263

In Vitro Cell. Dev. Biol.—Animal 42:263–272, September and October 2006� 2006 Society for In Vitro Biology1071-2690/06 $18.00+0.00

DEVELOPMENT OF A CONTINUOUS CELL LINE, PBLE, FROM AN AMERICAN EELPERIPHERAL BLOOD LEUKOCYTE PREPARATION

S. J. DEWITTE-ORR, K. LEPIC, S. P. BRYSON, S. K. WALSH, L. E. J. LEE, AND N. C. BOLS1

Department of Biology, University of Waterloo, 200 University Avenue, Waterloo, Ontario, Canada N2L 3G1 (S. J. D., K. L., S. P. B., N. C. B.)and Department of Biology, Wilfrid Laurier University, Waterloo, Ontario, Canada (S. K. W., L. E. J. L.)

(Received 30 April 2006; accepted 11 May 2006)

SUMMARY

A continuous cell line, PBLE, was developed from the adherent cells in a culture of peripheral blood leukocytes fromthe American eel, Anguilla rostrata. The cells were grown in Leibovitz’s L-15 basal medium supplemented with 20%fetal bovine serum (FBS). Under normal culture conditions at 18� C, the morphology of PBLE was fibroblast-like. Thecultures have been subcultured over 80 times and have been cryopreserved successfully. These cells have a diploidkaryotype of 38 chromosomes, survived temperatures from 5 to 36� C, and proliferated at temperatures from 5� C to atleast 30� C. PBLE underwent apoptosis in response to gliotoxin, but did not show a respiratory burst. Results suggestthat PBLE may have arisen from a circulating mesenchymal stem cell. PBLE was susceptible to Chum salmon reovirus(CSV) and supported CSV replication. Therefore this cell line should be useful in studying eel specific virus–hostinteractions.

Key words: Anguilla rostrata; stem cell; reovirus; gliotoxin; apoptosis; temperature tolerance.

INTRODUCTION

The freshwater eels of the genus Anguilla Schrank have longbeen subjects of scientific study because of their complex life his-tory and their commercial value. However, in the last decade eelpopulations have declined dramatically, intensifying interest in thisgroup of teleosts (Lecomte-Finiger, 2003; Stone, 2003). The popu-lation decline has been most studied for the only two species in theAtlantic Ocean, A. rostrata and A. anguilla (Haro et al., 2000; Wirthand Bernatchez, 2003). The cause of the declining eel numbers isa matter of debate, but recently viral diseases have been suggestedto play a role (van Ginneken et al., 2005). Unfortunately, relativelylittle is known about the eel immune system (Nielsen and Esteve-Gassent, 2006), and cell lines to study eel diseases are few. Twocontinuous cell lines have been derived from A. japonica (Chen andKou, 1987; Fryer and Lannan, 1994), and two early passage epi-thelial cell lines have been developed from A. rostrata (Garrick,2000).

Cell lines from fish are valuable for exploring many aspects offish biology. They are particularly useful in detecting viruses (Fryerand Lannan, 1994) and in studying the molecular and cellular basisof physiological processes and toxicological mechanisms (Bols etal., 2005; Garrick et al., 2005). Although a cell line from a fishspecies that is not of direct interest can be a useful surrogate, hav-ing lines from the species of interest is superior for many purposes.For example, susceptibility to some viruses can be species specificand some physiological processes can vary between species, leading

1 To whom correspondence should be addressed at e-mail:ncbols@scimail.uwaterloo.ca

to different sensitivities to stimuli. Thus, additional cell lines fromA. rostrata would be valuable for studying species specific responsesat the cellular level.

Over the last 16 yr, a small number of cell lines with valuableproperties has arisen spontaneously in vitro from the peripheralblood (PB) of fish. At this time it is unknown how common thisphenomenon is among fish, or vertebrates in general, the nature ofthe cells giving rise to them, and the full potential of the cell linesto differentiate and express functional properties. The first PB cellline was derived from the common carp (Faisal and Ahne, 1990),but most PB cell lines have been from the catfish, including celllines expressing properties of monocyte/macrophages (Vallejo et al.,1991), NK cells (Shen et al., 2002), T lymphocytes (Hogan et al.,1999), and B lymphocytes (Miller et al., 1994). A likely origin ofPB cell lines is from circulating stem cells, although contaminationof the blood with cells from the blood vessel wall or overlying tissueis another, albeit remote possibility, as any cells released from thewall when drawing blood would be small relative to the number inperipheral blood. In human PB, circulating stem cells include adultpluripotent stem cells, mesenchymal stem cells, and endothelialprogenitor cells (Zhao et al., 2003; Roufosse et al., 2004; Khakooand Finkel, 2005). However, the nature of circulating stem cells inthe blood of poikilotherms is unknown and unlikely to be demon-strated directly in the foreseeable future due to the absence of iden-tifying cell surface markers and the difficulty and cost of developingthem. By contrast, demonstrating the routine development of par-ticular types of cell lines from fish PB would be indirect and in-expensive evidence for the presence of specific kinds of circulatingstem cells, as well as providing cell lines for a variety of purposes,such as studying viruses.

264 DEWITTE-ORR ET AL.

FIG. 1. PBLE morphology in culture. (A)PBLE form a monolayer of fibroblast-like cellswhen grown at 21� C. Over time areas in themonolayer begin to grow and form ‘centers’ inwhich the cells are more densely packed (B).These centers develop into a phase brightmound of cells (C) that produce phase brightviable cells that can reattach in a new cultureflask (D). Arrow indicates forming (B) andmore mature (C) centers. Magnification: (A andC) �100, (B) �150, (D) �200.

FIG. 2. Attachment efficiency of PBLE at different cell densities. PBLEwere plated at increasing cell densities in normal growth media with or with-out FBS supplementation. The number of cells able to attach 4 h after platingwas measured using a Coulter particle counter. Results are reported as per-centage of efficiency of attachment, a percentage of the number of attachedcells compared to total number plated.

For these reasons, the culturing of blood cells from the Americaneel (A. rostrata) was explored, resulting in the development andcharacterization of a peripheral blood leukocyte eel cell line, PBLE.This cell line demonstrated unique temperature tolerance, suscep-tibility to CSV, a member of the genus Aquareovirus, and the abilityto die via apoptosis.

MATERIALS AND METHODS

Culture initiation. Peripheral blood leukocytes (PBLs) were obtained fromhealthy American eel, A. rostrata. The eels were anesthetized in a 100 mg/Lclove oil solution (Anderson et al., 1997). Blood was obtained by venipunc-ture of the caudal blood vessel, transferred to a 15-ml polypropylene centri-fuge tube (Falcon/Becton-Dickinson, Franklin Lakes, NJ) and put immedi-ately on ice. The cells were centrifuged in a swinging bucket rotor centrifugeat 150–200 � g for 8 min at room temperature. Centrifugation resulted inthe separation of blood into plasma and red blood cells with a buffy coatlayer of white blood cells, which was removed using a glass pasteur pipette.These cells were placed into a 50-ml polypropylene centrifuge tube (Falcon)with �45-ml collection media (Leibovitz’s L-15 medium with 2% FBS, 2%penicillin–streptomycin solution, and 10 IU heparin/ml; all Sigma, St. Louis,MO). After gentle mixing by inversion to distribute cells evenly, the cellsuspension was distributed into 6–15 ml centrifuge tubes, in approximately7-ml aliquots. Three milliliters of Histopaque� 1077 (Sigma) was layeredunder the 7-ml cell suspension using a 9-inch pasteur pipette. This layeredsolution was centrifuged in a swinging bucket centrifuge at 400 � g for 30min. The leukocytes, which form a white band at the interface of the His-topaque and L-15, were collected with a pasteur pipette, pooled, and washedin growing medium (L-15 containing 20% FBS, and 150 U/ml penicillin G,150 �g/ml streptomycin sulfate). The cells were resuspended to 2.5 � 107

cells in 5 ml of growth medium in 12.5-cm2 tissue culture flasks (Falcon),and incubated at 21� C. The flasks were observed on at least a monthly basis,and after a year, one flask had one densely packed colony about the size ofa small coin. This colony of fibroblast-like cells was removed by trypsiniza-tion and the cells added to two new flasks, where they grew to confluencywithin a week.

Routine cell culturing and maintenance. PBLE growing medium is L-15containing 20% FBS and 2% penicillin/streptomycin. When reaching con-fluency, which is approximately every 7–10 d, cells are subcultured at a ratio1:2 according to the standard trypsinization method, using 0.5 mg/ml trypsin.The PBLE cell line has been cryopreserved in liquid nitrogen when sus-pended in growing medium with 10% dimethyl sulfoxide (DMSO; Sigma) asdescribed previously for cell lines (Bols and Lee, 1994) and has been suc-cessfully thawed. PBLE cultures were found to be free of mycoplasma con-tamination, as determined using Hoechst 33258 staining, as previously de-scribed (Chen, 1977).

Cell growth characteristics. The optimal growing conditions for the PBLEcell line were obtained by determining attachment efficiency, as well as tem-perature and FBS concentration preferences.

265DEVELOPMENT OF A CONTINUOUS CELL LINE, PBLE

FIG. 3. Optimal growth conditions for PBLE. PBLE growth was measured over time in conditions of varying FBS concentration orvarying temperature. Cells were seeded at a starting cell density of 5.0 � 104 cells per well and cell numbers were measured at specifictime points using a Coulter particle counter. When grown at varying FBS concentrations, the temperature was kept constant at 21� C.When testing optimal temperature conditions, the FBS concentration was kept at 20%.

For attachment efficiency, the number of attached cells from the numberseeded (expressed as a percentage) was determined at different starting celldensities and FBS concentrations. PBLE cells were plated in 12-well tissueculture plates (Falcon) at 5.0 � 104, 1.0 � 105, 1.5 � 105, and 2.0 � 105

cells per well with three replicates per cell density. The total volume of eachwell was brought to 1 ml growing medium (L-15, 2% penicillin/streptomycin),one plate using growing medium with 20% FBS, and one plate without FBS.Plates were incubated at 21� C for 4 h to allow cells to attach. After whichadherent cells were detached using trypsin and counted using a Coulter par-ticle counter (Beckman, Mississauga, ON).

Optimal growing serum concentrations were determined by plating PBLEcells at a density of 5.0 � 104 cells per well in 12-well plates with 1 mlgrowing medium. After an incubation at 21� C for 24 h to allow cells toattach, the growing medium was removed and 1 ml new medium containingvarying concentrations of FBS was added (0, 10, 15, and 20% FBS). Therewere three replicates for each FBS concentration. The cells, incubated at 21�C, were collected by trypsinization at 3-d intervals and counted using aCoulter particle counter.

Temperature preference was determined using a similar experimental de-sign. Cells were plated in normal growth medium as described above. Thecells were incubated at 21� C for 24 h to allow the cells to attach, after whichthe cultures were moved to different temperatures: 5� C, 12� C, 21� C, 28�C, and 30� C. Temperature experiments were performed using normal growingmedium supplemented with 20% FBS. Cells were collected and counted asdescribed above at 3-d intervals. Cultures were also placed at 36� C forobservation, but cell counts were not performed.

Chromosome preparations. Metaphase chromosome spreads (n � 66) wereprepared and counted from PBLE cultures at 112 population doublings. Stan-dard procedures with some variations were used (Freshney, 1994; Ganassinet al., 1999). Briefly, cultures with dividing cells were exposed to 0.4 �g/mldemecolcine (Sigma) for 2.5 h, trypsinized and centrifuged at 400 � g for10 min. The supernatant was removed and the cell pellet was suspended in0.075 mol/L KCl (Sigma) for 8–10 min after which the cells were collectedby centrifugation. The pellet was fixed in 3:1 methanol–glacial acetic acidfor 6 h at 4� C, collected again by centrifugation, and resuspended again inthe fixative. Slides were prepared by dropping 2–3 drops of cell suspensiononto a glass slide (VWR, West Chester, PA), stained with Giemsa stain (Sig-ma), and observed using a Nikon Optiphot microscope (Japan). The numberof chromosomes present per cell was counted under oil immersion.

Respiratory burst assay. PBLE were plated at 3.0 � 105 cells per well ina 96-well plate and allowed to attach overnight. Cells were then challengedwith lipopolysaccharide (LPS; 0.01–100 �g/ml; Sigma) or phorbol 12-myris-tate 13-acetate (PMA; 0.2 ng/ml–2 �g/ml; Sigma) in Dulbecco’s phosphate

buffered saline (D-PBS; Sigma) supplemented with D-glucose (500 mg/L) andimmediately measured for a respiratory burst using the fluorescent indicatordye, 2�,7�-dichlorodyhydrofluorescein (H2DCF; Molecular Probes, Eugene,OR), as previously described (Brubacher and Bols, 2001). Oxidation ofH2DCF was measured using a fluorometric plate reader (Molecular Devices,SpectraMax). Excitation and emission wavelengths were 485 and 530 nm,respectively. Measurements were taken every 5 min for 4 h, and then everyhour for 24 h. Relative fluorescent units (RFUs) from control and treatedcultures were compared.

CSV propagation and replication assay. CSV was obtained from AmericanType Culture Collection (ATCC) (Manassas, VA) and routinely propagated onChinook Salmon Embryo Cells (CHSE-214) monolayers. Media was collected7 d post-infection (pi), passed through a 0.2-�m filter, and kept frozen at�80� C until used. For all experiments, PBLE were grown to 90% confluencyin 25-cm2 tissue culture treated flasks (Nunc, Roskilde, Denmark), and themedia was changed to 5% FBS immediately prior to virus infection. Cellswere infected with 104.3 tissue culture infectious dose (TCID)50/ml, as deter-mined using the Karber method (Karber, 1931).

A CSV replication assay was performed by adding CSV (104.3 TCID50/ml)to a confluent monolayer of PBLE for 2 h to allow for viral attachment. Themedia was then removed, the cultures were washed twice with D-PBS andfresh growth media was added to each culture, supplemented with 5% FBS.These cultures were incubated for either 4 or 7 d, after which the media wascollected, passed through a 0.2-�m filter, put on a fresh monolayer of CHSE-214, and incubated for 3 d to measure syncytia formation. Intracellular in-fectious particles were detected by lysing the cells by 3 freeze–thaw cycles(5 min at �80� C and 5 min at 37� C), followed by resuspension of the celllysate in L-15 with 5% FBS. This cell lysate was filtered (0.2 �m), placedon fresh CHSE-214 monolayers and assayed for syncytia formation. Syncytiaformation on CHSE-214 would suggest that PBLE supports CSV replication.

Cytochemical staining of syncytia. Cultures were stained to demonstratesyncytia formation using a Romanowsky-Giemsa stain. PBLE cultures ex-posed to CSV or CHSE-214 reporter cultures from the CSV replication assaywere fixed in anhydrous methanol for 10 min, washed in distilled water for1 min and allowed to air dry. The slides were then submerged in stain (3 mg/mlazure B bromide in DMSO, 1 mg/ml eosin Y in anhydrous methanol, diluted1:15 in 10 mM HEPES buffer [10 mM HEPES, 5% DMSO, pH 6.8]) for 30min. Azure B bromide was purchased from Manufacturing Chemists (Nor-wood, OH), and eosin Y was purchased from Allied Chemical (Morriston,NJ). Slides were then rinsed in distilled water for 1 min, and allowed to airdry. Slides were mounted using a nonaqueous mounting medium and visu-alized using a Nikon Optiphot microscope.

Cell viability. Two fluorescent indicator dyes, Alamar Blue� (Medicorp,

266 DEWITTE-ORR ET AL.

FIG. 4. PBLE cellular morphology at different growing temperatures. PBLE, normally grown at 21� C and demonstrate a fibroblastmorphology, changed cell shape at different temperatures. At 5� C PBLE were round, compared to 12� C where they were fibroblast-likebut sparse. At 30� C and 36� C the cells flattened to a more epithelial-like morphology. All pictures taken at 3 d using �150 magnification.

Montreal, PQ) and 5-carboxyfluorescein diacetate acetoxymethyl ester(CFDA-AM) (Molecular Probes) were used to evaluate cell viability. The Ala-mar Blue� dye measures cellular metabolism (resorufin) (O’Brien et al.,2000), while CFDA-AM measures intracellular esterase activity, thus con-firming membrane integrity (Schirmer et al., 1997). The levels of fluorescenceare measured using a microplate spectrofluorometer. Alamar Blue and CFDA-AM assays were performed as previously described (Ganassin et al., 2000).Briefly, cells were plated at 3.0 � 105 cells per well in a 96-well plate andallowed to attach overnight. Cells were then exposed to serial dilutions ofCSV (initial titre � 104.3 TCID50/ml) or gliotoxin (5 �g/ml to 0.5 ng/ml) fora specified time period prior to measuring cell viability. After this exposure,

the media was removed and the two fluorescent dyes were added to the wells,diluted in a minimal salt solution called L-15ex the composition of whichhas been described previously (Schirmer et al., 1997; Ganassin et al., 2000).The cells were incubated with the dyes for 60 min before the microwellswere read with a fluorometric plate reader (Molecular Devices, SpectraMax).Excitation and emission wavelengths were 530 and 595 nm for Alamar Blue�and 485 and 530 nm for CFDA-AM. Cytotoxicity was indicated by a declinein fluorescence units (FUs) for treated cultures relative to untreated controlcultures. FUs for culture wells with no cells were constant and substractedfrom FUs for experimental and control cultures. For graphical presentation,the results were plotted using SigmaPlot (Jandel Scientific).

267DEVELOPMENT OF A CONTINUOUS CELL LINE, PBLE

FIG. 5. Frequency distribution of PBLEchromosomes. It was found that after 112 pop-ulation doublings, PBLE had a modal chro-mosome number of 38 (n � 66 cells). Insetpicture representative of a chromosomespread. Magnification: �2000.

FIG. 6. Chum salmon reovirus (CSV) induced cytopathic effect (CPE) in PBLE. The CSV induced CPE in PBLE, syncytia formation,became evident by 3 d post-infection (pi) (B) and developed into mature syncytia by 5 d pi (C). Control cultures show a healthy monolayerof cells (A). All pictures taken at �200 magnification.

gDNA laddering assay. Cells were plated at 2.5 � 106 cells per 25-cm2

flask and allowed to attach over night. Cells were then exposed to either 104.3

TCID50/ml CSV with or without 50 �g/ml ribavirin or increasing concentra-tions of gliotoxin, for a specified length of time. After the specified treatment,cells were collected and genomic DNA was extracted using a GenElute�mammalian genomic DNA miniprep kit (Sigma), which is based on columnaffinity technology. The genomic DNA was eluted from the column using 100�l Milli-Q water, and 35 �l of this DNA was resolved by electrophoresis ona 2% (w/v) agarose gel for 4 h at 60 V. The gDNA was visualized by staininggels with 0.5 �g/ml ethidium bromide and photographed under a UV trans-illumination.

RESULTS

Culture morphology. PBLE was derived from a peripheral bloodleukocyte preparation using a technique to preferentially isolate themonocyte/macrophage population. Out of this culture arose a con-tinuous cell line of fibroblast morphology (Fig. 1A), which appearsto be homogenous in cell type. Older cultures produce phase brightfoci or ‘centers’ (Fig. 1B and C), that generate phase bright round

cells. These phase bright cells remain in suspension but are ableto reattach in a new culture flask (Fig. 1D).

Culture growth. The growth of cultures required that the L-15 besupplemented with FBS. Both cell attachment and proliferation wereinfluenced by FBS. At lower cell densities cells attached with sim-ilar efficiencies, with or without FBS; however, at higher cell den-sities more cells were able to attach in the presence of FBS (Fig.2). With time, cell number increased slightly or not at all in cultureswith L-15 alone and proliferated best with more FBS as tested to20% FBS (Fig. 3).

Temperature also influenced the growth of cultures. PBLE grewbest at 21� C and 28� C. At 30� C a slight increase in cell numberoccurred, and although cell number was not measured at 36� C,cells remained attached to the culture wells for at least 7 d. Cellsmaintained at 5� C and 12� C persisted but did not proliferate (Fig.3B). It was concluded that culture grew best with 15–20% FBSsupplementation (Fig. 3A) and at 21–28� C (Fig. 3B).

268 DEWITTE-ORR ET AL.

FIG. 7. Chum salmon reovirus (CSV) replication in PBLE. PBLE were exposed to CSV (104.3 TCID50/ml) for 2 h, after which the viruscontaining media was removed and fresh media was added. The cells were incubated for either 4 or 7 d post-infection (pi). The conditionedmedia or cell lysate (derived from freeze–thaw cycles) was added to new CHSE-214 monolayers and allowed to incubate for 3 d beforesyncytia formation was assessed by Romanowsky stain. (A) Whole plate pictures of CHSE-214 reporter cultures show the differencesbetween 4 and 7 d pi and between virions detected extracellularly (CSV) and intracellularly (CSV freeze/thaw) from infected PBLEcultures. (B) Romanowsky–Geimsa stained CHSE-214 show syncytia formation at 4 d pi (syncytia outlined with dashed circle) and thebreakdown of large syncytia by 7 d pi (�100 magnification).

Culture morphology changed with different incubation tempera-tures (Fig. 4). After a week at the temperature extremes, 5� C and36� C, monolayers of PBLE remained intact, although monolayerappearance changed subtly. With time at 5� C, the cells becameshorter and rounded, whereas at 30� C and 36� C the cells becamemore epithelial-like (Fig. 4).

Chromosome counts. Chromosome counts were performed after112 population doublings. The counts ranged from 28 to 47, witha modal chromosome number of 38 (Fig. 5), which is the normaldiploid chromosome number for the American eel (Sola et al.,1980).

Respiratory burst. PBLE did not undergo a respiratory burst in

response to LPS or PMA, as measured using the fluorescent indi-cator dye, 2�,7�-dichlorodihydrofluorescein (H2DCF) (data notshown).

Chum salmon reovirus susceptibility. PBLE were susceptible toCSV and demonstrated cytopathic effect (CPE) following viral ex-posure. CSV exposed PBLE began to fuse and form syncytia within3 d pi, and showed mature syncytia formation by 5 d pi (Fig. 6).Late syncytia do not lift off the growing surface, as is observed withother fish cell lines, but remain adherent with an intact cell mem-brane (data not shown). A replication assay demonstrated the abilityof PBLE to support CSV replication (Fig. 7). By 4 d pi viral par-ticles could be detected within the cells but very few in the media,

269DEVELOPMENT OF A CONTINUOUS CELL LINE, PBLE

FIG. 8. Chum salmon reovirus (CSV) induced cell death in PBLE by a mechanism other than apoptosis. Cell viability was assessedin PBLE cultures exposed to serial dilutions of CSV (initial titre � 104.3 TCID50/ml) for 4 or 7 d using two fluorescent indicator dyes (A).Genomic (g)DNA from control and virus-exposed cultures with or without 50 �g/ml ribavirin, an inhibitor of viral transcription, was runon a 2% agarose gel. gDNA laddering in 180-bp oligomers, a hallmark of apoptosis, could not be detected (B). An arrow points to the500-bp band in the lane containing the 100-bp DNA marker.

while by 7 d pi a higher tire of infectious virions could be detectedin both the media and within the cells. CSV-induced syncytia for-mation did result in cell death (Fig. 8A); however, this cell deathdid not appear to be via apoptosis as determined by gDNA ladder-ing (Fig. 8B).

Gliotoxin-induced cell death via apoptosis. PBLE died followinga 24-h exposure to gliotoxin, a fungal secondary metabolite, as mea-sured using fluorescent indicator dyes, Alamar Blue and CFDA-AM(Fig. 9A). This gliotoxin-induced cell death was via apoptosis asdetermined by gDNA laddering into 180-bp oligomers, a hallmarkof apoptosis (Fig. 9B) (Mesner and Kaufman, 1997).

DISCUSSION

PBLE is the first continuous cell line developed from a peripheralblood leukocyte preparation of A. rostrata. Cell lines have beenpreviously derived from the adherent cells of catfish and carp pe-ripheral blood; however, PBLE differs from them in several ways.Channel catfish peripheral blood leukocytes produced severalmonocyte-like cell lines (Vallejo et al., 1991). These cell lines hadthe characteristic morphology of monocytes/macrophages and werecapable of several monocyte/macrophage functions, includingphagocytosis, interleukin-1 secretion, and antigen presentation (Val-lejo et al., 1991). CLC is a cell line developed from the mononuclear

cells of carp peripheral blood (Faisal and Ahne, 1990). Even thoughthese cells did not have a typical monocyte/macrophage morphologythey did react to a monoclonal antibody raised against carp mac-rophages and expressed at least two macrophage functions: PMAand LPS caused a respiratory burst and CLC secreted interleukin-1 (Faisal and Ahne, 1990; Weyts et al., 1997). By contrast, PBLEdid not show typical monocyte/macrophage morphology nor werethey capable of a respiratory burst. Under routine culture conditionsPBLE had a bipolar or fibroblast-like morphology.

PBLE might be derived from fish equivalents of several types ofcirculating cells that have been identified in mammalian blood andthat grow in vitro as adherent cells with morphologies different fromtypical monocytes. Among these are circulating endothelial progen-itor cells (EPCs), as well as mature circulating endothelial cells(CEPs) (Khakoo and Finkel, 2005). Smooth muscle progenitor cells(SPCs) also have been found in human peripheral blood (Simper etal., 2002). Pluripotent stem cells have been grown in vitro fromhuman and mouse leukocytes, particularly from different subsets ofmonocytes. These include f-macrophages, which could be inducedto differentiate into five distinct cell lineages (Zhao et al., 2003),and mesenchymal stem cells or progenitor cells (MPCs) capable ofexpressing properties of osteoblasts, skeletal myoblasts, chondro-cytes, and adipocytes (Kuwana et al., 2003; Roufosse et al., 2004).

270 DEWITTE-ORR ET AL.

FIG. 9. Effect of gliotoxin, a fungal secondary metabolite, on PBLE cell viability. Cell viability was measured in PBLE cultures exposedto gliotoxin after 24-h exposures (A). gDNA laddering, a hallmark of apoptosis, was measured from PBLE cultures treated with increasingconcentrations of gliotoxin for 24 h (B). An asterisk identifies the position of the 180-bp fragment and an arrow points to the 500-bp bandin the lane containing the 100-bp DNA marker.

The relationship between these stem cell types is still incompletelyunderstood. All of them can be expanded in vitro and in some casescell lines have been developed (Conrad et al., 2002).

The absence of antibodies to cell surface antigens and to func-tional proteins of differentiated cells makes identifying the originof PBLE and their differentiation potential difficult to determine atthis time. However, PBLE share some characteristics with MPCs.PBLE could be removed from the growth surface by conventionaltrypsin treatment, which was true for MPCs but not for f-macro-phages (Zhao et al., 2003). Like the mesenchymal stem cell line,V54/2, (Conrad et al., 2002), PBLE cultures produced a small num-ber of viable, round floating cells, which were able to reattach andproliferate in a new culture flask.

Peripheral blood cells of fish definitely appear to be a potentialsource of cell lines but how easily and universally they can begenerated is still uncertain. From the success so far, the blood offish can be assumed to contain some circulating stem or progenitorcells. Thus, one contributing factor to success would be the numberand type of circulating stem cells, which might vary with the healthstatus of the fish. In humans for example, CECs are elevated inpatents with septicemia (Mutunga et al., 2001), and MPCs can bemobilized by treatment with granulocyte–monocyte colony stimulat-ing factor (GM-CSF) (Roufosse et al., 2004). Another factor for cellline development would be how easily the cells spontaneously im-mortalize once in culture. In the case of the catfish, the successrate appeared high as many cell lines were established (Vallejo etal., 1991), whereas with the eel, cell line development appeared tobe a rare event as only one cell line developed. In the case of humanand rodent peripheral blood cells, direct immortalization has been

carried out with SV40 large-T-antigen leading to mesenchymal stemcell lines (Conrad et al., 2002). Application of such direct immor-talization procedures to fish peripheral blood cells might lead tocell lines being obtained more frequently.

Regardless of their cellular origin or differentiation potential,PBLE offer for the first time a peripheral blood cell line from A.rostrata and a tool for studying virus–host interactions specific tothe American eel. Upon exposure to CSV, PBLE illustrated the CPEthat is often characteristic of aquareoviruses, which is the formationof syncytia (Lupiani et al., 1995). Such a clear microscopic endpointcould make the cell line useful for evaluating tissue extracts forviral presence. PBLE were also able to support CSV replication,making this cell line valuable for studying viral replication mech-anisms as well as CPE. Viral exposures also lead to cell death inPBLE cultures after 4 d exposures to CSV.

Apoptosis is a common form of cell death; however, not all celllines are capable of dying via this mechanism (Kolenko et al., 1999;DeWitte-Orr and Bols, 2005). PBLE was able to undergo apoptosisas determined by gliotoxin-induced gDNA laddering. Gliotoxin is afungal metabolite and cytotoxic agent that usually kills by apoptosisand is often most potent to monocyte/macrophages (DeWitte-Orr andBols, 2005; Stanzani et al., 2005). Apoptosis is a frequent form ofreovirus-induced cell death (Clarke and Tyler, 2003); however, thedeath of PBLE in response to CSV did not appear to be by apoptosisas no gDNA laddering was detected. Further experiments need tobe performed to determine the precise mechanism of CSV-induceddeath of PBLE, but the results do imply that apoptosis is not ab-solutely necessary for viral replication and release from this cellline.

271DEVELOPMENT OF A CONTINUOUS CELL LINE, PBLE

The PBLE cell line appears to be unique among fish cell linesin surviving at 5� C to 36� C and proliferating from room temper-ature to 30� C, which are temperatures that encompass the rangesfor cell lines of both coldwater and warmwater fish. Cell lines fromcoldwater fish proliferate at temperatures as low as 5� C and as highas up to 24–26� C (Bols et al., 1992; Fryer and Lannan, 1994). Celllines of warmwater fish generally grow optimally around 30� C (Chenet al., 1988). The American eel is found from 5� to 62�18� N latitude(Scott and Crossman, 1973), which is a geographical range thatwould contain habitats with a wide range of temperatures. The tem-perature tolerance of the American eel could be exceptional andPBLE could be useful in exploring cellular and biochemical mech-anisms responsible for this. Also, as pointed out by several authors(Nicholson et al., 1987; Kang et al., 2003), cell lines that grow overa wide range of temperatures are potentially valuable for isolatingand studying both coldwater and warmwater fish viruses. Thus,PBLE may prove to be a valuable tool for understanding many eelspecific characteristics of temperature tolerance, cell death path-ways and antiviral mechanisms.

ACKNOWLEDGMENTS

This research was supported by scholarships to S. J. D.-O., K. L., and S.K. W., as well as grants to L. E. J. L. and N. C. B. from the Natural Sciencesand Engineering Research Council of Canada.

REFERENCES

Anderson, W. G.; McKinley, R. S.; Colavecchia, M. The use of clove oil asan anesthetic for rainbow trout and its effects on swimming perfor-mance. North Am. J. Fish. Manage. 17:301–307; 1997.

Bols, N. C.; Dayeh, V. R.; Lee, L. E. J.; Schirmer, K. Use of fish cell linesin toxicology and ecotoxicology of fish. In: Moon, T. W.; Mommsen,T. P., ed. Biochemistry and molecular biology of fishes: environmentaltoxicology. Amsterdam: Elsevier Science; 2005:43–82.

Bols, N. C.; Lee, L. E. J. Cell lines: availability, propagation and isolation.In: Hochachka, P. W.; Mommsen, T. P., ed. Biochemistry and molec-ular biology of fishes. Amsterdam: Elsevier Science; 1994:145–159.

Bols, N. C.; Mosser, D. D.; Steels, G. B. Temperature studies and recentadvances with fish cells in vitro. Comp. Biochem. Physiol.103A:1–14; 1992.

Brubacher, J. L.; Bols, N. C. Chemically de-activated 2�,7�-dichlorodihydro-fluorescein diacetate as a probe of respiratory burst activity in mono-nuclear phagocytes. J. Immunol. Methods 251:81–91; 2001.

Chen, J. D.; Yew, F. H.; Li, G. C. Thermal adaptation and heat-shock responseof tilapia ovary cells. J. Cell. Physiol. 134:189–199; 1988.

Chen, S. N.; Kou, G. H. Establishment, characterization and application of14 cell lines from warm-water fish. In: Kuroda, Y.; Kurstak, E.; Mar-amorosch, K., ed. Invertebrate and fish tissue culture. Tokyo: JapanScientific Societies Press; 1987:218–227.

Chen, T. R. In situ detection of mycoplasma contamination in cell culturesby Hoechst 33258 stain. Exp. Cell Res. 104:255–262; 1977.

Clarke, P.; Tyler, K. L. Reovirus-induced apoptosis: a minireview. Apoptosis8:141–150; 2003.

Conrad, C.; Gottgens, B.; Kinston, S.; Ellwart, J.; Huss, R. GATA transcrip-tion in a small rhodamine 123(low)CD34() subpopulation of a pe-ripheral blood-derived CD34(-)CD105() mesenchymal cell line.Exp. Hematol. 30:887–895; 2002.

DeWitte-Orr, S. J.; Bols, N. C. Gliotoxin-induced cytotoxicity in three sal-monid cell lines: cell death by apoptosis and necrosis. Comp. Bio-chem. Physiol. C. Toxicol. Pharmacol. 141:157–167; 2005.

Faisal, M.; Ahne, W. A cell line (CLC) of adherent peripheral blood mono-nuclear leucocytes of normal common carp Cyprinus carpio. Dev.Comp. Immunol. 14:255–260; 1990.

Freshney, R. I. Characterization. In: Culture of animal cells: a manual ofbasic techniques. 3rd ed. New York: Wiley-Liss; 1994:197–217.

Fryer, J. L.; Lannan, C. N. Three decades of fish cell culture: a current listingof cell lines derived from fishes. Methods Cell Sci. 16:87–94; 1994.

Ganassin, R. C.; Sanders, S. M.; Kennedy, C. J.; Joyce, E. M.; Bols, N. C.Development and characterization of a cell line from Pacific herring,Clupea harengus pallasi, sensitive to both naphthalene cytotoxicityand infection by viral hemorrhagic septicemia virus. Cell Biol. Tox-icol. 15:299–309; 1999.

Ganassin, R. C.; Schirmer, K.; Bols, N. C. Cell and tissue culture. In: Os-trander, G. K., ed. The handbook of experimental animals: the lab-oratory fish. London: Academic Press; 2000:631–651.

Garrick, R. A. Isolation and culture of capillary endothelial cells from theeel, Anguilla rostrata. Microvasc. Res. 59:377–385; 2000.

Garrick, R. A.; Woodin, B. R.; Stegeman, J. J. Cytochrome P4501A induceddifferentially in endothelial cells cultured from different organs ofAnguilla rostrata. In Vitro Cell. Dev. Biol. Anim. 41:57–63; 2005.

Haro, A.; Richkus, W.; Whalen, K.; Hoar, A.; Busch, W. D.; Lary, S.; Brush,T.; Dixon, D. Population decline of the American eel: implicationsfor research and management. Fisheries 25:7–16; 2000.

Hogan, R. J.; Taylor, W. R.; Cuchens, M. A.; Naftel, J. P.; Clem, L. W.; Miller,N. W.; Chinchar, V. G. Induction of target cell apoptosis by channelcatfish cytotoxic cells. Cell. Immunol. 195:110–118; 1999.

Kang, M. S.; Oh, M. J.; Kim, Y. J.; Kawai, K.; Jung, S. J. Establishment andcharacterization of two new cell lines derived from flounder, Parali-chthys olilvaceus (Temminck & Schlegel). J. Fish Dis. 26:657–665;2003.

Karber, J. Beitrag zur kollektiven behandlung pharmakologischer reihenver-such. Arch. Exp. Pathol. Pharmakol. 162:480–483; 1931.

Khakoo, A. Y.; Finkel, T. Endothelial progenitor cells. Annu. Rev. Med.56:79–101; 2005.

Kolenko, V.; Uzzo, R. G.; Bukowski, R.; Bander, N. H.; Novick, A. C.; Hsi,E. D.; Finke, J. H. Dead or dying: necrosis versus apoptosis in cas-pase-deficient human renal cell carcinoma. Cancer Res.59:2838–2842; 1999.

Kuwana, M.; Okazaki, Y.; Kodama, H.; Izumi, K.; Yasuoka, H.; Ogawa, Y.;Kawakami, Y.; Ikeda, Y. Human circulating CD14 monocytes as asource of progenitors that exhibit mesenchymal cell differentiation. J.Leuk. Biol. 74:833–845; 2003.

Lecomte-Finiger, R. The genus Anguilla Schrank, 1798: current state ofknowledge and questions. Rev. Fish Biol. Fish. 13:265–279; 2003.

Lupiani, B.; Subramanian, K.; Samal, S. K. Aquareoviruses. Annu. Rev. FishDis. 5:175–208; 1995.

Mesner, P. W. Jr.; Kaufman, S. H. Methods utilized in the study of apoptosis.Adv. Pharmacol. 41:57–87; 1997.

Miller, N. W.; Rycyzyn, M. A.; Wilson, M. R.; Warr, G. W.; Naftel, J. P.; Clem,L. W. Development and characterization of channel catfish long-termB-cell lines. J. Immunol. 152:2180–2189; 1994.

Mutunga, M.; Fulton, B.; Bullock, R.; Batchelor, A.; Gascoigne, A.; Gillespie,J. I.; Baudouin, S. V. Circulating endothelial cells in patients withseptic shock. Am. J. Resp. Crit. Care Med. 163:195–200; 2001.

Nicholson, B. L.; Danner, D. J.; Wu, J. L. Three new continuous cell-linesfrom marine fishes of Asia. In Vitro Cell. Dev. Biol. 23:199–204;1987.

Nielsen, M. E.; Esteve-Gassent, M. D. The eel immune system: presentknowledge and need for research. J. Fish Dis. 29:65–98; 2006.

O’Brien, J.; Wilson, I.; Orton, T.; Pognan, F. Investigation of the Alamar Blue(resazurin) fluorescent dye for the assessment of mammalian cell cy-totoxicity. Eur. J. Biochem. 267:5421–5426; 2000.

Roufosse, C. A.; Direkze, N. C.; Otto, W. R.; Wright, N. A. Circulating mes-enchymal stem cells. Int. J. Biochem. Cell Biol. 36:585–597; 2004.

Schirmer, K.; Chan, A. G. J.; Greenberg, G. M.; Dixon, D. G.; Bols, N. C.Methodology for demonstrating and measuring the photocytotoxicityof fluoranthene to fish cells in culture. Toxicol. In Vitro11:107–119; 1997.

Scott, W. B.; Crossman, E. J. Freshwater fishes of Canada. Bull. Fish. Res.Board Can. 184:623–629; 1973.

Shen, L.; Stuge, T. B.; Zhou, H., et al. Channel catfish cytotoxic cells: aminireview. Dev. Comp. Immunol. 26:141–149; 2002.

Simper, D.; Stalboerger, P. G.; Panetta, C. J.; Wang, S.; Caplice, N. M. Smoothmuscle progenitor cells in human blood. Circulation 106:1199–1204;2002.

272 DEWITTE-ORR ET AL.

Sola, L.; Gentili, G.; Cataudella, S. Eel chromosomes—cytotaxonomicalinterrelationships and sex-chromosomes. Copeia 4:911–913;1980.

Stanzani, M.; Orciuolo, E.; Lewis, R.; Kontoyiannis, D. P.; Martins, S. L. R.;St John, L. S.; Komanduri, K. V. Aspergillus fumigatus suppresses thehuman cellular immune response via gliotoxin-mediated apoptosis ofmonocytes. Blood 105:2258–2265; 2005.

Stone, R. Freshwater eels are slip-sliding away. Science 302:221–222;2003.

Vallejo, A. N.; Ellsaesser, C. F.; Miller, N. W.; Clem, L. W. Spontaneousdevelopment of functionally active long-term monocyte like cell linesfrom channel catfish. In Vitro Cell. Dev. Biol. 27A:279–286; 1991.

van Ginneken, V.; Ballieux, B.; Willemze, K.; Coldenhoff, K.; Lentjes, E.;Antonissen, E.; Haenen, O.; van den Thillart, G. Hematology patternsof migrating European eels and role of EVEX virus. Comp. Biochem.Physiol., C 140:97–102; 2005.

Weyts, F. A. A.; Rombout, J. H. W. M.; Flik, G.; Verburg van Kemenade, B.M. L. A common carp (Cyprinus carpio L.) leucocyte cell line sharesmorphological and functional characteristics with macrophages. FishShell. Immunol. 7:123–133; 1997.

Wirth, T.; Bernatchez, L. Decline of North Atlantic eels: a fatal synergy?Proc. R. Soc. Lond. B. Biol. Sci. 270:681–688; 2003.

Zhao, Y.; Glesne, D.; Huberman, E. A human peripheral blood monocyte-derived subset acts as pluripotent stem cells. Proc. Nat. Acad. Sci.USA 100:2426–2431; 2003.