In Vitro Cell. Dev. Biol.--Animal 37:386-394, June 2001 9 2001 Society for In Vitro Biology 1071-2690/01 $10.00+0.00

CHARACTERIZATION OF KIDNEY EPITHELIAL CELLS FROM THE FLORIDA MANATEE, TRICHECHUS MANATUS LATIROSTRIS JAMES M. SWEAT, 1 DAVID D. DUNIGAN, AND SCOTT D. W R I G H T

Marine Mammal Pathobiology Laboratory, Florida Fish and Wildlife Conservation Commission, 3700 54th Avenue South, St. Petersburg, Florida 33711 (J. M. S., S. D. W.), Department of Pathobiology, College of Veterinary Medicine, Universityof Florida, 2015 SW 16th Avenue, Gainesville, Florida 32610 (J. M. S.), and Gulf Coast Research and Education Center, Universityof Florida, 13138 Lewis-GallagerRoad, Dover, Florida 33527 (D. D. D.) (Received 21 July 2000; accepted 17 February 2001)

SUMMARY

The West-Indian manatee, Trichechus manatus latirostris, is a herbivorous marine mammal found in the coastal waters of Florida. Because of their endangered status, animal experimentation is not allowed. Therefore, a cell line was developed and characterized from tissue collected during necropsies of the manatees. A primary cell culture was established by isolating single cells from kidney tissue using both enzymatic and mechanical techniques. Primary manatee kidney (MK) ceils were subcuhured for characterization. These cells were morphologically similar to the cell lines of epithelial origin. An imnmnocytochemistry assay was used to localize the cytokeratin filaments common to cells of epithelial origin. At second passage, epithelial-like cells had an average population-doubling time of 48 h, had an optimum seeding density of 5 X 103 cells/cm2, and readily attached to plastic culture plates with a high level of seeding efficiency. Although the epithelial-like cells had a rapid growth rate during the first three passages, the cloning potential was low. These cells did not form colonies in agar medium, were serum dependent, had a limited life span of approximately nine passages, and possessed cell-contact inhibition. These data suggest that the cells were finite (noncontinuous growth), did not possess transformed properties, and were of epithelial origin. These cells are now referred to as MK epithelial cells.

Key words: Florida manatee; cell culture; cytokeratins; endangered species. status limits studies to a postmortem examination of fresh carcasses. Establishing and characterizing a manatee-derived cell line will provide researchers with a tool that can be used in in vitro simulations of pathogenic and nonpathogenic processes. Hopefully, this study will contribute to a better understanding of the causes of changes in the metabolic, immunological, and physiological conditions in manatees. Currently, cell cultures of nine marine mammal species are listed in the American Type Culture Collection (ATCC) Cell Lines and Hybridomas catalogue. These include five cetaceans (the spotted dolphin Stennella plagiodon, the Pacific bottle-nosed dolphin Turslops gilli, the Pacific common dolphin Delphinus bairdi, the finback whale Balenoptera physalus, and the Pacific pilot whale Globicephala scammoni), 3-pinnepeds (the harbor seal Phoca vitulina, the northern fur seal Callorhinus ursinus, and a sea lion Zalophus californianus). Kidney epithelial cells from Atlantic bottle-nosed dolphins Tursiops truncatus have also been characterized (Carvan et al., 1994). The purpose of this study was to develop epithelial cell lines isolated from MKs and to characterize these ceils with respect to growthmedium requirements, morphology, lineage, normal growth rate, optimal seeding density and efficiency, cloning potential, occurrence of transformation, and susceptibility to human viral pathogens.

INTRODUCTION

The Florida manatee, Trichechus manatus latirostris, is a herbivorous marine mammal that inhabits the coastal waters of Florida and its neighboring states. The manatee was listed under the U.S. Endangered Species Act of 1973, and is also protected by the Florida Manatee Sanctuary Act initiated in 1978. These statutes prohibit the use of manatees for experimental purposes. Part of the manatee researcher's strategy for the recovery of manatee populations is to determine the causes of mortality and morbidity. Thus, as an alternative to live-animal studies, tissue-culture analyses were initiated at the Florida Department of Environmental Protection's (FDEP) Marine Mammal Pathobiology Laboratory (MMPL), to understand the pathogenic effects of naturally occurring microbes and biotoxins. Certain postmortem examination reports include the description of lesions that are indicative of a viral infection (Jones and Hunt, 1984), and deoxyribonucleic acid (DNA) in situ hybridization has confirmed the presence of the papilloma virus in wartlike lesions in the manatees (G. D. Bossart, pets. comm.). However, no viruses have been isolated from the manatees to date. A culture system that is started from manatee tissue would be advantageous in evaluating manatee-associated viruses. Although cell cultures have been extensively used in biomedical research to determine the effects of pathogens in animals, the Florida manatee's endangered

MATERIALS AND METHODS M K tissue. Kidney tissue was collected from the carcasses of five mana-

tees--MSE9501, MSE9504, MNE9502, MEC9535, and SeaWorld of Florida

To whom correspondence should be addressed.

386

MKE CELLS FROM FLORIDA MANATEE

Trichechus manatus (SWFTM) 9535--brought to the MMPL from various locations in Florida. The tissues were collected during necropsies of fresh carcasses, which allowed the isolation of viable cells. Several other attempts to start primary cultures from tissue obtained from slightly autolytic carcasses were unsuccessful. Most of the assays were conducted using kidney tissue from MSE9504. For the primary culture, a 10-g section of kidney was dissected from a calf (MSE9504) less than 12 h postmortem. The cause of death was attributed to bilateral pneumonia. The tissue was rinsed with 72% ethanol and placed in a sterile 50-inl centrifuge tube containing 20 ml of 1• Hank~ balanced salt solution (HBSS) with 2x antibiotics consisting of penicillin (10,000 units penicillin-G), streptomycin (10 mg/ml), neomycin (20 mg/ml), and gentamicin solution (100 mg/ml) (Sigma Chemical Co., St. Louis, MO). Tissue dissociation. Approximately 8 g of kidney tissue, which included the medulla and cortex, was rinsed with sterile phosphate-buffered saline (PBS) (Sigma) and transfmTed to three separate, sterile plastic petri dishes (100-mm diameter) containing approximately 15 ml of HBSS with 2• antibiotics. Kidney capsule and connective tissue was removed from the serosal surface of the tissue, and discarded. Renal vessels and white fibrous tissues were avoided, to enhance the probability of isolating epithelial cells. The kidney tissue was cut into approximately l-ram pieces, placed in a lO0-mm plastic culture dish (Falcon #1029) containing approximately 15 ml of Oulbeceo's minimum essential medimn (DMEM) Ham's F-12 (Sigma) with 2• antibiotics, and processed as previously described by Freshney (1994). The minced tissue was transfmTed to a 150-ml Erlenmeyer flask and digested enzynmtically with 50 ml of 1X trypsin (2.5 g porcine tiTpsin/L in HBSS) (Waymouth, 1974) for 30 min at 37 ~ C. The trypsin containing the partially digested tissue was pipetted up and down in order to dissociate the connective tissue and release the single cells. Cells and clusters of cells were pelleted by centrifugation at 500 • g for 5 rain at room temperature. The supernatant was poured off, and 20 ml of DMEM with 10% bovine fetal serum (BFS) was added to the cell pellet. The cells were dispersed by gentle pipetting and held on wet ice until they were pooled with other cells. Pieces of undigested tissue were then transfen'ed to a stainless-steel tissue grinder with a #60 mesh screen. The trypsinized tissue was ground with a glass pestle over a lO0-mm plastic petri dish containing 20 ml of fresh trypsin. Fresh trypsin was pipetted on to the tissue between grindings. The supernatant was collected into a 50-ml tube and centrifuged at 500 X g for 5 rain, after which the pelleted cells were resuspended in 20 ml of DMEM containing 10% BFS, and then placed on ice. A red blood cell (rbc)-lysing buffer containing ammonium chloride, 8.3 g/L in 0.01 M Tris-HCL (Sigma), was used to lyse the rbcs. The number of viable eells/ml was determined by filling the counting chamber (0.1-mm deep) of an improved Neubauer hemocytometer with cells in a trypan blue stain. Growth medium evaluation. Four different media were used to seed 55-cm2 plastic culture plates containing approximately 3.7 • 104 cells. The plates were incubated at 5% CO~ and 37 ~ C. Reconstituted powdered media, with varying quantities of amino acids, vitamins, cofactors, and minerals, were used to determine which (of these) would result in the most efficient growth and colony fornmtion of cells. Basal medium Eagle B9638, B9763, and minimum essential nmdium Eagle M0268 (Sigma) were purchased in powder form, reconstituted with culture-grade water, and filter-sterilized (0.45-b~m filter). DMEM/Ham's F-12 D6421 was obtained in solution form. Antibiotics (IX) were added to each medium which was further supplemented with 10% BFS. Primary cultures initiated from MK tissue were grown for 2 wk in each of the above media. Five milliliters were removed and replaced with fresh medium every 3 d during the 2-wk growth period. The efficacy of each medium was evaluated by the number of single cells and clusters of cells that survived in primary culture, the number of colonies formed, and the rate at which confluent monolayers were formed. MK ceils from primary, first, second, and third passages were frozen and stored at - 7 0 ~ C (Hayrick and Moorhead, 1961). The cells were dissociated from the culture plate by removing the medimn and washing with PBS without magnesium chloride (MgC1) and calcium chloride (CaC1), and incubated at 37 ~ C in 2 ml of 1 • trypsin-ethylenediaminetetraacetic acid (EDTA) solution (0.5 g trypsin and 0.2 g EDTA per liter of HBSS). The freezing medium was DMEM/Ham'S F-12 with 7.5% dimethylsulfoxide and 10% BFS. The cells were pipetted into sterile 1-ml Nalgene cryogenic storage vials at 5 • 105 cells/ml and placed in a cryostorage box within a 15 • 15-cm styrofoam cooler, which allowed the cells to cool slowly at about l~ in a 70 ~ C freezer (Leibo and Mazur, 1971; Harris and Griffiths, 1977).

387

Morphology. Single cells and clusters of cells in the primary culture were photographed using an Olympus CK2 inverted microscope. Cell growth was evaluated daily by noting the changes in morphology, number, and size of the colonies. In order to resolve the cytosolic structures, selected histological stains were used on confluent colonies. Growth medium was removed, and the cells were washed in 1X PBS with MgC1 (0.1 g/L) and CaC1 (0.133 g/L), twice each for 5 rain. Three milliliters of 100% methanol fixative was added to each culture plate for 10 min at room temperature. After the methanol was removed, the ceils were washed twice with 1 • PBS and stained with a modified Wright's stain o1"a 1% crystal violet solution for 1 min at room temperature (McKeehan et al., 1977). Lineage determination. Cell lineage was determined by using immunocytochemical staining with commercially available antibodies to protein markers. Primary MK cells, spontaneously immortalized mouse granulosa (SIG) cells, and human skin fibroblast (HSF) cells were subcultured once and grown to confluence before immunofluorescence staining assays were completed. The technique was adapted from the product-use protocol 'Indirect Fluorescence LabeIing of Fibroblast for Vimentin' (Sigma). Indirect irnmunocytochemistry. Glass coverslips (12-ram diameter) were steam sterilized and placed in 2 cm2/well polystyrene culture wells (Corning #29820) prior to cell hm'vest. Twelve wells containing glass coverslips were seeded at 1 x 104 cells/well with MK, HSK and SIG. HSFs and SIGs were used as negative and positive controls for cytokeratin expression, respectively (Franke et al., 1979; Stein et al., 1991). Cells were grown to subconfluency in a 5% CO2 incubator at 37 ~ C. After removing the growth medium, the monolayer was rinsed twice in PBS containing 1 mM MgC1 and 0.1 mM CaC1. Cells were fixed in 100% methanol for 10 rain at room temperature, The fixative was removed, and the cells were rinsed three times with PBSiifor 5 min each. Excess PBS was removed, and the fixed cells were permeabil{zed with 0.1% Triton X-IO0 in PBS for 2 min at room temperature. The TritonX solution was removed, and the cells were washed three times in PBS for 10 rain each. After removing the excess buffer, a blocking agent made of 10% bovine serum albumin (BSA) fraction V 96% crude (Sigma) in RBS was used to wash the fixed cells thrice for 5 rain each. The cells were blocked for 10 nfin at romn temperature with BSA and incubated for 1 8 h at 37 ~ C with anticytokeratin and antivimentin (1:200 dilution) and antidesmosomal cytokeratin (1:100 dilution) (Sigma) in PBS with 1% BSA. After an 18-h incubation period, the culture plate was removed from the incubator and held at 4 ~ C for 30 rain to stabilize the antibody and antigen association. Excess primary antibody solution was removed, and the cells were rinsed thrice for 15 min each in PBS. The cells were blocked again as previously described. An appropriate fluorescein-labeled secondary antibody, conjugate diluted to 1:256 in PBS, was added to the ceils. The cells were incubated at room temperature for 1 h in the dark. The cells were rinsed in PBS thrice for 5 min each. The coverslips were removed from the wells and blotted dry. A mounting medium consisting of 1 mg of m-phenylenediamine in 10 ml of a 50:50 soiution of glycerul and distilled water was used to retard the fading of the fluorescein dye. The coverslips were placed on microscope slides for observation under an epiftuorescence microscope (Olympus BH2 with fluorescence). MK cells that had not been stained were screened for autofluorescence. Antibodies. The monoclonal mouse IgG2a anticytokei'atin (clone #8.13), mouse IgM antivimentin (clone #VIM 13.2), mouse IgM monoclonal antidesmosomal cytokeratin (clone #KD80.20), goat anti-mouse IgG (Fab specific) fluorescein isothiocyanate (FITC) conjugate, and goat anti-lnouse IgM (Muchain specific) FITC conjugate were obtained from Sigma. Cell lines. The HSF cell line was a gift from Dr. Bill Ross, Eckerd College, St. Petersburg, Florida, and is also available from the ATCC. The SIC cell line was a gift from Dr. Robert Burghardt, Texas A&M University, College Station, Texas (Stein et al., 1993). Growth analysis. Three different seeding densities were used for this assay. Duplicate wells were seeded at each density. MK cells were at the second passage and grown to approximately 75% confluence. The cells were trypsinized, dispersed, counted, and held on ice. Three 24-well culture plates were seeded with cell densities of 2.5, 5, and 8.7 x 10a cells/cm 2 and incubated in a 5% C02 incubator at 37 ~ C. A cell count was determined for each density every 24 h for 9 d. Population-doubling times (PDTs) were determined for MK cells as they progressed from a primary culture through the fifth passage. Seeding-density determination. MK cells, at second passage, were plated onto a 6-well (9.6 cm2/well) plate. The cells were seeded at 1, 2, and 3 X 10a

388

SWEAT ET AL.

and at 1 and 4 • 104 cells/cm z, and incubated in a CO2 incubator at 37 ~ C. The medium was renewed every 3 d. After 8 d of growth, the cells were trypsinized and counted by using a hemocytometer. The number of population doublings (PDs) was an index of growth at each density. Seeding efficiency. Two seeding-efficiency assays were conducted for the MK celts. The first was conducted using cells at the second passage, and a second assay was completed at the fifth passage. In each assay, a 24-well culture plate was seeded with 2.1 • 104 cells/cm 2. The medium containing unattached ceils was removed from the wells 30, 120, and 300 min after seeding. As the number of unattached cells was low, the supematant was centrifuged and resuspended in 200 txl of medium before enumeration using a hemocytometer. The seeding efficiency was calculated as follows: (number of attached cells/number of cells seeded) • 100. Cloning efficiency. Second-passage MK cells were grown to approximately 75% confluence and used for a cloning-efficiency assay. After a dilute-trypsin treatment (50%), the epithelial cell colonies were preferentially aspirated from the plate using a fine-tipped glass pipet. The epithelial cell colonies were viewed using an inverted microscope to locate cells with an epitheliallike morphology during the harvesting step. Three colonies of epithelial-like ceils were removed from the culture plate and placed in a test tube with 1 ml of DMEM containing 10% BFS. Four test tubes containing 2 ml of cloning medium (DMEM and conditioned medium mixed 2:1 supplemented with 30% BFS) were prepared. Four 1:10 serial dilutions of cells were made and plated onto a 96-well plate. The plates were placed in a 5% CO2 incubator at 37 ~ C and left undisturbed for 3 d. The medium was renewed every 3 d during the 2-wk incubation period. The plates were visually assessed every third d to estimate the approximate number of cells in each well. A similar culture was initiated from small distinct colonies from primary cultures as well. Cell-transformation potential. For growth in soft agar assay, a base underlay preparation of 1.2% agarose gel was made and poured into 60- and lO0-mm culture plates and stored at 4 ~ C. Four 60-mm culture plates containing previously prepared agar underlay were placed in an incubator for 10 min and allowed to stabilize at 37 ~ C. An agar-cell overlay was prepared by trypsinizing second-passage MK cells and positive control cells (transformed rat ovarian surface epithelial [ROS-T] cells) (Hoffman et al., 1993) and suspending them in 2• DMEM Ham's F-12 supplemented with 20% BFS and 2• antibiotics. The MK and ROS-T cells were diluted to 2.4 • 104 and 1.3 • 105 cells/ml, respectively. Equal amounts of 2• DMEM and 1.2% melted agar were mixed in a 50-ml centrifuge tube to produce a 0.6% agarmedium suspension, and placed in a 45 ~ C water bath. MK cells in 2• medium and 0.6% agar medium cooled to 37 ~ C were mixed in equal amounts and plated onto the four 60-mm culture plates containing an agar underlay. The plates containing ceils were allowed to solidit~r at room temperature for 10 min. The cells were placed in a CO~ incubator at 37 ~ C. Alter 14-21 d, colonies of cells were counted by using an inverted light microscope. DMEM supplemented with 0, 1.25, 2.5, and 5% BFS was prepared and inoculated with MK cells at second passage for the serum-dependency growth assay. The MK cells were seeded at a density of 8.3 • 103 cells/cm 2 in each cuhure well. A separate 60-mm plate was seeded with cells fi'om the same passage and with the same density, and grown in a separate stock medium of DMEM supplemented with 10% BFS as a positive control (Lindgren et at., 1975). The cells were grown in a CO2 incubator at 37 ~ C for 8 d. The medium was renewed on days 3 and 6 of the growth period. After 8 d, the cells were photographed, harvested, and counted by using an improved Neubauer hemocytometer. Susceptibility to human viral pathogens. In order to determine if MK ceils were permissive for the replication of various human viruses in vitro, a series of cultures were inoculated at a tissue-culture infective dose known to cause either a cytopathic effect (CPE) or a replication in a permissive control culture as previously determined. Duplicate cultures of 80% confluent MK and buffalo green monkey kidney (BGM) cells were washed twice with PBS and inoculated either with 0.1 ml of a 1:10 virus dilution or with growth medium alone (negative controls) for 1 h at 37 ~ C. The viruses were human isolates and included: adenovirus, eco-9 and 22, coxsackie B5 and CA9, influenza A and B, parainfluenza, respiratory syncytial virus (RSV), reo, and herpes-1. Afterwards, culture media were added to each well, and the cultures were evaluated over the following 8 d. Visible signs of CPE were noted and assigned the following values: +1 = small degree of cytoplasmic vacuolation and/or swelling; +2 = 25% ceils show distinct cytoplasmic vacuolation and/or swelling; +3 = 75-95% cells show attenuated cytoplasmic processes; and

TABLE 1

VARIOUS CULTURE MEDIA WERE USED TO EVALUATE THE GROWTH POTENTIAL OF PRIMARY MKE CELLS. THE EFFICACY OF EACH GROWTH MEDIA TO SUSTAIN SINGLE-CELL VIABILITY, PROMOTE COLONY FORMATION, AND REACH EXPONENTIAL GROWTH AND COLONY FORMATION WAS DETERMINED

Medium DMEM F-13 BME 9638 BME 9763 MEM 0268

Single-cell viability ++ + + +

No. of days to reach No. of days to exponential reach Colonyformation growth confluency ++ ++

6 9

14 -

Cells passing through mitosis (+ +), viable cells (+), no-growth response (-). +4 = 100% cells show cytoplasmic swelling attenuated processes and/or are detached (cells are nonviable). Following culture for 8 d, hemagglutination (HA) tests were conducted on wells inoculated with influenza A and B and parainfluenza as previously described (Hsiung et at., 1994). Wells inoculated with RSV and medium alone were used as negative controls for the HA tests. Briefly, cells were washed twice with PBS. Guinea-pig rbcs (0.5 ml) at 0.5% in PBS were added to each well and allowed to incubate undisturbed at room temperature for 15 min. Afterwards, the ceils were observed for the presence of adherent rbc or rosettes that were not associated with the cell cytoplasm. RESULTS Approximately 80% of the cells removed from the kidney tissue were viable using a trypan blue-exclusion stain. Two enzymatic treatments with trypsin for 30 rain digested the connective stroma from around single ceils and clusters. A stainless-steel-mesh sieve freed single ceils from the tissue after two treatments with trypsin. Morphological observations after 8 d in culture revealed a mixture of epithelial- and fibroblast-like cells. The MK cells readily attached to the polystyrene substrate surface after 1 h. DMEM Ham's F-12 with 10% BFS was found to be the most effective in supporting ceils to mitosis and in promoting colony formation, and was subsequently used as the primary growth and maintenance medium. Other media tested had varying results for cell growth and colony formation. The various other test media were able to sustain single cells and clusters of cells released from whole tissue. However, cells incubated in either B9638 or B9763 did not show appreciable growth resulting in colony formation or confluence (Table 1). After the initial characterization assays were completed, it was noted that MK ceils grew well in DMEM D2902 with 0.584 g/L L-glutamine, 1.0 g/L glucose, 0.004 g/L pyridoxine, supplemented with 10% newborn calf serum. Morphological differentiation between epithelial and fibroblast cells, viewed under a phase-contrast microscope, was not obvious until the cells had been in culture for at least 4 - 5 d. The MK cells had a polygonal shape, frequently appeared in small clusters surrounded by fibroblastoid cells during primary culture, and formed a confluent monolayer at approximately 8 - 1 4 d after subculture (Fig. 1). MK cells at higher passages (7-9) had an increase in size and an irregular shape. The cytoplasm became star-shaped and serrated at the periphery and developed cytoskeletal striations that resembled those of the dysplastic cells previously reported (Merchant, 1990) (Fig. 2).

MKE CELLS FROM FLORIDA MANATEE

FIG. 1. Confluent monolayer of MK cells. Cells are morphologicallyepithelial-like. Cells were photographed using a phase-contrast microscope after 9 d in culture, as described in the "Materials and Methods" section. Magnification: •

FIG. 2. MK cells with a dysplastic appearance photographed using a phase-contrast microscope. The MK cells at the fifth passage, which were 25-30 pan in diameter, had a serrated boundary and distinct cytoskeletal structures. Magnification: •

389

FIG. 3. MK cells stained with fluorescein-conjugated Mabs to keratin proteins, clone #8.13, demonstrate a pattern of perinuclear staining of cytokeratins. Cells were photographed using an epifluorescence microscope. Magnification: •

Indirect immunofluorescence staining assays revealed the reactivity of MK cells with antibodies to cytokeratin filaments. MK and SIG both had a diffuse cytoplasmic and perinuelear staining pattern (Fig. 3) when the antieytokeratin clone 8.13 was used. MK cells incubated with monoelonal antibodies (Mabs) to desmosomal cytokeratins revealed a staining pattern sinfilar to SIG (Stein et al., 1991, 1993). Pinpoint aggregates of FITC-labeled antibodies demonstrated areas of desmosomal eytokeratins located along cell-tocell gap-junction complexes in MK cells (Fig. 4). These cells also had a negative reactivity to Mabs against vimentin proteins. No fluorescence was noted in the cells that were not exposed to primary or secondary Mabs alone. Therefore, fluorescence was a result of antibody specificity and not of autofluorescenee of MK cells. MK cells at the second passage averaged a PDT of 1.9 d and reached a saturation density of 4 X 104 eells/em2 after 13 d of growth. The exponential growth period of a primary culture of MK cells had an average PDT of 24 h, whereas passages one through three increased to 48 h. MK cells grew rapidly as primary cultures, but slowed with subsequent passages. By the fifth passage, the PDT was 3.8 d (Fig. 5). Based on PD attained during an 8-d culture period, the seeding density for MK cells was optimal at approximately 8 • 103 cells/cm2 (Fig. 6). However, MK cells were subeuhured routinely at a seeding density of 5 • 103 cells/cmz, which resulted in PD comparable to those of the seeding-density-determination assay. MK cells seeded at 1 x 104 cells/eraz reached confluence rapidly and did not undergo a PD. Cells seeded at 4 X 104 cell/cm2 attached to the plate, but appeared vacuolated, had roughened cytoplasmic edges, and were not viable by day 8 of culture. Upon inspection, the seeding efficiency of MK cells at the second

390

SWEAT ET AL.

FIG. 4. MK cells stained with fluorescein-conjugated Mabs to desmosomal cytokeratins. Intense punctate loci of immunofluorescence staining reveal desmosomes located along the cytoplasmic: boundary of adjacent cells (arrows). MK ceils were grown to confluence on glass slides betore being stained and photographed using an epifluoreseenee microscope. Magnification: •

and fifth passages was not significantly difl>rent. However, because of the small number of data points, the minimum detectable diffel~ ence was 12.5% with a Power of 0.8 and alpha of 0.05, as determined by two-way analysis of variance (Sigma slat). The mean percentage of the attached cells for 30, 120, and 300 rain after plating

was and 88, 90, and 95, and 90, 97, and 97 for the second and fifth passages, respectively (Fig 7). MK cells subcuhured two or more times and seeded at relatively low densities (less than 1 • 1@ cells/cm 2) failed to produce confluent monolayers. Wells containing a single cell from the second or third passages did not produce colonies after 2 wk. Primary cultures initiated at a low seeding density produced large colonies (estimated at over 3 • 103 cells/colony) that had epithelial cell morphology. The appearance and location of single cells were noted in primary cultures. These gave rise to colonies of cells that had a similar morphology and were clearly demarcated from the surrounding colonies. Using a single-cell pick technique, cells from these isolated colonies were used for clonal expansion with limited success. A small number of cells were grown, but a confluent colony was not obtained from the single-cell cultures after 2 wk. Based on the difference in appearance of the two cell types after 10 d of a primary culture, fibroblast-cell contamination was ruled out. Conditioned media (0.45 ~m filtered-sterilized medium from confluent third-passage MK ceils) resulted in an increase in both MK and manatee lung cell primary cultures (data not shown). MK cells grown in a soft agar medium appeared to be viable for up to 5 d in cuhure. After 5 d, the cells began to appear vacuolated, and the cytoplasm was opaque. The cells did not form colonies, and appeared to be lysed by day 12 of culture. Data in Fig. 8 indicates that MK cells were serum dependent for growth in culture. There was a noticeable decrease in cell viability when the concentration of BFS was reduced from 10 to 5%. Cells grown in reduced-serum media had attenuated cytoplasmic processes, increased vacuolization, and roughened cytoplasmic-membrane edges (Fig. 9). MK cells appeared to be permissive for various human viral pathogens based on the presence of CPE and HA of guinea-pig rbc. Table 2 shows the temporal distritmtion of CPE in MK and BGM cultures. On day 1 postinoculation (PI), MK cells inoculated with herpes-I virus showed indications of cytoplasmic swelling (cyto-

FIG. 5. PDT of MK cells. The number of PDs, which occurred during culture periods ranging from 7 to 10 d, was used to determine the PDT of MK cells. Error bars represent standard deviations of the mean.

MKE CELLS FROM FLORIDA MANATEE

391

FIG. 6. Seeding-density-determination assay for MK cells. Cells were plated at various densities to determine which one ~sulted in an optimum growth response. This assay showed that 8 • ]03 cells/cm 2 produced the greatest number of PDs, representing the optimum seeding density.

FIG. 7. Seeding-efficiency assay for MK cells. MK cells were seeded at 1.1 • 104 cells/cm 2. Nonadherent cells were harvested at 30, 120, and 300 min after seeding, and counted. Error bars represent standard deviations of the mean.

plasmic rounding and enlargement), and some cells had become detached from the plate. On day 4 PI, MK ceils inoculated with eco-9 virus revealed a slight degree of cytoplasmic vacuolation (these cells had a foamy appearance). Those inoculated with influenza A did not appear to have a clear indication of CPE, while approximately 75% of those with influenza B had attenuated cytoplasmic processes. Reo virus appeared to result in a mild CPE on day 4. Herpes-1 virus appeared to have infected approximately 95% of the ceils by day 4 in cuhure. On day 5 PI, cells inoculated with coxsackie B5 virus appeared swollen with a vacuolated cytoplasm.

A small number of cells inoculated with parainfluenza virus appeared swollen with a slight CPE, while nearly 100% of those infected with reo virus had a reduced cell-body size with cytoplasmic processes that were detached at the distal end. By day 5 PI, cultures inoculated with herpes-1 and reo showed complete CPE (100% of the cells were either swollen and detached in the case of herpes-} or vacuolated and partially detached in the case of reo). On day 8 PI, nearly all the ceils inoculated with coxsackie B5 had a swollen cytoplasm, and approximately 75% of those infected with the parainfluenza virus indicated a CPE. Interestingly, despite the lack of

392

SWEAT ET AL.

FIG. 8. Serum-dependency assay fur MK cells. MK cells were cultured in Dulbecco modified Eagle medium with various concentrations of BFS to determine serum dependence. This graph shows that MK cells cultured in less than 10% BFS had a dramatic decrease in growth and viability. Error bars represent standard deviations of the mean.

DISCUSSION

FIG. 9. MK cells growing in DMEM with a reduced amount (2.5%) of BFS. This phase-contrast photograph shows cells that had attenuated cytoplasmic boundaries and increased vacuolization. Magnification: •

obvious CPE in the cultures inoculated with influenza A, HA with 0.5% guinea-pig rbc was noted on day 8 PI. MK cells inoculated with parainfluenza and influenza B hemagglutinated guinea-pig rbcs, while the negative controls (RSV-infected cultures and culture medium alone) showed no, or a small amount of, background/nonspecific HA.

In this study, we defined the requirements for isolating, characterizing, and growing an epithelial cell line derived from MK. Serial cultivation of epithelial and fibroblast cells from various mammalian sources has been well documented, with results similar to those obtained in this study (Rheinwald and Green, 1975; Hennings et al., 1980; De Jong et al., 1993; Carvan et al., 1994). Kidneys of most mammals contain difterent cell types, including modified epithelial ceils: podocytes, simple squamous epithelial cells, simple cuboidal epithelial cells, and fibroblastoid cells (Ross et al., 1985). In this study, no distinction was made as to the type of epithelial cell that was isolated from MK. The type of epithelial cell isolated for the development of this cell line was not investigated. The cells isolated and cultured fl'om MK tissue had a morphological appearance similar to epithelial cell strains. The polygonal cells that formed confluent monolayers had a positive reactivity to Mabs against keratin proteins, which is an established characteristic of epithelial cells (Moll et al., 1981; Lane, 1982; Adams and Watt, 1988). There are human, monkey, cow, dog, rabbit, and hamster cell lines that react with antibodies to vimentin proteins (Franke et al., 1979). In this study, the HSF and SIG cell lines were reactive to Mobs against vimentin and cytokeratin expression, respectively', and served as appropriate controls. 3"his suggests that the cells that were cultured from MK tissue and possessed the above characteristics were epithelial-like. Therefore, these cells are now referred to as manatee kidney epithelial (MKE) cells. PDTs can vary from 12 to 15 h in neoplastic cells and can be as long as 72 h in noncontinuous, finite cell lines (Freshney, 1992, 1994). Carvan et al. (1994) reported a bottle-nosed dolphin kidney cell line with a PDT of 1.2 d. MKE cells were determined to be a rapidly growing cell strain during primary culture and the first three passages. These cells appeared to grow well in DMEM Ham's F-12 supplemented with 10% BFS. DMEM Ham's F-12 has a greater

393

MKE CELLS FROM FLORIDA MANATEE TABLE 2

MKE CELLS WERE INOCULATED WITH DIFFERENT HUMAN PATHOGENS TO DETERMINE IF THEY WOULD SUPPORT REPLICATION. THE CPE OF INOCULATION WITH THESE VIRUSES WAS DOCUMENTED OVER AN 8-D PERIOD No. of darp PI Virus 0 1 2 3 4 5 6 7 8 Eco-9 Influenza B Reo Herpes-1 Cox-B5 Influenza A Parainflu RSV

. . . . . . .

. . .

. . .

. . .

+

++

.

+++

.

.

. .

. .

.

+ +++ + +++

. .

.

. .

.

.

+++ ++++ ++++ + .

. .

.

.

.

+++ 0 0 +

+++ 0 0 ++

+

+

x 0 0 +++ • +++x

.

Cells showing no CPE (-), cells showing mild CPE (+), 25% cells showing CPE (+ +), 75%-95% cells showing CPE (+ + +), 100% showing CPE (+ + + +), no viable cells (0), hemagglutination (•

variety of cofactors, vitamins, inorganic salts, and amino acids than the other test media. Growth of MK cells in DMEM Ham's F-12 is consistent with that of other cell-culture systems with respect to rapid colony formation and confluent monolayers (Lane, 1982; De Jong et al., 1993; Stein et al., 1993; Carvan et al., 1994). Cellular senescence and reduced proliferative capacity have been well described in regenerative diploid cells derived fi'om somatic tissues (Hayflick and Moorhead, 1961). MK cells were morphologically similar to epithelial cells isolated fi'om other species and underwent distinct changes consistent with a finite cell-line PDT, seeding-density requirements, and seeding efficiency. Furthermore, these changes were consistent with those reported elsewhere (Stanulis-Praeger, 1987; Carvan et al., 1994) and were not indicative of pathological changes. A rapidly expanding population of MKE cells was most easily obtained from a primary culture or first passage, but seemed to require the presence of fibroblast cells in later passages at low seeding densities. Growth factors produced by cells in culture are important mediators of cellular signaling and are required to maintain homeostasis (Freshney, 1994). It appeared that fibroblasts were important during cloning to aid in the growth of MKE cells seeded at low density. It was noted that MKE cells, seeded at low densities, reached confluency in less time when fibroblast-like cells or conditioned medium collected from cultured manatee fibroblast-like ceils were also present. This is consistent with the findings of studies comparing the effects of conditioned media from epithelial and from fibroblast cells (Stanulis-Praeger, 1987). MKE cells did not appear to possess the properties of transformed cells. They had a limited passage number before they reached senescence, demonstrated cell-contact inhibition, did not form colonies in agar, and were serum dependent. Reducing the serum in cultured cells results in prolonged periods in a G1 phase, and their subsequent reentry into the S phase in which DNA synthesis occurs (Alberts et al., 1994). The mechanism responsible for the quiescence and death of MK cells in medium with a reduced amount of sermn was not determined. However, the response of MK cells to serum supplementation after serum deprivation and the effects of growth in manatee fetal calf serum are potential subjects for future studies. Unless a transformed cell can be developed, MKE cell availability will limit its use for assays that require serially passaged cell cuhures. Primary cultures started from newborn or stillborn manatee calves, grown subsequent to this study, had a greater potential for rapidly expanding colonies during the first through third

passages (data not shown). These ceils have been frozen for further characterization studies. MKE cells appeared to support a permissive infection with human viruses, as indicated by CPE in the case of herpes-l, eco-9, influenza B, reo, coxsackie B5, and parainfluenza virus, and the HA of rbc as noted in cultures infected with influenza A and B virus. Inoculation of MKE cells with RSV, however, did not result in a productive cytopathic infection. This exercise was not intended to characterize the entire spectrum of permissive infections supported by cultured manatee tissue. However, this does highlight the possible use of such a cell line in the determination of zoonotic pathogens. To date, a continuous or transformed cell line from manatee tissue has not been reported, but would be extremely useful during long-term studies where viral isolates must be passed several times before a CPE is present. There are ongoing collaborative-research initiatives with respect to manatee biology, including serology, histology, pathology, basal metabolism, genetics, neurophysiology, parasitology, immunology, and behavior. Much remains to be learned about the effects of putative microbial pathogens on manatees in the wild or in captivity. Necropsies of several manatees have revealed gross and microscopic lesions compatible with viral infection (G. D. Bossart and S. Wright, pers. comm.). Serological studies have reported the exposure of manatees to morbillivirus (Duignan et al., 1995), but no virus has been isolated and characterized from manatee tissue. Wart-like lesions have been confirmed to be the resuh of infection with a papilloma virus by in situ DNA hybridization with a human-papilloma probe (G. D. Bossart, in prep.). Additionally, the cellular and molecular mechanisms in response to brevetoxicosis from exposure to the red-tide organism, Gynodidium brevi, in manatees are not clearly understood. Therefore, characterization of the MKE and other manatee-derived cell lines will aid in further research involving putative-virus isolation and characterization, interspecies cross-reactivity of immunological markers, ongoing cytogenetics, effects of growth factors in wound repair, and evaluation of biochemical and physiological responses to biological and industrial toxins. ACKNOWLEDGMENTS This study was supported by the Florida Fish and Wildlife Conservation Commission, MMPL, under permit PRT-773494. FDEP field-station biologists and MMPL staff had an integral part in the collection and transport of fresh carcasses used during this study. We are grateful to Dr. Mike Walsh at

394

SWEAT ET AL.

SeaWorld of Florida for providing tissue samples. Our thanks are due to Dr. My Lien Dao for providing help with immunocytochemical techniques. Drs. Robert Burghardt and Bill Ross provided cell lines used as controls during this project. Dr. Lillian Stark of the Tampa Branch Diagnostic Laboratory kindly provided her expertise during the human-pathogen i~ffection studies. This manuscript is a partial report of the thesis requirement of J. M. S. for an M.S. in microbiology at the University of South Florida. REFERENCES Adams, J. C.; Watt, J. M. An unusual strain of human keratinocytes which do not stratify or undergo terminal differentiation in cuhure. J. Cell Biol. 107:1927-1938; 1988. Alberts, B.; Bray, D.; Lewis, J.; Raft, M.; Roberts, K.; Watson, J. D. The celldivision cycle. In: Molecular biology of the cell. 3rd ed. New York: Garland Publishing; 1994:864-865. Carvan, M. J.; Santostefano, M.; Safe, S.; Busbee, D. Characterization of a bottlenosed dolphin Tursiops truncatus kidney epithelial cell line. Mar. Mamm. Sci. 10:52-69; 1994. De Jong, P.; Van Sterkenburg, M. A.; Kempenaar, J. A.; Dijman, J. H.; Ponec, M. Serial culturing of human bronchial epithelial cells derived from biopsies. In Vitro Cell. Dev. Biol. 29A:379-387; 1993. Duignan, P. J.; House, C.; Walsh, M., et al. Morbillivirus infection in man~ atees. Mar. Mamm. Sci. 11(4):441~1~45; 1995. Franke, W. W.; Schmid, E.; Winter, S.; Osborn, M.; Weber, K. Widespread occurrence of intermediate-sized filaments of the vimentin-type in cultured cells from diverse vertebrates. Exp. Cell Res. 123:25-~6; 1979. Freshney, R. I. In: Culture of epithelial cells. New York: Wiley-Liss; 1992. Freshney, R. I. In: Culture of animal cells, a manual of basic technique. New York: Wiley-Liss; 1994. Harris, L. W.; Griffiths, J. B. Relative effects of cooling and warming rates on mammalian cells during the freeze-thaw cycle. Cryobiology 14:662-669; 1977. Hayrick, L.; Moorhead, P. S. The serial cultivation of human diploid cell strains. Exp. Cell Res. 25:585~521; 1961. Hennings, H.; Micheal, K.; Cheng, K.; Steinert, P.; Holln~ok, K.; Yuspa, S. Calcium regulation of growth and difterentiation of mouse epidermal cells in culture. Cell 19:245-254; 1980.

Hoffman, A. G.; Burghardt, R. C.; Tilley, R.; Auersperg, N. An in vitro model of ovarian epithelial carcinogenesis and changes in cell-cell communication and adhesion occurring during neoplastic progression. Int. J. Cancer 54:828-838; 1993. Hsuing, G. D.; Fong, C. K. Y.; Landry, M. L. Hemagglutination and hemagglutination-inhibition test. In: Hsuing's diagnostic virology. 4th ed. London: Yale University Press; 1994:69-75. Jones, T. C.; Hunt, R. D. In: Veterinary pathology. 5th ed. Philadelphia: Lea and Febiger; 1984:1270-1274. Lane, E. B. Monoclonal antibodies provide specific intermolecular markers for the study of epithelial tonofilament organization. J. Cell Biol. 92:665-673; 1982. Leibo, S. P.; Mazur, P. The roll of cooling rates in low temperature preservations. Cryobiology 8:447452; 1971. Lindgrend, A.; Westermark, B.; Ponten, J. Serum stimulation of stationary human glia and glioma cells in culture. Exp. Cell Res. 95:311-319; 1975. McKeehan, W. L.; McKeehan, K. A.; Hammond, S. L.; Ham, R. G. Improved medium for clonal growth of human diploid ceils and low concentration of serum protein. In Vitro 13:399-416; 1977. Merchant, D. J. Terminally differentiating epithelial tissues in primary explant cuhure: a model of growth and development. In Vitro Cell. Dev. Biol. 26:543-553; 1990. Moll, R.; Franke, W. W.; Schiller, D. L. The catalogue of human cytokeratins: patterns of expression in normal epithelia, tumors and cultured cells. Cell 31:11-24; 1981. Rheinwald, J. G.; Green, H. Serial cultivation of strains of human epidermal keratinocytes: the formation of keratinizing colonies from single cells. Cell 6:331-344; 1975. Ross, M. H.; Reith, E. J. In: Histology, a text and atlas. New York: Harper and Row; 1985:25-35. Stanulis-Praeger, B. Cellular senescence revisited: a review. Mech. Aging Dev. 38:148; 1987. Stein, L. S.; Stein, D. W. J.; Echols, J.; Burghardt, R. C. Concomitant alterations of desmosomes, adhesiveness, and diffusion through gap junction channels in a rat ovarian transformation model system. Exp. Cell Res. 207:19-32; 1993. Stein, L. S.; Stoica, G.; Tilley, R.; Burghardt, R. C. Rat ovarian granulosa cell cuhure: a model system tot the study of cell-cell comnmnication during muhi-step transformation. Cancer Res. 51:696-706; 1991. Waymouth, C. rib disaggregate or not to disaggregate. Injury and cell disaggregation, transient or permanent? In Vitro 10:97-111; 1974.