Establishment of Epidermal CellLines Derived From the Skin of the

Atlantic Bottlenose Dolphin(Tursiops Truncatus)

JIN YU,1 MARK S. KINDY,1–3 BLAKE C. ELLIS,1,2 JOHN E. BAATZ,2,4

MARGIE PEDEN-ADAMS,2,4 TARA J. ELLINGHAM,5 DAYNNA J. WOLFF,5

PATRICIA A. FAIR,2,6 AND SEBASTIANO GATTONI-CELLI2,3,7*

1Department of Neurosciences and Neuroscience Institute, Medical University ofSouth Carolina, Charleston, South Carolina

2Marine Biomedicine and Environmental Sciences Center, Medical University ofSouth Carolina, Charleston, South Carolina

3Ralph H. Johnson VA Medical Center, Charleston, South Carolina4Department of Pediatrics, Medical University of South Carolina, Charleston,

South Carolina5Department of Pathology and Laboratory Medicine, Medical University of South

Carolina, Charleston, South Carolina6National Oceanic and Atmospheric Administration/National Ocean Service, Center forCoastal Environmental Health and Biomolecular Research, Charleston, South Carolina

7Department of Radiation Oncology, Medical University of South Carolina,Charleston, South Carolina

ABSTRACTThe Atlantic bottlenose dolphin (Tursiops truncatus), a marine mammal found off the Atlantic coast, has become the

focus of considerable attention because of an increasing number of mortality events witnessed in this species over the lastseveral years along the southeastern United States. Assessment of the impact of environmental stressors on bottlenosedolphins (BND) has been difficult because of the protected status of these marine mammals. The studies presented hereinfocused on establishing epidermal cell cultures and cell lines as tools for the in vitro evaluation of environmental stressorson BND skin. Epidermal cell cultures were established from skin samples obtained from Atlantic BND and subjected tokaryotype analysis. These cultures were further characterized using immunohistochemical methods demonstrating expres-sion of cytokeratins. By two-dimensional polyacrylamide gel electrophoresis (2D-PAGE), we observed that the proteomicprofile of BND skin tissue samples shared distinct similarities with that of skin-derived cultures. Epidermal cell cultureswere transfected with a plasmid encoding the SV40 small t- and large T-antigens, as well as the neomycin-resistance gene.Five neomycin-resistant clones were isolated and expanded, and all of them proliferated at a faster rate than nontransfectedBND epidermal cultures, which exhibited signs of senescence. Cell lysates prepared from two transfected clones were shownto express, by Western blot analysis, both SV40 tumor antigens. These experimental results are consistent with the conceptthat transfected clones expressing SV40 tumor antigens represent immortalized BND cell lines. Epidermal cell lines derivedfrom Tursiops truncatus will provide a unique tool for studying key features of the interaction occurring between dolphinsand the environment in which they live at their most crucial interface: the skin. © 2005 Wiley-Liss, Inc.

Key words: Tursiops truncatus; bottlenose dolphin; skin; cell lines; cytokeratin

Grant sponsor: National Oceanic and Atmospheric Administra-tion Dolphin HERA Project.

*Correspondence to: Sebastiano Gattoni-Celli, Department ofRadiation Oncology, Medical University of South Carolina, StromThurmond Biomedical Research Building, Room 338C, 114Doughty Street, Charleston, SC 29403. Fax: 843-876-5099.E-mail: gattonis@musc.edu

Received 10 January 2005; Accepted 11 August 2005

DOI 10.1002/ar.a.20266Published online 10 November 2005 in Wiley InterScience(www.interscience.wiley.com).

THE ANATOMICAL RECORD PART A 287A:1246–1255 (2005)

© 2005 WILEY-LISS, INC.

The bottlenose dolphin (Tursiops truncatus) is a cosmo-politan species that inhabits tropical and temperate re-gions, ranging from inshore to pelagic waters (Mead andPotter, 1990; Shane, 1990). Both natural and human-related factors pose threats to the well-being of marinemammals such as bottlenose dolphins (BND), especiallythose in coastal inland waters, due to exposure to runofffrom agricultural and industrial chemicals, sewage efflu-ents, and ingestion of contaminated prey. The increase innumber and frequency of BND mortalities over the last 2decades has led to increased concern about the potentialrole of environmental pollutants in marine mammal die-offs (Aguilar and Borrell, 1994; Mossner and Ballschmiter,1997). These events are of concern not only for the localmarine mammal populations but also because they may beindicators of the health of coastal ecosystems.

While recent mass mortalities among several marinemammal populations have been attributed to morbillivi-rus infection, the role of contaminants is also a possiblecontributing factor (Ross, 2002). The observation of skinpathologies and diseases previously unseen in BND(Bossart et al., 2003) suggests that BND may be an envi-ronmental sentinel for the coastal waters and an indicatorfor the health of the human communities inhabiting theEastern Seaboard of the United States, since the environ-ments of the coastal waters and the seashore clearly over-lap. As top-level predators, marine mammals accumulatehigh levels of persistent organic pollutants (POPs) in theirlipid-rich blubber tissues (Dietz et al., 1989; Aguilar andBorrell, 1994), and this appears also to be true for BND(Kuehl and Haebler, 1995). As protected species, there areethical and logistical reasons that limit opportunities toevaluate the effects of hazardous substances using con-ventional investigative techniques on these species. Invitro methods are necessary to evaluate toxicological ef-fects of pollutants in protected marine mammal species.

Cytogenic evaluations using fibroblasts and kidney cellshave been utilized in a number of different cetacean spe-cies such as beluga whale (Jarrell and Arnason, 1981) andthe BND (Bielec et al., 1997, 1998). Cultured cells fromvarious cetaceans have previously proven useful in assess-ing susceptibility to viruses (Cecil and Nigrelli, 1970;Nielsen et al., 1989; Kadoi et al., 1992; Dotzauer et al.,1994) and toxicants (DeGuise et al., 1998; Pfeiffer et al.,2000; Wang and Pfeiffer, 2001; Hu et al., 2002). Recentmolecular studies performed on DK1/CDK bottlenose dol-phin kidney cells demonstrate the value of cell cultures fortesting relevant environmental agents such as haloge-nated aromatic hydrocarbons and the polynuclear aro-matic hydrocarbons (Carvan et al., 2002).

Cetacean skin exhibits special adaptations to theaquatic environment. The epidermis serves as a primaryprotective barrier and represents the interface betweenthe organism and its environment (Pfeiffer and Menon,2002). However, investigations of cetacean skin are lim-ited and understanding its physiology and role in expo-sures to biological and chemical agents remains minimal.

We focused our effort on establishing epidermal cellcultures and cell lines, as they would provide a unique toolfor studying key features of the interaction occurring be-tween BND and the environment in which they live, attheir most crucial interface: the skin. These cultureswould provide a representative cellular system for assess-ing the effects of some environmental agents and marinebiotoxins. In this study, we report the development of

several BND skin cell cultures and cell lines and the use ofseveral techniques for characterizing them. Such in vitrosystems are important for future evaluations of environ-mental stressors on the skin of these protected mammals.

MATERIALS AND METHODSTissue Collection

Skin biopsy samples were obtained from five wild BNDtaken during capture-release health assessment studiesconducted in the Indian River Lagoon, Florida, and inCharleston, South Carolina, during summer 2004 underNational Marine Fisheries Service scientific research per-mits 998-1678 and 1008-1637-01. One skin sample wasobtained from a stranded BND.

Epidermal CulturesSkin tissue samples were processed as previously de-

scribed (Newton et al., 2000). Briefly, samples retrievedfrom each dolphin were transferred to a flask containingice-cold sterile Hanks’ balanced salt solution (HBSS) andgentamicin (50 �g/mL) and transported to the laboratory.In a biosafety level 2 laminar flow hood under sterileconditions, each skin tissue sample was minced into thesmallest pieces possible using sterile tweezers, scissors,and scalpel blades. While processing the skin tissue spec-imens, great attention was paid to separating out anyblubber material from the epidermis. This task was madefairly easy by the clear difference between blubber (white)and epidermis (black). Minced tissue fragments werewashed twice with HBSS containing gentamicin (50 �g/mL) and transferred to flasks containing 50–100 mL of anenzymatic solution made of RPMI-1640 medium contain-ing type I collagenase (1 mg/mL; Worthington Biochemi-cal, Lakewood, NJ), DNase I (0.1 mg/mL; Sigma Chemical,St. Louis, MO), and gentamicin (50 �g/mL), followed byincubation at room temperature overnight on a magneticstirring plate at low speed. The digested cell suspensionwas filtered through a sterile nylon filter (mesh size, 95�m) to exclude undigested fragments. The cell suspen-sions were transferred into centrifuge tubes, spun at 250gfor 5 min, washed twice with HBSS, checked for viabilityusing the trypan blue exclusion test, and resuspended incomplete medium. Complete medium was made up ofequal parts of RPMI-1640 medium and Keratinocyte-SFM, supplemented with Bovine Pituitary Extract andhuman recombinant Epidermal Growth Factor (catalognumber 17005-042; Invitrogen-Gibco), as well as 10% fetalbovine serum (FBS; Hyclone). Complete medium also con-tained penicillin (50 �g/mL), streptomycin (50 �g/mL),and ciprofloxacin (20 �g/mL). Cell suspensions wereplated in tissue culture flasks (Corning) with 0.2 �m ventcap and incubated at 37°C in a humidified atmosphere of5% CO2 in air. No bacterial or fungal contaminations wereobserved in any of the dolphin cell cultures obtainedthrough this protocol.

Histology and ImmunostainingDolphin skin samples were fixed in 10% buffered forma-

lin and paraffin-embedded. The sections from dolphin skinwere placed on 1% gelatin-coated slides and stained withhematoxylin and eosin. For histochemistry studies, epi-dermal cells were plated on eight-well glass chamberslides (Lab-Tek), washed twice in phosphate-buffered sa-line (PBS), and fixed in 4% paraformaldehyde for 30 min.

1247DOLPHIN SKIN CELL LINES

After fixation, cells were washed twice more with PBS andtreated for 30 min with 0.3% H2O2. In parallel, skin sec-tions were dewaxed, hydrated, and treated with 3% H2O2for 30 min. Epidermal cells on chamber slide and skinsections on coated slides were treated with normal horseserum to block nonspecific sites before overnight incuba-tion at 4°C with the primary antibody, mouse antihumancytokeratin AE1/AE3 monoclonal antibody (Dako Cytoma-tion M3515). Peroxidase mouse IgG kit (Vector Laborato-ries, Burlingame, CA) was used as recommended, with3,3�-diaminobenzidine (DAB) as the chromogen, followedby counterstaining with hematoxylin. All digital imageswere obtained by means of a Nikon Eclipse E800 micro-scope.

Two-Dimensional Polyacrylamide GelElectrophoresis (2D-PAGE)

Skin tissue or cultured skin cell pellets were incubatedwith lysis buffer consisting of 10 mM sodium dodecylsulfate (SDS), 195 mM dithiothreitol (DTT), 60 mM Tris,and 3.5 mM MgCl2. Samples were briefly sonicated whileon ice three times for 5 sec each. One ml of acetone wasadded to the suspension, incubated on ice for 20 min, thencentrifuged for 20 min at 10,000g at 4°C. The acetonelayer was decanted and the pellet air-dried for 10–15 min.Pellets were resuspended in 0.1% Triton X-100 using avortex. Undissolved material was pelleted by centrifuga-tion at 10,000 g. The supernatant was isolated and dia-lyzed against deionized water at 4°C overnight to desaltsamples using Pierce 3500 MW dialysis cups. Total pro-tein concentration was determined by the Micro BCA Pro-tein Assay Kit (Pierce). Samples were dried via centrifugalvacuum and resuspended in appropriate volumes ofDeStreak rehydration solution (GE Healthcare) to obtainequal protein concentration.

Both skin-derived and cell culture-derived sampleswere cup-loaded onto Immobiline pH gradient (IPG) gelstrips, pH 3–10, 11 cm. Strips were rehydrated overnightin 200 �l/strip of DeStreak rehydration solution (GEHealthcare) in a reswelling tray prior to sample loading.Isoelectrofocusing (IEF) was performed with an IPGphorsystem (GE Healthcare/Amersham) as follows: 500 V gra-dient for 1 min, 4,000 V gradient for 1.5 hr, and finally8,000 V step-and-hold until 15,000 Vhr was reached. Fol-lowing IEF, IPG strips were stored at �80°C overnight.The second-dimension SDS-PAGE was performed with aprecast 10–20% Tris-HCl (Bio-Rad) using Tris/Glycine/SDS running buffer at 200 V constant voltage for approx-imately 1 hr. Gels were then stained with Sypro Rubyprotein stain (Molecular Probes, Eugene, OR) and imagedwith Alpha Innotech FluorChem 8900 with UV transillu-mination at 300 nm and a SyproRuby emission filter.

Immunoblot AnalysisCell lysates from two G418-resistant clones, DS-T3 (T3)

and DS-T4 (T4), and from nontransfected controls (DS)were resolved by PAGE. Separated proteins were trans-ferred to a nitrocellulose membrane. The membrane wasblocked for 1 hr with gentle shaking in tris-buffered salinewith 0.1% Tween-20 (TTBS) with 5% nonfat dried milkand subsequently washed three times for 5 min each inTTBS. The membrane was then incubated for 1 hr inTTBS with monoclonal antibody Pab 108 (Santa Cruz)specific for SV40 large T- (LT) and small t- (ST) antigens.

Three 5-min washes in TTBS were performed followed byincubation with secondary antibody (goat antimouse an-tibodies conjugated with ZyMax (Zymed Laboratories) inTTBS with 0.5% nonfat dried milk for 1 hr with gentleshaking. The membrane was washed three times withTTBS for 5 min and then reacted with luminescent chro-mophore before being exposed to film for less than 1 min.

Karyotyping and Chromosome BandingChromosome preparations were derived from epidermal

cells of a female bottlenose dolphin and cultured in RPMI-1640 medium supplemented with 20% FBS. Cultures wereincubated in 5% carbon dioxide in air with 97% relativehumidity at 37°C. Colcemid was added to subconfluentcultures at a concentration of 0.05 �g/ml for 4 hr. Cellswere rinsed with Hank’s balanced salt solution (HBSS)and detached from the flask by treating the culture with0.25% trypsin-EDTA for 1–2 min at 37°C. Cells were har-vested and spun down at 4°C. The cell pellet was resus-pended in 5 ml of hypotonic solution (1:1 v/v of 75 mM KCland 0.8% sodium citrate) for 7 min and prefixed with 1 mlof Carnoy’s fixative (3:1 of methanol:glacial acetic acid).The cells were fixed three times using Carnoy’s fixativeand then dropped on wet slides at 24°C and 50% relativehumidity to obtain metaphase preparations.

Slides were kept at 90°C for 1 hr and chromosomes wereG-banded using trypsin and Giemsa. Room temperatureslides were trypsinized for 7 sec in a solution of 1:50 2.5%trypsin:Earle’s balanced salt solution. The trypsin wasrinsed off in a pH 6.8 Gurr’s buffer and cell preparationswere stained (1:50 Haleco Giemsa stain:Gurr’s buffer) for2.5 min. Slides were rinsed in tap water and allowed toair-dry.

Transfection of BND Cell CulturesBND epidermal cultures were transfected with a mix-

ture of lipofectamine (Invitrogen) and plasmid pZIP-776-1(a generous gift by Dr. William Hahn of the Dana-FarberCancer Institute). As shown in Figure 5, this plasmidencodes the SV40 small t- and large T-antigens as well asthe neomycin-resistance gene. Two days after transfec-tion, cells were changed with complete medium containing100 �g/ml of the neomycin analog G418 (Geneticin, In-vitrogen). This was the lowest concentration of G418 ableto kill 100% of BND cultured epidermal cells, whichproved extraordinarily sensitive to G418 selection. Fiveweeks after transfection, several G418-resistant coloniesbecame visible; cells from these colonies were harvestedby trypsinization using sterile cloning cylinders and ex-panded for further characterization. Growth curves oftransfected clones and control (nontransfected) cultureswere obtained as follows: cells were seeded at 50,000 cellsper well on six-well plates and grown in complete mediumfor 48, 72, 96, 120, 144, and 168 hr under normal cultureconditions. Cells were harvested with trypsin and broughtup to a final volume of 1 ml, from which cell counts wereobtained. Aliquots for counting were mixed with an equalvolume of trypan blue to exclude nonviable cells from thefinal count. Values were obtained from duplicate samplesusing a hemacytometer.

RESULTSUnder the auspices of the National Ocean Service, a

federal agency that is part of the National Oceanic and

1248 YU ET AL.

Atmospheric Administration (NOAA), and of the HarborBranch Oceanographic Institute, we had access to freshtissue samples of epidermis from wild BND at two loca-tions: the coastal waters of South Carolina and the coastalwaters of eastern central Florida (Indian River Lagoonecosystem). We also obtained tissue samples from BNDstranded along the South Eastern Seaboard.

Epidermal cell cultures (DS1-6) were establishedfrom six out of six BND specimens (one Florida strand-ing, two Indian River Lagoon controlled captures, threeCharleston controlled captures). Following digestion of

skin specimens as described above, cell suspensionswere plated in culture flasks in complete medium andincubated at 37°C. Within 24 hr, individual cells hadattached to the plastic surface and their morphologywas either fibroblastic or epithelioid. After a few days,we could observe several colonies of epidermoid-lookingcells, which eventually came to represent the vast ma-jority, if not the totality, of the cell culture. The initialcell suspension had a dark color due to the considerableamount of melanin pigment in BND skin. However, cellspropagated in culture flasks appeared to quickly lose

Fig. 1. Phase-contrast photomicrographs of DS1 epidermal cell cultures. Cell monolayer with epithelial/squamous morphology was photographed at 100� (A) and 200� (B) magnification. Note the several mitoticcells in each field. Scale bar � 100 �m (A); 50 �m (B).

1249DOLPHIN SKIN CELL LINES

the ability to produce melanin. No obvious differenceswere noted among the cell cultures initiated from thesix individual dolphins.

Figure 1 shows phase contrast images of DS1 cells atconfluence, at two different magnifications (100� in Fig.1A and 200� in Fig. 1B). The morphology of many cells isindicative of squamous epithelial/epidermal cells; nucleiwith prominent nucleoli are clearly visible. Grainlikestructures are also visible in the cytoplasm of many cells;these structures do not contain melanin, which disappearsearly on during in vitro propagation. Several cells under-going mitosis can be observed, consistent with a sustainedrate of proliferation, especially in early-passage cultures.Some of the cultures have undergone more than 60 celldivisions, at which point we observed slower proliferation,associated with cell senescence.

In an attempt to verify that cultured cells were derivedfrom skin epidermal cells, sections of BND skin and DS1cells grown on slides were subject to fixation and immu-nostaining with a mouse antihuman cytokeratin AE1/AE3

monoclonal antibody. We assumed that because of thehigh degree of conservation, BND cytokeratin would berecognized by this monoclonal antibody, which is routinelyused in clinical histopathology. Figure 2 shows that BNDskin sections and skin-derived in vitro cultures are immu-noreactive to AE1/AE3 monoclonal antibody, indicatingthat BND skin cultures comprise epidermal cells. Figure2A shows BND skin section stained with hematoxylin andsecondary antibody alone (peroxidase mouse IgG kit) as anegative control for immunostaining (dark granules in thecytoplasm of most cells represent melanin pigment). Fig-ure 2B demonstrates strong immune recognition of epi-dermal cells by AE1/AE3, indicating that BND and humancytokeratin can be recognized by the same anticytokeratinmonoclonal antibody. Figure 2 also shows DS1 cellsstained either with hematoxylin and secondary antibodyalone (peroxidase mouse IgG kit) as a negative control(Fig. 2C), or with hematoxylin, antihuman cytokeratinAE1/AE3 monoclonal antibody, and secondary antibody(Fig. 2D). Even though fibroblasts from the dermis were

Fig. 2. Immunostaining of tissue sections and cell cultures. A showsBND skin section stained with hematoxylin plus secondary antibody asa negative control for immunostaining. Note the abundance of melanin(dark) granules in the cytoplasm of most cells. B shows strong immunerecognition of sectioned epidermal cells by antihuman cytokeratin AE1/

AE3 monoclonal antibody. C shows cultured DS1 cells stained withhematoxylin plus secondary antibody as a negative control for immuno-staining. D shows positive immunostaining of cultured DS1 cells reactedwith antihuman cytokeratin AE1/AE3 monoclonal antibody. For all im-ages, magnification was 400�. Scale bar � 50 �m.

1250 YU ET AL.

certainly present in the initial cell suspension, virtuallyall cultured cells were positive for cytokeratin expression.Therefore, the positive immunostaining is consistent withthe epidermal origin of these BND skin-derived culturedcells.

Karyotyping of BND skin-derived cultured cells, shownin Figure 3, was accomplished by comparison of six sepa-rate metaphases to the karyotypes reported by Bielec etal. (1997, 1998). The 44 chromosomes were paired andseparated into four groups plus the sex chromosomes.Groups were defined according to chromosome length andcentromere position (arm ratio). The first group of subte-lomeric chromosomes consists of pairs 1 and 2 with an armratio of 3.0–7.0. The second group consists of pairs 3–11with an arm length ratio of 1.7–3.0 and classified as thesubmetacentrics. Group 3, the hardest to classify, wasmade of metacentric chromosomes (pairs 12–17) with anarm ratio of 1.0–1.67. The last group includes the telo-meric chromosomes (pairs 18–21) with an arm ratio of 7.0to infinity. The sex chromosomes, although kept separate,are classified as part of the metacentric group. The G-banded arrangement generated by our laboratory is con-sistent with the R-banded karyotype reported by Bielec etal. (1997).

Cell lysates from cultured dolphin (Tursiops truncatus)skin cells and isolated dolphin skin were subjected to2D-PAGE because of the extraordinary sensitivity of thistechnology to identify similarities in protein expression(proteomic profiles) between the two samples. These pro-files are evidence of a common origin or link. As illustratedin Figure 4, the protein profiles (as indicated by proteinspot patterns) for isolated skin (Fig. 4A) and DS1 cells(Fig. 4B) share several similarities. For instance, the vir-tually identical linear arrays of protein spots of � 75and � 40–45 kDa in both gels could not be observedunless the two samples shared a fundamental biologicalcommonality. It is important to note that absoluteamounts of various proteins as indicated by spot densitymay differ between samples, but the protein profile signa-ture is indicative of similarity between cultured cells andisolated whole skin. In addition, two subtypes of cytoskel-etal keratin type I, one with Mr of 64 kDa and pI � 5.8 andthe other with Mr of 55 kDa and pI of � 5.4, as identifiedin Figure 4, were both found in isolated skin and culturedskin cells via 2D-PAGE immunoblot (data not shown).

Oncogene-mediated phenotypic transformation is gen-erally utilized to overcome the finite lifespan of most pri-mary cultures, leading to cell immortalization and gener-

Fig. 3. GTG-Banded karyotype of a female Atlantic bottlenose dolphin. The 44 chromosomes arearranged into four previously described groups (subtelocentric, submetacentric, metacentric, and telocentric)according to centromeric position and arm ratio.

1251DOLPHIN SKIN CELL LINES

ation of cell lines that do not undergo senescence. In anattempt to obtain cell lines, BND epidermal cultures weretransfected with a mixture of lipofectamine (Invitrogen)and pZIP-776-1 (Fig. 5), a plasmid containing the neomy-cin-resistance gene and the early region of the SV40 DNAtumor virus; this region encodes for two oncogenic pro-teins: the small t- and the large T-antigens. Two days aftertransfection, cells were changed with complete mediumcontaining 100 �g/ml of the neomycin analog G418 (Ge-neticin, Invitrogen). This was the lowest concentration ofG418 that killed 100% of BND cultured epidermal cells,which are very sensitive to G418 selection. Five weeksafter transfection, several G418-resistant colonies becamevisible; these colonies were harvested by trypsinizationand expanded.

Two clones, DS-T3 and DS-T4, were selected for furtherstudies. Their morphology was not very different from

that of nontransfected parental cells (Fig. 6), but theyappeared to grow and reach confluence faster than thesecells. The results of Western blot analysis of cell lysatesfrom DS-T3 and DS-D4 clones demonstrate expression ofboth small t- and large T-antigens encoded by the earlyregion of SV40 (Fig. 7). Moreover, preliminary observa-tions indicate that these transfected cells express cytoker-atins and their karyotype is altered in comparison to thenormal karyotype because of the presence of abnormal/marker chromosomes (data not shown).

Cultures from all transfected clones proliferated at afaster rate than nontransfected BND epidermal cultures,which over time exhibited signs of senescence because oftheir reduced rate of proliferation. Figure 8 shows growthcurves of wild-type BND epidermal cultures (DS wt) com-pared with DS-T3- and DS-T4-transfected clones. Cellcounts were recorded at 48, 72, 96, 120, 144, and 168 hr

Fig. 4. Two-dimensional gel electrophoresis of isolated dolphin skin (A) and cultured DS1 cells (B) bothstained by SyproRuby stain. Vertical axis indicates approximate molecular weight, while horizontal axisindicates pH gradient of the IPG strip (approximate protein pI). Arrows indicate positions of two protein spotsthat were identified as cytoskeletal keratin type I proteins via immunoblot of 2D-PAGE.

1252 YU ET AL.

after the initial plating of cells. The two transformed celllines (DS-T3 and DS-T4) displayed a marked increase ingrowth rate compared to the nontransfected cultures.DS-T4 cells reached 100% confluence after only 72 hr,DS-T3 cells after 96 hr, and the control cultures not until144 hr. Moreover, DS-T3 and DS-T4 cells appeared toreach a higher saturation density than nontransfectedcultures (DS wt), as observed with phenotypically trans-formed and immortalized cell lines.

DISCUSSIONWe have developed skin-derived cell cultures (DS1-6)

from tissue specimens of BND. We have also establishedBND epidermal cell lines by transfecting some of thesecultures with a plasmid encoding the SV40 small t- andlarge T-antigens, as well as the neomycin-resistance gene.A similar approach has already been successfully utilizedto immortalize BND kidney cells (Pine et al., 2004).

Newly established epidermal cell cultures exhibit squa-mous/epithelial morphology, express cytokeratins, andgrow very well for at least 60 or so doublings before theirgrowth rate slows down, presumably as a result of senes-cence. Chromosomal analysis of skin-derived cell culturesfrom BND revealed an apparently normal female karyo-type with a modal number of 44 chromosomes. It is worthnoting that all karyotypes from BND epidermal cultureswere euploid, even after multiple passages. Thus, thesecell cultures may be useful for assessing genotoxic agentsand for chromosomal breakage analysis.

By DNA-mediated gene transfer, we obtained five neo-mycin-resistant clones expressing the large and small tu-mor antigens of the SV40 tumor virus; two of these cloneswere characterized further. Both exhibited a significantlyhigher growth rate than nonimmortalized control cul-tures. These new cell lines will also be useful for investi-gating the effects of marine biotoxins and other environ-mental stressors on dolphin skin.

Research into the impact of environmental stressors onmarine mammals has been difficult because of the obsta-cles posed to experimentation by the protected status ofthese animals. Understanding the effects of toxic pollut-ants on BND tissues and organs is important to the pro-tection and management of these animals; however, theresponse of BND to environmental stressors is not well

Fig. 5. Map of pZIP-776-1 plasmid encoding the SV40 small t- andlarge T-antigens, as well as the neomycin-resistance gene.

Fig. 6. Phase-contrast photomicrographs of DS-T3 and DS-T4 celllines. Subconfluent monolayers of DS-T3 (A) and DS-T4 (B) cells werephotographed at 100� magnification. Scale bar � 100 �m.

1253DOLPHIN SKIN CELL LINES

characterized or understood, especially when compared tohumans or rodents. We do not know which chemical con-taminants in the environment have the potential to be themost detrimental to dolphin health. While rodent andhuman cultures are available, they may not have the sameresponse or provide as accurate a tool for appropriatelyassessing potential risk. BND skin cell cultures and celllines are now available for testing the vulnerability ofdolphins to different environmental contaminants. There-fore, the cultures obtained represent an alternative invitro system for studying protected species such as marinemammals.

These epidermal cell cultures and cell lines will allowthe development of in vitro models for mechanistic studieson the effect that various contaminants may have on dol-phin skin; they will also allow a more systematic study ofthe skin, which is a crucial interface between marinemammals and the ocean in which they live. Moreover, invitro responses of BND cell lines to environmental stres-sors will help us understand biological effects seen at thewhole organism level. Cell lines will permit us to validateexisting assays, to establish new and more sensitive as-says for measuring BND health, and to predict the poten-

Fig. 7. Immunoblot of transfected BND cell cultures. Cell lysatesfrom two G418-resistant clones DS-T3 (T3) and DS-T4 (T4) and fromnontransfected controls (DS) were resolved by PAGE. Separated pro-teins were transferred to nitrocellulose membrane and reacted withmonoclonal antibody Pab 108 (Santa Cruz) specific for SV40 large T- (94kDa) and small t- (21 kDa) antigens. The membrane was then exposed to

goat antimouse antibodies conjugated with ZyMax (Zymed Laboratories)and subsequently reacted with luminescent chromophore, before beingexposed to film for less than 1 min. Both G418-resistant clones expresslarge T- and small t-antigens coded by the pZIP-776-1 plasmid, while nosignal is detected in nontransfected controls.

Fig. 8. Growth curve of dolphin skin control cultures and transformedcell lines. Cells were harvested at 48, 72, 96, 120, 144, and 168 hr afterplating. The two transfected cell lines (DS-T3 and DS-T4) displayed amarked increase in growth rate compared to the nontransfected cul-

tures. DS-T4 cells reached 100% confluence after only 72 hr, DS-T3cells after 96 hr, and the control cultures not until 144 hr. Values repre-sent the average of duplicate samples.

1254 YU ET AL.

tial impact of ocean environmental stressors on humanhealth.

To the best of our knowledge, only kidney-derived celllines from BND have been reported (Carvan et al., 1994;Pine et al., 2004); therefore, there is an urgent need togenerate additional research tools to better study thisimportant marine mammal. The establishment of celllines from BND skin will be of enormous benefit to dolphinresearchers and to other parties interested in the health ofthe marine environment. By testing environmental agentson BND epidermal cell lines, comparisons can be made toother marine species, as well as human cells, to betterunderstand responses of organisms to environmentalstressors. Moreover, the stability of the karyotype in ourepidermal cell cultures, even after many generations, sug-gests their possible use as biosensors for the ecosystem bytesting their ability to metabolize a series of environmen-tal pollutants and to repair DNA damage due to environ-mental carcinogens and/or mutagens.

ACKNOWLEDGMENTSThe authors thank the many individuals who partici-

pated in the Dolphin Health and Risk Assessment (HERA)Project during dolphin capture and release operations inCharleston and Florida, making possible the collection ofthese samples. The authors are grateful to Dr. GregoryBossart for collection of these samples under his NationalMarine Fisheries Permit Number 998-1678. Also, the au-thors thank Wayne McFee and the SC Marine MammalStranding Network for assisting with sample collectionfrom stranded animals. The authors extend their grati-tude to Drs. John Wise and Fred Holland for the conductof the cell work under his National Marine Fisheries Ser-vice Permit Number 1008-1637-01.

LITERATURE CITEDAguilar A, Borrell A. 1994. Abnormally high polychlorinated biphenyl

levels in striped dolphins (Stenella coeruleoalba) affected by the1990–92 Mediterranean epizootic. Sci Total Environ 154:237–247.

Bielec P, Riggs P, Womack J, Busbee D. 1997. A high resolutionGBG-banded karyotype of the Atlantic bottlenose dolphin, Tursiopstruncatus: generation of an ideogram, and NOR localization byfluorescence in situ hybridization. Cytogenet Cell Genet 78:6–11.

Bielec P, Gallagher D, Womack J, Busbee D. 1998. Homologies be-tween human and dolphin chromosomes detected by heterologouschromosome painting. Cytogenet Cell Genet 81:18–26.

Bossart GD, Meisner R, Varela R, Mazzoil M, McCulloch SD, Kil-patrick D, Friday R, Murdoch E, Mase B, Defran RH. 2003. Patho-logic findings in stranded Atjantic bottlenose dolphins (Tursiopstruncatus) from the Indian river Lagoon, Florida. Florida Scie 66:226–238.

Carvan M, Santostefano M, Safe S, Busbee D. 1994. Characterizationof a bottlenose dolphin (Tursiops truncatus) kidney epithelial cellline. Mar Mammal Sci 10:52–69.

Carvan M, Millar S, Busbee D. 2002. Cellular and molecular investi-gation utilizing a bottlenose dolphin epithelial cell line. In: PfeifferCJ, editor. Molecular and cell biology of marine mammals. Melabar,Florida: Krieger Pub. Co. p 313–327.

Cecil J, Nigrelli R. 1970. Virus susceptibility of cell lines derived fromtwo species of marine mammals. In Vitro 6:226.

DeGuise S, Martineau D, Beland P, Fournier M. 1998. Effects of invitro exposure of beluga whale leukocytes to selected organochlo-rines. J Toxicol Environ Health 55:479–493.

Dietz R, Heide-Jorgensen M, Harkonen T. 1989. Mass deaths of har-bor seals (Phoca vitulina) in Europe. Ambio 18:258–264.

Dotzauer A, Feinstone S, Kaplan G. 1994. Susceptibility of nonpri-mate cell lines to hepatitis A virus infection. J Virol 68:6064–6068.

Hu W, Jones P, Upham B, Trosko J, Lau C, Giesy J. 2002. Inhibitionof gap junctional intercellular communication by perfluorinatedcompounds in rat liver and dolphin kidney epithelial cell lines invitro and Sprague-Dawley rats in vivo. Toxicol Sci 68:429–436.

Jarrell G, Arnason U. 1981. Banded karyotypes of a belukha whale,Delphinapterus leucas. Hereditas 95:37–41.

Kadoi K, Mochizuki A, Ikeda T, Kamata H, Yukawa M, Inoue Y. 1992.Susceptibility of a line of dolphin kidney cell culture to severalherpesviruses. J Basic Microbiol 32:227–232.

Kuehl D, Haebler R. 1995. Organochlorine, organobromine, metal,and selenium residues in bottlenose dolphins (Tursiops truncatus)collected during an unusual mortality event in the Gulf of Mexico,1990. Arch Environ Contam Toxicol 28:494–499.

Mead J, Potter C. 1990. Natural history of bottlenose dolphins alongthe central Atlantic coast of the United States. In: Leatherwood S,Reeves R, editors. The bottlenose dolphin. San Diego, CA: AcademicPress. p 165–195.

Mossner S, Ballschmiter K. 1997. Marine mammals as global pollu-tion indicators for organochlorines. Chemosphere 34:1285–1296.

Newton D, Romano C, Gattoni-Celli S. 2000. Semiallogeneic cell hy-brids as therapeutic vaccines for cancer. J Immunother 23:246–254.

Nielsen O, Kelly R, Lillie W, Clayton J, Fujoka R, Yoneyama B. 1989.Some properties of a finite cell line from beluga whale (Delphi-napterus leucas). Can J Fish Aquat Sci 46:1472–1477.

Pfeiffer C, Sharova L, Gray L. 2000. Functional and ultrastructuralcell pathology induced by fuel oil in cultured dolphin renal cells.Ecotoxicol Environ Safety 47:210–217.

Pfeiffer C, Menon G. 2002. Cellular ultrastructural and biochemicalspecializations in the cetacean epidermis. In: Pfeiffer C, editor.Molecular and cell biology of marine mammals. Melabar, Florida:Krieger Pub. Co. p 396–411.

Pine M, Schroeder M, Greer K, Hokanson R, Busbee D. 2004. Gener-ation and partial characterization of a transformed cetacean cellline. Aquat Toxicol 67:195–202.

Ross P. 2002. The role of immunotoxic environmental contaminants infacilitating the emergence of infectious diseases in marine mam-mals. Hum Ecol Risk Assess 8:277–292.

Shane S. 1990. Comparison of the bottlenose dolphin behavior inTexas and Florida, with a critique of methods for studying dolphinbehavior. In: Leatherwood S, Reeves R, editors. The bottlenosedolphin. San Diego, CA: Academic Press. p 541–558.

Wang A, Pfeiffer C. 2001. Cytopathology induced by mercuric chlorideand methylmercury in cultured renal cells of the Atlantic spotteddolphin (Stenella plagiodon). J Submicrosc Cytol Pathol 33:7–16.

1255DOLPHIN SKIN CELL LINES