Reporter Gene Expression in Fish Following CutaneousInfection with Pantropic Retroviral Vectors

T.A. Paul,1 J.C. Burns,2 H. Shike,2 R. Getchell,1 P.R. Bowser,1 K.E. Whitlock,3 and J.W. Casey1,*

1Department of Microbiology and Immunology, Cornell University, Ithaca, NY 14853, U.S.A.2Department of Pediatrics, UCSD School of Medicine, La Jolla, CA 92093-0830, U.S.A.3Department of Molecular Biology and Genetics, Cornell University, Ithaca, NY 14853, U.S.A.

Abstract: A central issue in gene delivery systems is choosing promoters that will direct defined and sustainable

levels of gene expression. Pantropic retroviral vectors provide a means to insert genes into either somatic or

germline cells. In this study, we focused on somatic cell infection by evaluating the activity of 3 promoters

inserted by vectors into fish cell lines and fish skin using pantropic retroviruses. In bluegill and zebrafish cell

lines, the highest levels of luciferase expression were observed from the 58 murine leukemia virus long terminal

repeat of the retroviral vector. The Rous sarcoma virus long terminal repeat and cytomegalovirus early pro-

moter, as internal promoters, generated lower levels of luciferase. Luciferase reporter vectors infected zebrafish

skin, as measured by the presence of viral DNA, and expressed luciferase. We infected developing walleye

dermal sarcomas with retroviral vectors to provide an environment with enhanced cell proliferation, a condi-

tion necessary for integration of the provirus into the host genome. We demonstrated a 4-fold to 7-fold increase

in luciferase gene expression in tumor tissue over infections in normal walleye skin.

Key words: retroviral vector, promoter, experimental infection, zebrafish, walleye dermal sarcoma virus.

INTRODUCTION

Retrovirus-associated neoplastic diseases in walleyes (Sti-

zostedion vitreum) provide a dynamic model to investigate

mechanisms of oncogenesis and tumor progression because

of the disease’s unique property of developing and regress-

ing seasonally (Bowser et al., 1988). Degenerate reverse

transcriptase polymerase chain reaction (RT-PCR) amplifi-

cation strategies identified 3 complex walleye retroviruses:

walleye dermal sarcoma virus (WDSV); walleye epidermal

hyperplasia viruses type 1 (WEHV 1); and type 2 (WEHV 2)

(Martineau et al., 1992; LaPierre et al., 1998a). All these

retroviruses contain two 38 accessory genes (orf A and orf B)

that share homology with the cell-cycle regulator cyclin D

(LaPierre et al., 1998b). These cyclin D–like transcripts are

thought to play a critical role in tumorigenesis because they

are expressed exclusively in developing tumors (Quacken-

bush et al., 1997). Further, it was shown recently that cell-

type-specific expression of WDSV orf A results in the de-

velopment of hyperplastic lesions in transgenic mice (Lair-

more et al., 2000). The mechanisms by which WDSV,

WEHV 1, and WEHV 2 induce disease remain unknown.

Characterization of the walleye retroviruses and their

putative oncogenes has been impeded by the lack of a cell-

culture system for viral propagation and difficulty working

with captive wild fish. Zebrafish (Danio rerio) are an attrac-

Received January 31, 2001; accepted March 30, 2001.

*Corresponding author. Current address: Department of Microbiology and Immunology,

C5145 VMC, Cornell University, Ithaca, NY 14853, U.S.A.; telephone 607-253-3579; fax

607-253-3384, e-mail jwc3@cornell.edu

Mar. Biotechnol. 3, S81–S87, 2001DOI: 10.1007/s10126-001-0029-y

© 2001 Springer-Verlag New York Inc.

tive alternative system in which to study oncogenesis be-

cause of their established genetics and development of

transgenic technology (Amsterdam et al., 1999; Linney et

al., 1999). Also, they develop tumors both naturally and

after experimental induction (Beckwith et al., 2000). Pan-

tropic retroviral vectors in which oncogenes can be ex-

pressed via experimental infection are being developed.

These vectors are replication-deficient murine leukemia vi-

rus (MLV)–based retroviral vectors pseudotyped with ve-

sicular stomatitis virus (VSV) G protein, which confirs a

broad host range. To design optimal vectors, we studied

promoter activity in the context of retroviral vector infec-

tion of cells from zebrafish and bluegill (Lepomis macrochi-

rus). We explored the feasibility of inserting genes into ze-

brafish with vectors by experimentally infecting their skin.

We also tested retroviral vector infection in developing tu-

mors in walleyes.

MATERIALS AND METHODS

Plasmid and Vector Preparation

The plasmids pLLRNL (for long terminal repeat [LTR]–

Luciferase–Rous sarcoma virus [RSV] LTR–Neo-LTR) and

pLNRLL (LTR-Neo–RSV LTR–Luciferase-LTR) were previ-

ously described (Boulo et al., 2000) (Figure 1). The plasmid

pLNCLL (LTR-Neo-CMV-Luciferase-LTR) was con-

structed by cloning the cytomegalovirus (CMV) promoter

from pcDNA/HisMax (Invitrogen) into pLN(MCS)LL

(LTR-Neo–Multiple Cloning Site–Luciferase-LTR) using

HindIII and BglII sites (Figure 1). Concentrated pantropic

retroviral vector stocks LLRNL, LNRLL, and LNCLL were

prepared as previously described (Burns et al., 1993; Yee et

al., 1994). Virus titers were determined by infecting the

208F rat fibroblast cell line in the presence of polybrene (8

µg/ml; Sigma) and selecting with G418 (400 µg/ml; Gene-

ticin, Gibco-BRL). Virus titers were determined by the

number of colony-forming units (per milliliters).

Cell Lines

ZF4 is a stable cell line derived from zebrafish embryos as

described (ATCC CRL-2050) (Driever and Rangini, 1993).

BF-2 cells are a fibroblast cell line isolated from bluegill

sunfish (ATCC CCL-91) (Wolf et al., 1966). ZF4 and BF-2

cells were grown at room temperature in L-15 medium with

10% fetal bovine serum and penicillin/streptomycin (pen/

strep) (100 U/ml and 100 µg/ml). Human transformed em-

bryonic kidney cell line 293 expressing Moloney MLV gag

and pol (Burns et al., 1993) and 208F cells were grown in

Dulbecco’s modified Eagle’s medium (DMEM) with high

glucose supplemented with 10% fetal calf serum, 2 mM

L-glutamine, and pen/strep (100 U/ml and 100 µg/ml) at

37°C with 10% CO2.

Cell-Culture Experiments

ZF4 cells were seeded in 3 replicate wells at 1.5 × 105 cells

per well in 24-well plates 1 day prior to infection. The cells

were infected with 1 × 104 cfu of each vector (MOI = 0.1)

in 200-µl DMEM in the presence of polybene (2 µg/ml;

Sigma). An additional 300 µl of DMEM was added after 1

hour, and the medium was changed after 7 hours. Cells

were lysed 72 hours after infection in 115-µl lysis buffer

(Promega). Luciferase expression was assayed on 100 µl of

lysate with 100 µl of luciferase substrate (Promega) using a

luminometer (Turner Designs). Soluble protein concentra-

tion was determined by Bradford Assay (Bio-Rad) and cal-

culated using a bovine serum albumin standard curve. Data

were reported as light units (LU) per milligram of protein.

Similar experiments were carried out with BF-2 cells.

Cutaneous Infection Experiments

Vector stocks were diluted in 0.1× Hanks balanced salt so-

lution to a final titer of 6 × 105 cfu/ml. Adult zebrafish were

provided by K. Whitlock (Cornell University). Cutaneous

infections of zebrafish were adapted from previously de-

scribed WDSV infection of walleyes (Bowser et al., 1996).

Figure 1. Genetic organization of pantropic retroviral vectors.

Promoters are designated by arrows. LTR indicates long terminal

repeat; RSV LTR, Rous sarcoma virus LTR; Neo, neomycin phos-

photransferase; CMV, immediate early promoter of human cyto-

megalovirus.

S82 T.A. Paul et al.

Zebrafish were placed on separate wet cotton cloths, and a

small area, approximately 5 × 20 mm, on the right side was

rubbed 5 times with a cotton swab. Then 100 µl of vector

mixed with polybrene (2 µg/ml) was pipetted onto the

abraded area and distributed by swabbing. This entire pro-

cedure took less than 1 minute for each fish. Five zebrafish

were infected with each vector and 2 uninfected fish served

as controls. Five days after infection the fish were eutha-

nized by an overdose of tricaine methanesulfonate (MS-

222; Sigma), and the infected skin and underlying muscle

were removed. Infected tissue was homogenized on ice in a

glass grinder containing 160 µl of luciferase lysis buffer

(Promega), centrifuged 15,000 g for 20 seconds at 4°C, and

100 µl of supernatant was combined with 100 µl of lucifer-

ase substrate to assay luciferase in a luminometer (Ber-

thold). Soluble protein concentrations were determined us-

ing the Bradford assay. DNA was isolated from remaining

tissue and supernatant from infected and uninfected

samples using a QIAGEN Blood Kit. DNA from LLRNL-

infected and LNRLL-infected ZF4 cells was used as a posi-

tive control. One microgram of DNA was used for poly-

merase chain reaction (PCR) with luciferase specific prim-

ers, forward 58-TGGGCTCACTGAGACTACATCAG-38 and reverse

58-AACTGGCGGACTTCAGAGACT-38 under the following condi-

tions: 94°C for 5 minutes (1 cycle), 94°C for 30 seconds,

60°C for 30 seconds, 72°C for 1 minute (35 cycles), and

72°C for 10 minutes (1 cycle). PCR reactions were run on a

2% agarose gel (Seakem LE; BMA) and stained with ethid-

ium bromide (Sigma).

Walleye Infections

Young-of-the-year walleyes, approximately 14 weeks old,

the progeny of feral Oneida Lake walleyes, were obtained

from New York State Department of Environmental Con-

servation. They had been trained at the hatchery to accept a

pellet ration, a feeding practice we continued. Young wall-

eyes were maintained in a 625-L fiberglass tank (Frigid

Units, Inc.) containing aerated 15°C dechlorinated munici-

pal water.

Dermal sarcoma tissue was obtained from adult wall-

eyes by cutting the superficial nodules from the skin. A

cell-free tumor homogenate was prepared by thawing

WDSV tissue, homogenizing it with a glass grinder, and

suspending it in a 1:3 (weight-to-volume) dilution in sterile

0.01 M phosphate-buffered saline (pH 7.2). The suspension

was sonicated for 1 minute at 20 kHz and 28 W while in an

ice bath. Then, it was briefly centrifuged at 1200 g and

passed through a 0.45-µm filter. Approximately 100 µl of

filtrate was applied topically with a cotton swab along the

right lateral line of 5 walleye fingerlings. Six weeks after

infection fish were anesthetized with MS-222, and cross-

sections of tissue were collected from the epaxial muscula-

ture in the inoculation area. Samples were fixed in neutral

buffered formalin, sectioned, and stained in hematoxylin

and eosin to examine the early stages of neoplasia (Martin-

eau et al., 1990).

Two WDSV-infected walleyes were coinfected in the

region of developing tumors by topical application of ap-

proximately 100 µl of LLRNL (1 × 107 cfu/ml) in polybrene

(2 µg/ml). One WDSV-uninfected walleye was infected

along the right lateral line with LLRNL by similar methods

and reared in a separate tank. After 7 weeks, fish were

euthanized by an overdose of MS-222, and tumors on coin-

fected fish and tissue from controls were similarly collected,

homogenized, and analyzed by luciferase assay. DNA was

prepared for PCR from normal walleye skin, LLRNL-

infected normal walleye skin, and LLRNL-infected walleye

tumors. DNA from LLRNL-infected ZF4 cells was used as a

positive control.

RESULTS

Promoter Activity in Zebrafish (ZF4) and BluegillSunfish (BF-2) Cell Lines

The activities of several promoters were tested in the con-

text of pantropic retroviral infection of an embryonic ze-

brafish cell line (ZF4) and a fibroblastic cell line from blue-

gill sunfish (BF-2). Luciferase reporter assays in ZF4 cells

demonstrated that the murine leukemia virus LTR had 1.5-

fold higher activity than a Rous sarcoma virus LTR internal

promoter, and 10-fold more than a cytomegalovirus inter-

nal promoter (Figure 2). In BF-2 cells, the MLV promoter

had a 7-fold higher luciferase expression than the RSV pro-

moter and about 40-fold higher than the CMV promoter

(Figure 2).

Vector Skin Infection of Zebrafish

To establish the feasibility of transducing adult zebrafish in

vivo, the ability of retroviral vectors to infect adult zebrafish

skin was tested using the luciferase reporter vectors. PCR

for provirus demonstrated successful infection of 5 out of 5

fish with the LLRNL vector (Figure 3, B) and 3 out of 5 fish

with the LNRLL vectors (Figure 3, C). Low levels of lucif-

Gene Expression After Retroviral Infection in Fish S83

erase expression were detected from the MLV promoter in

LLRNL-infected fish (169 LU/mg of protein); however, ex-

pression from the RSV LTR internal promoter (102 LU/mg

of protein) could not be differentiated from that in control

tissue (95 LU/mg of protein) (Figure 3, A). This result is

consistent with the stronger promoter function of MLV

LTR versus the RSV LTR as demonstrated in ZF4 cells.

Vector Infection of Developing Walleye Tumors

A limitation of pantropic retroviral vectors is their require-

ment for dividing cells in order to integrate into the target

cell genome. Owing to the low numbers of proliferating

cells in zebrafish skin, low levels of in vivo infection were

expected. To provide an environment of sustained cell di-

vision in which infection efficiency could be increased and

the luciferase signal amplified, retroviral vectors were in-

fected onto developing tumors in walleyes.

Histological examination of tissue from the 6-week

WDSV-infected walleye at the time of LLRNL vector infec-

tion revealed the development of the early stages of dermal

sarcoma. Consistent with previously described histological

criteria, fibroblast-like cells appeared, originating from the

external surface of the scales separated by a moderate

amount of collagen (Figure 4, B). Macroscopic cutaneous

tumors collected 13 weeks after WDSV infection were

highly cellular or densely fibrous with no signs of metastasis

(Martineau et al., 1990) (Figure 4, A and C).

Tumors collected on 2 separate fish with skin coin-

fected with LLRNL and WDSV showed 4-fold and 7-fold

(422 and 712 LU/mg of protein) increases in luciferase ex-

pression compared with LLRNL-infected normal walleye

skin (139 LU/mg of protein) (Figure 5, A). PCR for provirus

demonstrated successful infection of both tumors and LL-

RNL-infected skin of normal walleyes (Figure 5, B).

DISCUSSION

We assessed the potential for utilizing experimental infec-

tion mediated by pantropic retroviral vectors to introduce

genes into adult zebrafish. We successfully transduced su-

perficial tissue or cutaneous tissue in 8 out of 10 zebrafish

and measured luciferase expression from the MLV LTR

promoter 5 days after infection. Although we demonstrated

Figure 2. Luciferase activity in transduced fish cell lines. The ze-

brafish embryonic cell line (ZF4) (hatched bars) and bluegill sun-

fish fibroblastic cell line (BF-2) (shaded bars) were infected with

the retroviral vectors LLRNL, LNRLL, or LNCLL for 7 hours in the

presence of polybrene and assayed for luciferase expression 72

hours later. Luciferase activity is reported in light units per milli-

gram of protein. Bars denote 1 SD.

Figure 3. Pantropic retroviral vector infection of adult zebrafish

skin. Retroviral vectors LLRNL or LNRLL were applied topically to

the skin of 5 adult zebrafish. A: Five days after infection, skin

sections were homogenized and assayed for luciferase expression.

Graph represents the average and 1 SD of light units per milligram

of protein from 5 separately infected fish for each vector and 2

controls. B: Provirus was detected by PCR in LLRNL-infected

zebrafish skin. C: In LNRLL-infected zebrafish skin. Lane 1, un-

infected zebrafish skin; lanes 2–6, vector-infected zebrafish skin;

lane 7, LLRNL-infected or LNRLL-infected ZF4 cells.

S84 T.A. Paul et al.

that the proviral genome was present, future work will need

to examine quantitative levels of infection.

Initially, we obtained relatively low levels of luciferase

expression in zebrafish infections, so we chose an alternative

approach to amplify the reporter signal. A limitation to

successful integratation of MLV-based vectors into the host

genome is the requirement for cell division (Miller et al.,

1990). By infecting WDSV-induced tumors, we provided an

environment with substantial cell proliferation, which

would likely enhance infection and provirus integration.

Proliferation of these tumor cells was predicted to result in

higher luciferase activity, which was, indeed, observed.

We assayed the activity of several promoters in the

context of retroviral vector infection of zebrafish and blue-

gill cells in cell culture. The MLV LTR had a higher level of

expression than the RSV LTR or the CMV internal pro-

moter. Expression levels in cell culture correlated with the

trends seen in infected skin of zebrafish.

Several factors influence the choice of promoters to

mediate transgene expression. Interference from transcrip-

tional signals in the flanking retroviral LTRs may affect the

expression and regulation of internal regulatory sequences

in a retroviral vector. Promoters derived from nonpiscine

species may also contain elements that are not optimally

recognized by zebrafish (Higashijima et al., 1997). There

also has been concern for the long-term expression of trans-

genes due to host-silencing mechanisms (Amsterdam et al.,

1995), which may be resolved using piscine promoters with

their lower GC content, in conjunction with strong insula-

tor elements (Gibbs and Schmale, 2000). Interestingly, we

were able to detect expression up to 7 weeks after experi-

mental infection of early-stage tumors in walleye from the

MLV LTR.

Pantropic retroviral vectors will be a useful tool to

study the function of putative oncogenes from newly dis-

covered piscine retroviruses, and to verify the function of

established oncogenes, using either a transgenic or somatic

cell approach in fish models.

ACKNOWLEDGMENTS

We thank Alan Eaglesham for a critical review of this manu-

script, Mark Roberson for assistance with luciferase assays,

and Shelley Bakalis for technical assistance with retroviral

vectors. T.A.P. was supported by National Institutes of

Health grant 5T32CA09682.

Figure 4. Dermal sarcoma in walleyes at early and late stages of

development. A: Walleye dermal sarcoma virus (WDSV) applied

topically to young walleyes causes dermal sarcoma approximately

13 weeks after infection. B: Photomicrograph of early develop-

mental stages of experimentally transmitted walleye dermal sar-

coma (arrows) at 6 weeks after infection. C: At 13 weeks after

infection. Tumors appear highly cellular or densely fibrous and do

not appear to metastasize. S indicates scales.

Gene Expression After Retroviral Infection in Fish S85

REFERENCES

Amsterdam, A., Lin, S., and Hopkins, N. (1995). The Aequorea

victoria green fluorescent protein can be used as a reporter in live

zebrafish embryos. Dev Biol 171:123–129.

Amsterdam, A., Burgess, S., Golling, G., Chen, W., Sun, Z.,

Townsend, K., Farrington, S., Haldi, M., and Hopkins, N. (1999).

A large-scale insertional mutagenesis screen in zebrafish. Genes

Dev 13:2713–2724.

Beckwith, L.G., Moore, J.L., Tsao-Wu, G.S., Harshbarger, J.C., and

Cheng, K.C. (2000). Ethylnitrosourea induces neoplasm in ze-

brafish. Lab Invest 80:379–385.

Boulo, V., Cadoret, J.P., Shike, H., Shimizu, C., Miyanohara, A.,

and Burns, J.C. (2000). Infection of cultured embryo cells of the

pacific osyter, Crassostrea gigas, by pantropic retroviral vectors. In

Vitro Cell Dev Biol Anim 36:395–399.

Bowser, P.R., Wolfe, M.J., Forney, J.L., and Wooster, G.A. (1988).

Seasonal prevalence of skin tumors from walleye (Stizostedion vit-

reum) from Oneida Lake, New York. J Wildl Dis 24:292–298.

Bowser, P.R., Wooster, G.A., Quackenbush, S.L., Casey, R.N., and

Casey, J.W. (1996). Comparison of fall and spring tumors as in-

ocula for experimental transmission of walleye dermal sarcoma. J

Aquat Anim Health 8:78–81.

Burns, J.C., Friedmann, T., Driever, W., Burrascano, M., and Yee,

J.K. (1993). Vesicular stomatitis virus G glycoprotein pseudotyped

retroviral vectors: concentration to very high titer and efficient

gene transfer into mammalian and nonmammalian cells. Proc Natl

Acad Sci USA 90:8033–8037.

Driever, W., and Rangini, Z. (1993). Characterization of a cell line

derived from zebrafish (Brachydanio rerio) embryos. In Vitro Cell

Dev Biol Anim 29A:749–754.

Gibbs, P.D.L., and Schmale, M.C. (2000). GFP as a genetic marker

scorable throughout the life cycle of transgenic zebrafish. Mar

Biotechnol 2:107–125.

Higashijima, S., Okamoto, H., Ueno, N., Hotta, Y., and Eguchi, G.

(1997). High-frequency generation of transgenic zebrafish which

reliably express GFP in whole muscles or the whole body using

promoters of zebrafish origin. Dev Biol 192:289–299.

Lairmore, M.D., Stanley, J.R., Weber, S.A., and Holzschu, D.L.

(2000). Squamous epithelial proliferation induced by walleye der-

mal sarcoma retrovirus cyclin in transgenic mice. Proc Natl Acad

Sci USA 97:6114–6119.

LaPierre, L.A., Holzschu, D.L., Wooster, G.A., Bowser, P.R., and

Casey, J.W. (1998a). Two closely related but distinct retroviruses

are associated with walleye discrete epidermal hyperplasia. J Virol

72:3484–3490.

Figure 5. Pantropic retroviral vector

infection of WDSV-induced tumors. A:

LLRNL was applied topically to walleye

dermal sarcomas induced by WDSV at

an early stage of development or to

normal walleye skin of the same age.

After 7 weeks, 2 LLRNL-infected dermal

sarcomas from 2 different fish and

LLRNL-infected normal tissue from 1

fish were assayed for luciferase activity.

B: PCR detection of provirus in

LLRNL-infected walleye normal skin

(lane 2) and WDSV-induced tumors

(lanes 3 and 4). Uninfected walleye skin

(lane 1) and LLRNL-infected ZF4 cells

(lane 5) were used as controls.

S86 T.A. Paul et al.

LaPierre, L.A., Casey, J.W., and Holzschu, D.L. (1998b). Walleye

retroviruses associated with skin tumors and hyperplasias encode

cyclin D homologs. J Virol 72:8765–8771.

Linney, E., Hardison, N.L., Lonze, B.E., Lyons, S., and DiNapoli, L.

(1999). Transgene expression in zebrafish: a comparison of retro-

viral-vector and DNA-injection approaches. Dev Biol 213:207–216.

Martineau, D., Bowser, P.R., Wooster, G.A., and Forney, J.L.

(1990). Histologic and ultrastructural studies of dermal sarcoma of

walleye (Pisces: Stizostedion vitreum). Vet Pathol 27:340–346.

Martineau, D., Bowser, P.R., Renshaw, R.R., and Casey, J.W.

(1992). Molecular characterization of a unique retrovirus associ-

ated with a fish tumor. J Virol 66:596–599.

Miller, D.G., Adam, M.A., and Miller, A.D. (1990). Gene transfer

by retroviral vecors occurs only in cells that are actively replicating

at the time of infection. Mol Cell Biol 10:4239–4242.

Quackenbush, S.L., Holzschu, D.L., Bowser, P.R., and Casey, J.W.

(1997). Transcriptional analysis of walleye dermal sarcoma virus

(WDSV). Virology 237:107–112.

Wolf, K., Gravell, M., and Malsberger, R.G. (1966). Lymphocystis

virus: isolation and propagation in centrarchid fish cell lines. Sci-

ence 151:1004–1005.

Yee, J.K., Friedmann, T., and Burns, J.C. (1994). Generation of

high-titer pseudotyped retroviral vectors with very broad host

range. Methods Cell Biol 43:99–112.

Gene Expression After Retroviral Infection in Fish S87