Journal of Immunological Methods 363 (2010) 90–94

Contents lists available at ScienceDirect

Journal of Immunological Methods

j ourna l homepage: www.e lsev ie r.com/ locate / j im

Technical Note

A novel method for procuring a large quantity of mature murine eosinophilsin vivo

Wonyoung Kim a, Juyang Kim b, Changshik Shin c, Hyunju Kim a, Younkyung Doh c,Hong R. Cho b, Byungsuk Kwon a,b,⁎a School of Biological Sciences, University of Ulsan, Ulsan, Republic of Koreab Biomedical Research Center, Ulsan University Hospital and School of Medicine, University of Ulsan, Republic of Koreac School of Nanobiotechnology and Chemical Engineering, Ulsan National Institute of Science and Technology, Ulsan, Republic of Korea

a r t i c l e i n f o

⁎ Corresponding author. School of Biological ScienceSan 29Mukeo-dong Nam-ku, Ulsan, Republic of Korea. Tfax: +82 52 259 2740.

E-mail address: bkwon@ulsan.ac.kr (B. Kwon).

0022-1759/$ – see front matter © 2010 Elsevier B.V.doi:10.1016/j.jim.2010.09.033

a b s t r a c t

Article history:Received 19 April 2010Received in revised form 14 September 2010Accepted 23 September 2010Available online 1 October 2010

Eosinophils are rare hematopietic cells that normally constitute only 1~3% of peripheral bloodleukocytes. It would be of help for the purpose of research to obtain a large quantity ofeosinophils. In this study, we wanted to develop a novel strategy to induce massive expansionofmurine eosinophils in vivo, based on the observation showing that treatment of IL-33 induceseosinophilia in mice. We generated an EL-4 lymphoma cell line (herein named EL-4-IL-33) thatwas engineered to secrete an active form of IL-33. We found that Siglec-F+ granulocytenumbers increased by 1850-fold in the peritoneal cavity 10 days after inoculation with 1×107

EL-4-IL-33 cells. This number corresponds to 74-fold increase, as compared with the number ofSiglec-F+ granulocytes in mice that received wild-type EL-4 cells. Siglec-F+ granulocytesexpanded by IL-33 had the circular nucleus and expressed eosinophil-specific genes. They alsoshowed some functional characteristics of eosinophils in that they had the ability to respond toIL-5 for survival and eotaxin-1 for chemoattaxis and to produce bioactive eosinophilperoxidase, suggesting that these cells are genuine eosinophils. Our results indicate thatsustained secretion of IL-33 by lymphoma cells in the peritoneal cavity is highly effective inincreasing peritoneal eosinophil numbers. Therefore, our simple method to obtain eosinphilson a large scale might be of value for eosinophil studies.

© 2010 Elsevier B.V. All rights reserved.

Keywords:EosinophilInterleukin-33Eosinophilia

1. Introduction

Eosinophils are end-stage cells that play critical roles ininflammatory processes involved in defense mechanismsagainst parasitic helminth infections and allergic reactions(Rothenberg and Hogan, 2006). Eosinophils are developedfrom hematopoietic stem cells andmature eosinophils migrateinto the peripheral blood and to tissues such as the gastroin-testinal mucosa. Upon various stimuli, eosinophils or theirprogenitors are recruited from the circulation into inflamma-

s, University of Ulsan,el.: +82 52 259 2860;

All rights reserved.

tory siteswhere theymay function asmodulators of innate andadaptive immunity. Effectormechanismsof eosinophils includethe release of cytotoxic mediators and the secretion of an arrayof proinflammatory cytokines (IL-2, IL-4, IL-5, IL-10, IL-12,IL-13, IL-16, IL-18, and TGF-α/β), chemokines, and lipidmediators (Blanchard and Rothernberg, 2009).

IL-33 is a recently describedmember of the IL-1 family. IL-33signals through ST2, the receptor for IL-33, and mediates thepathophysiololgical process of allergies by activating Th2 cells,mast cells, basophils, and eosinophils (Liew et al., 2010). Recentstudies suggest IL-33 as an alarmin, similarly to HMGB1(highmobility group box 1) and IL-1α, to alert cells of the innateimmune system in response to tissue damage during trauma orinfection (Haraldsen et al., 2009). Under apoptotic conditions,however, IL-33 undergoes cleavage into a less bioactive form bycaspase-3 and -7, and the cleaved IL-33 remains in the cell (Luthi

91W. Kim et al. / Journal of Immunological Methods 363 (2010) 90–94

et al., 2009). Accumulating results suggest that IL-33 might becritical in mediating a variety of inflammatory processes.

Even though murine eosinophils are different from humanones, murine models are very useful for the dissection ofesosinophil physiology in vivo (Legrand et al., 2008). However,a great obstacle tomurine eosinophil studies is in the difficulty inobtaining a large number of eosinophils for their functionalcharacterization. Here we report a novel method to expandeosinophils in the peritoneal cavity in vivo.

2. Materials and methods

2.1. Generation of the EL-4-IL-33 cell line

An active form of mouse IL-33 (amino acids 109–266)(Schmitz et al., 2005) fused to the OPG (TNFRSF11b) signalpeptide (MNKWLCCALLVLLDIIEWTTQ) (Lacey et al., 1998) wasamplified by PCR using sense (5′-CTAGCTAGCGACCACAATGAA-CAAGTGGCTGTGCTGCGCACTCCTGGTGCTCCTGGACATCATT-GAATGGACAACCCAGAGCATCCAAGGAACTTCACTT-3′) andantisense (5′-CCGCTCGAGCTACTTATCGTCGTCATCCTTGTAATC-GATTTTCGAGAGCTTAAACATAAT-3′) primers, cut with NheI and Xho I, and cloned into the pcDNA3.1/zeocin (Invitrogen,Carlsbad, CA) vector (pcDNA3.1/OPG-IL-33). pcDNA3.1/OPG-IL-33 was transfected into EL-4 cells using the MicroPoratorelectroporator (Digital Bio, Seoul, Korea). Zeocin (50 μg/ml)(Gibco BRL, Carlsbad, CA) was added to DMEM culture media toselect transfectant cells, starting on day 3 after transfection. Onecell line named EL-4-IL-33 was cloned by limiting dilution andmaintained in complete DMEM containing 50 μg/mL of zeocin.

2.2. Western blot analysis and ELISA (enzyme-linkedimmunosorbent assay)

EL-4 and EL-4-IL-33 cells (1×107)were lysed in ice-cold RIPAbuffer (Pierce, Rockford, IL) and centrifuged at 10,000×g at 4 °Cfor 10 min. Whole cell lysates were fractionated by 15%SDS-PAGE (sodium dodecyl sulfate-polyacrylamide gel electro-phoresis) before transfer to the PROTRAN nitrocellulose transfermembrane (WhatmanGmbH, Dassel, Germany). Themembranewas blocked by PBS (phosphate-buffered saline) containing0.05% Tween-20 and 5% ProPure non-fat dried milk (Amresco,Solon, OH). Immunoblotting was performed with anti-murineIL-33 polyclonal antibody (AdipoGen, Seoul, Korea) and horse-radish peroxidase-coupled anti-rabbit IgG (Jackson ImmunoR-esearch, West Groves, PA). Immunoreactive bands werevisualized using the enhanced chemiluminescence detection kit(GE Healthcare, Piscataway, NJ). EL-4 and EL-4-IL-33 cells(5×105 cells per well) were cultured in 6-well plates at a totalvolume of 3 ml per well for 3 days. Western blot analysis forculture supernatants was done as described above. Culturesupernatants were also used for ELISA. The Quantikine mouseIL-33 immunoassay kit was used for ELISA, according to themanufacturer's recommendation (R&D Systems, Minneapolis,MN).

2.3. In vivo expansion, staining, isolation, and culture of eosinophils

EL-4 and EL-4-IL-33 cells (1×107 cells per mouse) wereinoculated into the peritoneal cavity of C57BL/6 mice (Orient,Seoul, Korea). Ten days later, peritoneal lavage was conducted

with20 mlof sterile PBS tocollectperitoneal cells. Peritoneal cellswere further processed for either FACS (fluorescence-activatedcell sorter) analysis or isolation of eosinophils. For FACS analysis,peritoneal cells were first Fc (crystallizable fragment)-blockedwith anti-CD16/32 Ab (antibody) (2.4G2) for 15 min. Cells weresubsequently either double-stained with anti-Siglec-F-PE(phytohemagglutinin) (E50-2440) and anti-Gr-1-FITC(fluores-cein isothiocyanate) (RB6-8C5)or triple-stainedwithanti-Siglec-F-PE, anti-Gr-1-PerCP-CY (peridinin chlorophyll protein-cychrome) plus anti-MHC (major histocompatibility complex)class II-FITC (M5/114.15.2), anti-CD80-FITC (16-10A1), anti-CD86-FITC (GL1) or anti-CD137-FITC (1AH2)mAbs (monoclonalantibodies) (All mAbs were purchased from BD Biosciences, SanJose, CA). Labeled cells were analyzed using the FACS CANTO II(BD Biosciences, San Jose, CA). For purification of eosinophils,peritoneal cellswere stainedwithanti-Siglec-F-PEandanti-Gr-1-FITC mAbs, and Gr-1intSiglec-F+ cells were sorted using theMoFlo™ XDP cell sorter (Beckman Coulter, Fullerton, CA). Someof the sorted eosinophils were spun onto the glass slide for MayGrunwald staining (Diff Quik; Sysmex Corporation, Kobe, Japan)and the rest were used for culture and RNA extraction. Purifiedeosinophils were cultured by seeding at 5×105/mL in 48-welltissue culture plate containing RPMI 1640 medium supplemen-ted with 10% FBS and cultured in the presence or absence of10 ng/ml of mouse IL-5 (R&D Systems, Minneapolis, MN). After48 h, cells were harvested and washed twice in cold PBS, andviable cells were counted using a trypan blue exclusion assay.Apoptosis of eosinophils was assayed by analyzing annexinV-stained cells using the FACS CANTO II.

2.4. RT-PCR (reverse transcriptase-polymerase chain reaction)

Total RNA was extracted from the sorted eosinophils usingthe RNeasy-kit (Qiagen, Valencia, CA) and 1 μg RNAwas reversetranscribed using Superscript II reverse transcriptase at 42 °Cfor 1 h (Invitrogen). PCR was performed on serial dilutions ofcDNA using GoTaq polymerase (Promega, Madison, WI). PCRprimers used were as follows: β-actin forward (5′-ATGGGTCA-GAAGGACTCCTAT-3′) and reverse (5′-ATCTCCTGCTCGAAGTC-TAGA-3′); IL-5Rα forward (5′-GCCATTGACCAAGTGAATCC-3′)and reverse (5′-GTGGAATTTCCCATGACTTC-3′); C/EBPα forward,(5′-GAACAGCAACGAGTACCGGGTA-3′) and reverse (5′-CCCATGGCCTTGACCAAGGAG-3′); eosinophil peroxidase (EPO)forward (5′-CCTACTTCAAACAGCCAGTA-3′) and reverse (5′-ATCATTGCGTGGGATCTTGA-3′) primers.

2.5. Chemotaxis assay

Recombinant mouse eotaxin-1/ccl11 (Peprotech, RockyHill, NJ) dissolved in 0.5% BSA (bovine serum albumin)/PBS ata concentration of 10 mg/mlwas diluted in RPMI 1640mediumcontaining 10 ng/ml of mouse IL-5 until the concentrationsused in the assay were achieved. The chemotaxis assay wasperformed in a transwell plate with a 5.0-μm pore sizepolycarbonate membrane (Costar, Corning, NY) by placing100 μl of eotaxin-containing medium was placed in the lowerwell and 100 μl ofmedium containing50,000 eosinophils in theupper chamber. Cells were incubated at 37 °C for 3 h to permitmigration across the membrane in response to eotaxin-1.

Fig. 1. In vivo expansion of Gr-1intSiglec-F+ granulocytes by inoculation of EL-4-IL-33 into the peritoneal cavity. (A) Production of IL-33 in EL4-IL-33. (Left panel)Immunoblot of IL-33 in whole cell lysates and culture supernatant. (Right panel) ELISA analysis of IL-33 secreted by EL-4-IL-33. IL-33 was not detected (N.D.) in theculture supernatant of EL-4 cells. ***pb0.001 between the two groups. (C) FACS analysis of peritoneal cells of naïve mice, and mice inoculated with EL-4 or EL-4-IL-33cells. (B) Quantification of peritoneal cells based on the FACS analysis shown in (C). (Left panel) Total numbers of peritoneal cells. (Middle panel) Percentage of Gr-1+

Siglec-F+ cells. (Right panel) Gr-1+Siglec-F+ cell numbers. ***pb0.001, between the indicated twogroups (n=5 for naïvemice andn=11 formice that receivedEL-4orEL-4-IL-33 cells, respectively).

92 W. Kim et al. / Journal of Immunological Methods 363 (2010) 90–94

2.6. EPO activity assay

Measurement of EPO activity was done as described(Adamko et al., 2004). In brief, substrate solutionwas preparedby adding 800 μl of 5 mM o-phenylene-diamine (OPD) (Sigma,Oakvile, ON) and 1.25 μl of 30% H2O2 to 4 ml of 1 M Tris buffer(pH 8.0). OPD solutionwas adjusted to a final volume of 10 ml.Purified eosinophils or splenocytes (negative control) wereresuspended in phenol-red free RPMI 1640 medium at aconcentration 150,000 cells/ml and 15,000 cells (100 μl) wereallocated to each well. Cells were lysed with 0.2% SDS at 37 °Cfor 30 min. OPD solution was added for 20 min at roomtemperature. The reaction was terminated with 100 μl of 4 MH2SO4. EPO activity was determined by reading the plate at492 nm.

2.7. Statistics

Statistical significance was determined by the two-tailedunpaired Student's t test.

3. Results and discussion

We transfected EL-4 cells with mammalian vectorpcDNA3.1/zeocin containing a DNA segment coding for anactive form of IL-33 (amino acids 109–266) (Schmitz et al.,2005) fused to OPG signal sequence (Lacey et al., 1998) togenerate cell lines constitutively expressing a secreted form ofIL-33 (Fig. 1A). The EL-4-IL-33 cell line was established byselecting the transfected EL-4 cells in the presence of zeocin,followed by limiting dilution of the survived cells. Western blot

Fig. 2. Phenotypes of Gr-1intSiglec-F+ granulocytes expanded by inoculationwith EL-4-IL-33. Peritoneal cells were harvested 10 days after inoculationwith EL-4-IL-33cells into theperitoneal cavity. (A)MayGrunwaldGiemsa staining ofGr-1intSiglec-F+ cells sorted by aMoFlo cell sorter. (B) RT-PCRanalysis of eosinophil-specific genesin sorted Gr-1intSiglec-F+ cells. (C) FACS analysis of cell surface molecules on Gr-1intSiglec-F+ gates.

93W. Kim et al. / Journal of Immunological Methods 363 (2010) 90–94

analysis showed that IL-33 protein with molecular weight of18 kDa was detected in the cell lysates and the culturesupernatant of EL-4-IL-33 cells (Fig. 1A). This molecular size isconsistent with the molecular weight of recombinant IL-33protein which was produced in E. coli. EL-4-IL-33 cells seededwith 1×105 cells/ml produced IL-33 at the concentration of1,200 pg/ml in culture media after 3-day culture.

Fig. 3. Functional assay for purified eosinophils. (A) EPO activity for purified eosinoeosinophils after culture for 48 h in the presence or absence ofmouse IL-5. (C) Chemotaas percentage of cells migrating over control (no eotaxin-1).

We observed that intraperitoneal administration of IL-33induces massive eosinophilia in the bone marrow and in theperipheral blood (our unpublished data). Furthermore, subcu-taneous tumor masses formed by EL-4-IL33 cells contained ahigh percent of eosinophils (our unpublished data). Our datasuggested that IL-33 should play a role in the expansion ofeosinophils. Intraperitoneal inoculation of EL-4-IL-33 (1×107

phils and splenocytes measured by OPD. (B) Analysis of survival for purifiedxis assay for purified eosinophils towardmouse eotaxin-1. Datawere presented

94 W. Kim et al. / Journal of Immunological Methods 363 (2010) 90–94

cells per mouse) markedly increased peritoneal cell numbers.A high portion of peritoneal cells (30%)were cells that had highgranularity and expressed high levels of Siglec-F, a marker foreosinophils, and intermediate levels of Gr-1 (Fig. 1B). Theabsolute number of these Gr-1intSiglec F+ granulocytes presentin the peritoneal cavity was 7.4×107 per mouse 10 days afterchallengewith EL-4-IL-33, which corresponds to 1850-fold and74-fold increase compared with the number of peritonealGr-1intSiglec F+ granulocytes in naïve mice and in mice thatwere inoculated with wild-type EL-4 cells, respectively(Fig. 1C). It is noteworthy that 2 out of 11 mice that wereinoculatedwith EL-4-IL-33 had no tumor growth. This was dueto the anti-tumor effect of IL-33 (our unpublished data).

We purified Gr-1intSiglec F+ granulocytes using a MoFlo cellsorter to confirm that these cells are genuine eosinophils. Gimsastaining clearly showed the typical circular nucleus of a mouseeosinophil (Fig. 2A). We further examined the expression ofeosinophil-specific genes such as c/EBP, IL-5Rα, and EPO. RT-PCRanalysis demonstrated that purifiedGr-1intSiglec F+granulocyteshad clear expression of transcripts for those eosinophil-specificgenes (Fig. 2B). In addition, FACS analysis showed expression ofcell surface proteins including MHC class II, CD80, CD86, CD137,and ST2, which are known to be expressed in eosinophils(Rothenberg and Hogan, 2006) (Fig. 2C). Purified eosinophilsseem to have functional characteristics for general eosinophils inthat they showed a higher peroxidase activity compared withsplenocytes and they had the responsiveness to IL-5 (increase insurvival) and eotaxin (chemotaxis) (Fig. 3).

A large quantity of mouse eosinophils can be obtained fromIL-5 transgenic or parasitized mice that exhibit eosinophilia(see references in Rothenberg and Hogan, 2006). In this study,we demonstrated that challenge of IL-33-producing cells intothe peritoneal cavity induces massive accumulation of perito-neal eosinophils. We don't know the mechanism underlyingIL-33-mediating eosinophil expansion. Since target cells forIL-33, including mast cells, Th2 cells, and eosinophils, are alsotarget cells for IL-5, a cytokine most specific for eosinophildevelopment, it is possible that IL-33 promotes the generationof eosinophils by enhancing IL-5 production (our unpublisheddata). In our experimental system, IL-5 and other cytokinessuchas IL-3 andGM-CSF, onceupregulatedby IL-33,mighthavea synergistic action on esoinophil development together withIL-33. The effect of IL-33 on hemapotoiesis is currently underinvestigation in our laboratory.

Recently, Dyer et al. (2008) have developed a highlyefficient ex vivo culture system that generates eosinophils.Even though we did not compare the functional superiority ofeosinophils obtained using the twomethods, it seems that theyfunction equallywell in that their development is induced in anenvironment enrichedwith IL-5. On the other hand, our system

is unique as an experimental system. For example, it can beused to study the extramedullary eosinophil differentiation atEL-4-IL-33 inoculation sites, since we consistently observe thatIL-33 secreted by EL-4-IL-33 have no systemic influence oneosinophil differentiation in the bone marrow beyond inocu-lation sites. It also provides a good system for analyzing theanti-tumor activity of eosinophil within tumor.

4. Conclusion

The protocol shown in this manuscript is simple and easy toprocure a large number of eosinophils in vivo. Ourmethodmightmake it possible to perform in vitro and in vivo experimentswhich need high numbers of eosinophils, therefore contributingto eosinophil studies in the future.

Acknowledgements

This workwas supported by grants from the Nation ResearchFoundation of Korea (NRF) funded by the Ministry of Education,Science and Technology (R322009000200040 and BRL2009-0087350).

References

Adamko, D.J., Wu, Y., Gleich, G.J., Lacy, P., Moqbel, R., 2004. The induction ofeosinophil peroxidase release: improved methods of measurement andstimulation. J. Immunol. Meth. 291, 101.

Blanchard, C., Rothernberg, M.E., 2009. Biology of the eosinophil. Adv.Immunol. 101, 81.

Dyer, K.D., Moser, J.M., Czapiga, M., Siegel, S.J., Percopo, C.H., Rosenberg, H.F.,2008. Functional competent eosinophils differentiated ex vivo in highpurity from normal mouse bone marrow. J. Immunol. 181, 4004.

Haraldsen, G., Balogh, J., Pollheimer, J., Sponheim, J., Kuchler, A.M., 2009.Interleukin-33-cytolineofdual functionofnovel alarmin?Trends Immunol.30, 227.

Lacey, D.L., Timms, E., Tan, H.L., Kelley, M.J., Dunstan, C.R., Burgess, T., Elliott, R.,Colombero, A., Elliott, G., Scully, S., Hsu, H., Sullivan, J., Hawkins, N., Davy, E.,Capparelli, C., Eli, A., Qian, Y.X., Kaufman, S., Sarosi, I., Shalhoub, V., Senaldi,G., Guo, J., Delaney, J., Boyle,W.J., 1998. Osteoprotegerin ligand is a cytokinethat regulates osteoclast differentiation and activation. Cell 93, 165.

Legrand, F., Woerly, G., Driss, V., Capron, M., 2008. Innate immune function ofeosinophils. Meth. Mol. Biol. 415, 215.

Liew, F.Y., Pitman, N.I.,McInnes, I.B., 2010.Disease-associated functionsof IL-33:the new kid in the IL-1 family. Nat. Rev. Immunol. 10, 103.

Luthi, A.U., Cullen, S.P., McNeela, E.A., Duriez, P.J., Afonina, I.S., Sheridan, C.,Brumatti, G., Taylor, R.C., Kersse, K., Vandenabeele, P., Lavelle, E.C.,Martin, S.J., 2009. Suppression of Interleukin-33 bioactivity throughproteolysis by apoptotic caspases. Immunity 31, 84.

Rothenberg, M.E., Hogan, S.P., 2006. The eosinophil. Annu. Rev. Immunol. 24,147.

Schmitz, J., Owyang, A., Oldham, E., Song, Y., Murphy, E., McClanahan, T.K.,Zurawski, G., Moshrefi, M., Qin, J., Li, X., Gorman, D.M., Bazan, J.F., Kastelein,R.A., 2005. IL-33, an interleukin-1-like cytokine that signals via the IL-1receptor-related protein ST2 and induces T helper type 2-associatedcytokines. Immunity 23, 479.