Assessment of common housekeeping genes as reference for gene expression studies using RT-qPCR in mouse choroid plexus. Sci Rep 11, 3278.

10.

Kumar

V.E.

, Cherupanakkal

, Catherine

, Kadhiravan

, Parameswaran

, Rajendiran

, et al. (2018). Endogenous gene selection for relative quantification PCR and IL6 transcript levels in the PBMC's of severe and non-severe dengue cases. BMC Res Notes 11, 550.

11.

Lee

, Sivalingam

, Lim

Z.R.

, Chia

, Shi

L.G.

, Roberts

, et al. (2018). Review: in vitro generation of red blood cells for transfusion medicine: progress, prospects and challenges. Biotechnol Adv, 36, 2118–2128.

12.

, Yang

, Bai

, Yang

, Zhong

, and Wang

(2015). Identification of optimal reference genes for quantitative PCR studies on human mesenchymal stem cells. Mol Med Rep, 11, 1304–1311.

13.

Liu

, Wu

, Lancelot

, Deng

, Gao

, Roback

J.D.

, et al. (2021). BMI1 enables extensive expansion of functional erythroblasts from human peripheral blood mononuclear cells. Mol Ther, 29, 1918–1932.

14.

Y.N.

, Zhang

, Yu

H.C.

, and Zhang

J.W.

(2010). CTD small phosphatase like 2 (CTDSPL2) can increase epsilon- and gamma-globin gene expression in K562 cells and CD34+ cells derived from umbilical cord blood. BMC Cell Biol 11, 75.

15.

Merryweather-Clarke

A.T.

, Tipping

A.J.

, Lamikanra

A.A.

, Fa

, Abu-Jamous

, Tsang

H.P.

, et al. (2016). Distinct gene expression program dynamics during erythropoiesis from human induced pluripotent stem cells compared with adult and cord blood progenitors. BMC Genomics 17, 817.

16.

Panina

, Germond

, and Watanabe

T.M.

(2020). Analysis of the stability of 70 housekeeping genes during iPS reprogramming. Sci Rep 10, 21711.

17.

Pfaffl

M.W.

, Tichopad

, Prgomet

, and Neuvians

T.P.

(2004). Determination of stable housekeeping genes, differentially regulated target genes and sample integrity: BestKeeper—Excel-based tool using pair-wise correlations. Biotechnol Lett, 26, 509–515.

18.

Radonic

, Thulke

, Mackay

I.M.

, Landt

, Siegert

, and Nitsche

(2004). Guideline to reference gene selection for quantitative real-time PCR. Biochem Biophys Res Commun, 313, 856–862.

19.

Rallapalli

, Guhathakurta

, Narayan

, Bishi

D.K.

, Balasubramanian

, and Korrapati

P.S.

(2019). Generation of clinical-grade red blood cells from human umbilical cord blood mononuclear cells. Cell Tissue Res, 375, 437–449.

20.

Royer

, Begin

A.G.

, Plawinski

, Levesque

, Durrieu

M.C.

, and Laroche

(2018). Validation of reference genes for real-time PCR of cord blood mononuclear cells, differentiating endothelial progenitor cells, and mature endothelial cells. Exp Cell Res, 370, 389–398.

21.

Silver

, Best

, Jiang

, and Thein

S.L.

(2006). Selection of housekeeping genes for gene expression studies in human reticulocytes using real-time PCR. BMC Mol Biol 7, 33.

22.

van de Moosdijk

A.A.

, and van Amerongen

(2016). Identification of reliable reference genes for qRT-PCR studies of the developing mouse mammary gland. Sci Rep 6, 35595.

23.

Vandesompele

, De Preter

, Pattyn

, Poppe

, Van Roy

, De Paepe

, et al. (2002). Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 3, RESEARCH0034.

24.

Vossaert

, O'Leary

, Van Neste

, Heindryckx

, Vandesompele

, De Sutter

, et al. (2013). Reference loci for RT-qPCR analysis of differentiating human embryonic stem cells. BMC Mol Biol 14, 21.

25.

Wang

, Duan

, Chong

, Su

, Shan

, Yi

, et al. (2020). Systematic selection and validation of suitable reference genes for quantitative real-time PCR normalization studies of gene expression in Nitraria tangutorum. Sci Rep 10, 15891.

26.

Wojda

, Noel

, and Miller

J.L.

(2002). Fetal and adult hemoglobin production during adult erythropoiesis: coordinate expression correlates with cell proliferation. Blood, 99, 3005–3013.

27.

Xia

, Liu

, Chen

, and Wang

(2021). Selection and verification of the combination of reference genes for RT-qPCR analysis in rat adrenal gland development. J Steroid Biochem Mol Biol 208, 105821.

28.

Xie

, Xiao

, Chen

, Xu

, and Zhang

(2012). miRDeepFinder: a miRNA analysis tool for deep sequencing of plant small RNAs. Plant Mol Biol, 80, 75–84.

29.

Yan

, Xie

, Li

, Hu

, Tang

, and Shen

(2020). Identification and validation of reference genes selection in ovarian cancer exposed to hypoxia. Onco Targets Ther, 13, 7423–7431.

30.

Zamani

, Yaghoubi

, Naimi

, Hassanzadeh

, Pourakbari

, Aghebati-Maleki

, et al. (2021). Humanized culture medium for clinical-grade generation of erythroid cells from umbilical cord blood CD34(+) cells. Adv Pharm Bull, 11, 335–342.

31.

Zhang

, Wang

, Shen

, Guan

, Tian

, et al. (2017). Large-scale ex vivo generation of human red blood cells from cord blood CD34(+) cells. Stem Cells Transl Med, 6, 1698–1709.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB

Evaluation of Reliable Reference Genes for In Vitro Erythrocyte Generation from Cord Blood CD34 + Cells

Abstract

Introduction

Materials and Methods

CD34+ cell isolation and culture and erythropoietic differentiation

Flow cytometry

Assessment of cell morphology

RNA samples and cDNA synthesis

Candidate genes and primers for qRT-PCR

Real-time quantitative reverse transcription-PCR

Analysis of the expression stability of candidate genes

Results

Flow cytometric analysis of erythrocytes at different developmental stages

Selection of novel candidate reference genes from published microarray data

Identification of the specificity of primers

Variation in expression levels of the candidate reference genes

geNorm analysis for candidate reference gene stability

NormFinder analysis of candidate reference gene stability

RefFinder analysis of candidate reference gene stability

Discussion

Conclusions

Footnotes

Author Contribution Statement

Disclosure Statement

Funding Information

Supplementary Material

References

Supplementary Material

CD34⁺ cell isolation and culture and erythropoietic differentiation