The Ethics Committees of Hanyang University (HYG-11-019-1) and Soonchunhyang University Bucheon Hospital (SCHBC-2016-09-004) approved the protocols for the use of human CB and peripheral blood (PB), respectively. CD34⁺ cells were immunomagnetically purified from CB mononuclear cells using a MACS CD34⁺ MicroBead kit (Miltenyi Biotec, Auburn, CA, USA) and induced to differentiate toward eosinophils, as previously described (18). CD34⁺ cells were cultured in IMDM (Welgene, Gyeongsan, Korea) containing 10% fetal bovine serum (FBS), 100 U/ml penicillin, and 100 µg/ml streptomycin and supplemented with a cytokine cocktail of stem cell factor (SCF), Flt-3 ligand (Flt-3L), granulocyte-macrophage colony-stimulating factor (GM-CSF), IL-3, and IL-5 for the first 6 days. Cells were then plated in 12-well plates in medium supplemented with IL-3 and IL-5, and cultured with 50% medium change for an additional 6 days. The cells were further cultured in medium supplemented with IL-5 for an additional 12 days, with 50% medium change every 3 days. Cell differentiation was assessed by granule formation and nucleus shape after staining with Diff-Quick (Sysmex, Kobe, Japan). PB eosinophils and neutrophils were isolated from slightly atopic individuals by centrifugation on a Percoll solution (1.070 g/ml) followed by negative selection using anti-CD16 monoclonal antibody-conjugated microbeads (Miltenyi Biotec). Their purities were greater than 95%, as evidenced by Diff-Quick staining. Eosinophilic leukemia cell line (EoL-1) eosinophilic cells were maintained in RPMI 1640 medium (Welgene) supplemented with 10% FBS, penicillin (100 U/ml), and streptomycin (100 μg/ml).

Total RNA was isolated using TRI reagent (Molecular Research Center, Cincinnati, OH, USA) and treated with DNase I (Invitrogen Life Technologies, Carlsbad, CA, USA). cDNA was reverse-transcribed from 2 μg of total RNA using SuperScript II RNase H-Reverse Transcriptase in a 20-μl reaction containing random hexadeoxynucleotide primers (Invitrogen Life Technologies), deoxynucleotide triphosphates (dNTPs) (0.5 mM), magnesium chloride (MgCl₂) (2.5 mM), and DTT (5 mM). Reverse transcription was carried out at 42°C for 1 h, followed by heat inactivation at 70°C for 15 min. The resulting cDNA was amplified with the Accupower HotStart PCR Premix kit (Bioneer, Daejon, Korea). Real-time PCR was performed with SYBR Green mix (Roche, Mannheim, Germany) using a QuantStudio™ 3 Real-Time PCR system (Applied Biosystems, Foster City, CA, USA). mRNA levels were normalized by comparison with the level of PPIA mRNA as a reference gene. Data were expressed as relative expression using the comparative cycle threshold (ΔΔCT) method. The primers used for real-time PCR are listed in Table 1.

Cells were lysed in radioimmunoprecipitation assay (RIPA) buffer, resolved by SDS-polyacrylamide gel electrophoresis (PAGE), and transferred to polyvinylidene difluoride (PVDF) membranes. Blots were probed with anti-CTSG (Abcam, Cambridge, MA, USA), anti-MPO (R&D Systems, Minneapolis, MN, USA), anti-MBP1 (Abcam), anti-oligodendrocyte transcription factor 2 (Olig2) (IBL International, Hamburg, Germany), and anti-GAPDH (Santa Cruz Biotechnology, Santa Cruz, CA, USA) antibodies, followed by incubation with anti-rabbit or anti-mouse horseradish peroxidase (HRP)-conjugated secondary Ab (Cell Signaling Technology, Beverly, MA, USA). Immunostained proteins were detected using an ECL detection system (Amersham Pharmacia Biotech Inc., Piscataway, NJ, USA).

Developing CB cells were fixed in 4% paraformaldehyde, permeabilized in Tris-buffered saline (TBS) containing 0.1% saponin for 20 min, and incubated with 2% bovine serum albumin (BSA) for 1 h. The cells were stained with anti-CTSG (Abcam) and anti-human MBP1 antibody (Atlas Antibodies AB, Stockholm, Sweden) and then with Rhodamine-conjugated anti-mouse IgG antibody (Jackson ImmunoResearch, West Grove, PA, USA) or ALEXA488-conjugated anti-rabbit IgG antibody (Jackson ImmunoResearch). Isotype-matched mouse IgG2b or rabbit IgG was used as a control Ab. The cells were mounted in VECTASHIELD^® Mounting Medium with DAPI (Vector Laboratories, Burlingame, CA, USA). Expression of CTSG and MBP1 was visualized using a ZEISS LSM 800 confocal microscope (ZEISS, Oberkochen, Germany) and determined using ImageJ software (National Institutes of Health, Bethesda, MD, USA) (19). For measurement fluorescence intensity, 10 cells were randomly selected at days 6, 12, 18, and 24 during differentiation of CB cells.

All data were analyzed by independent t-test or analysis of variance using SPSS software (SPSS Inc., Chicago, IL, USA). Differences with a p-value <0.05 were considered statistically significant. Results are expressed as the mean±standard error of the mean.

We previously performed an RNA-seq analysis of differentially expressed genes at the late stages (days 18 and 24) when CB CD34⁺ hematopoietic stem cells were induced to differentiate toward eosinophils for 24 days (17). When we analyzed genes with reduced expression at day 24 versus day 18, the NPGP genes were enriched in the list of genes that were most downregulated (Table 2). These genes included PRTN3 (13^th), MPO (14^th), CTSG (27^th), LYZ (35^th), AZU1 (44^th), and ELANE (81^th), where the numbers in parenthesis indicate the rank of the most downregulated expression (the Gene Expression Omnibus database, accession number: PRJNA274725). Thus, the NPGP genes were drastically downregulated at the latest differentiation stage examined compared with the immediately preceding stage. In contrast, the transcripts of neutrophil secondary (LTF) and tertiary (MMP-2, MMP-9, and ARG1) granule protein-encoding genes remained virtually unchanged and were present at much lower abundance by 1 or 2 orders of magnitude compared with the NPGP transcripts. CD11b, which is used as an eosinophil marker (20), was upregulated as eosinophilic differentiation progressed, whereas CD16, which is not expressed by eosinophils (21), remained unchanged during these periods. Expression of GFI1 also remained unchanged.

We examined the expression of NPGP genes, together with genes for ESGPs and relevant transcription factors, over the 24-day culture period by quantitative real-time PCR. Transcripts of NPGP genes, including PRTN3, MPO, CTSG, AZU1, and ELANE were considerably expressed at day 6, reached peak expression at day 12, and sharply declined at days 18 and 24 (Fig. 1). These results are consistent with the finding that expression of these NPGPs is restricted at myeloblast and promyelocyte stages (1). LYZ, which is known to be expressed in a broader window from the myeloblast to metamyelocyte stages during neutrophil terminal differentiation (1), showed extension of maximum expression to day 18 (Fig. 1). C/EBPA and GFI1 mRNA levels were highest at day 6 and decreased thereafter, thus preceding expression of NPGPs. As expected, levels of C-C motif chemokine receptor 3 (CCR3) and ECP mRNAs increased as eosinophilic differentiation progressed. MBP1 mRNA expression increased up to day 18 during CB eosinophilopoiesis and then decreased to a modestly detectable level at day 24. Low expression of MBP1 mRNA was previously demonstrated in fully mature eosinophils such as PB eosinophils (2223). C/EBPE, GATA1, and GATA2 mRNAs were progressively upregulated, as eosinophilic cells became predominant in numbers, which was commensurate with upregulation of eosinophil-specific genes. Thus, these mRNA analyses are noteworthy in that neutrophil-lineage genes display abundant expression at the early stage and then a rapid decrease from the mid stage over the course of the 24-day culture, while expression of eosinophil lineage genes progressively increases. In addition, expression of the transcription factors that regulate the lineage-specific genes kinetically precedes or parallels that of their target genes. Immature eosinophilic EoL-1 cells has been used a useful in vitro model to study human eosinophils (24). Our and other studies have previously shown that transcripts of eosinophil-specific genes including MBP1, CCR3, and IL-5 receptor alpha (IL5RA) were upregulated upon stimulation of EoL-1 cells with butyric acid and dibutyryl-cyclic adenosine monophosphate (dbcAMP) (2526). We examined whether the expression patterns of NPGP and ESGP genes were similar to those seen in differentiating CB eosinophils. Four (PRTN3, MPO, CTSG, and ELANE) of the 6 NPGP genes were abundantly present in the absence of dbcAMP but were then markedly downregulated upon the addition of this agent, while eosinophil-lineage genes (CCR3, ECP, and MBP1) were upregulated. GFI1, C/EBPA, C/EBPE, GATA1, and GATA2 transcripts remained relatively unaltered or slightly increased as EoL-1 cells differentiated (Fig. 2). Although expression of these transcription factors in the cell line was not consistent with that in differentiating CB cells, the expression patterns of genes encoding NPGPs and ESGPs were almost identical in both cultures. Therefore, it should be emphasized that the eosinophilic lineage of clonogenic nature does express NPGP genes at the immature state and then turn off these genes with concomitant increases in ESGP gene expression upon maturation. Collectively, these data clearly show coexpression of NPGP and ESGP genes in a temporal manner during terminal differentiation of the eosinophil lineage.

We next examined whether the expression of NPGP genes was due to contamination of neutrophilic cells in the cultures of developing CB cells or whether a single eosinophilic cell changed its expression identity over the course of the 24-day culture. We performed immunofluorescence analysis over the culture period using CTSG and MBP1 as markers for neutrophils and eosinophils, respectively (Fig. 3A). At day 6, a small proportion of CB cells clearly expressed MBP1 but very weakly expressed CTSG. At days 12 and 18, the majority of CB cells robustly expressed both CTSG and MBP1, although the 2 proteins were not colocalized. At day 24 (Fig. 3A) and beyond (data not shown), MBP1 expression was maintained, but that of CTSG was markedly reduced to a relatively low level. Of note, there was also a small proportion of cells that were only CTSG-positive (see the CTSG-staining at day 24). For reference, expression of the 2 granule proteins was examined in PB eosinophils and neutrophils (Fig. 3B and C). MBP1 and CTSG were strongly detected in PB eosinophils and neutrophils, respectively, as expected. In contrast, CTSG in PB eosinophils and MBP1 in PB neutrophils were weakly detected by the immunofluorescence analysis. As control antibodies did not even produce faint signals (data not shown), these data indicate reciprocal expression of the lineage-specific protein in trace quantities that is assumed not to be expressed in eosinophils or neutrophils. These data suggest that immature eosinophils coexpress neutrophilic as well as eosinophilic granule proteins.

For immunoblot analysis, we examined the granule proteins that had been analyzed by immunofluorescent staining. CTSG was prominently expressed at days 12 and 18, the time points displaying strong expression of CTSG by immunofluorescence. A longer exposure showed weak CTSG expression at day 6 and/or day 24 (data not shown). A similar expression pattern was observed with anti-MPO Ab (Fig. 4A), in which MPO was expressed at days 12 and 18. Of note, EPX, an ESGP, increased with a decrease in MPO. The genes MPO and EPX form a cluster on human chromosome 17 with a similar genomic structure (27), and their gene products share 70% amino acid identity (28). The anti-MPO Ab used reacted to EPX (50 kDa) as well as MPO (55 kDa), as previously demonstrated (17). In contrast, MBP1 steadily increased as differentiation proceeded, while Olig2, a late eosinophil differentiation marker (17), was detected only at day 24. PB neutrophils strongly expressed CTSG and MPO, whereas PB eosinophils expressed MBP1 and EPX, but not CTSG or MPO (Fig. 4B). Even longer exposure failed to detect CTSG and MBP1 expression in PB eosinophils and neutrophils, respectively. These results indicate that protein levels of CTSG and MBP1 in differentiating CB cells correlate with their mRNA levels and the findings from immunofluorescent staining.