Expression of Human Glucose Transporter Type 1 and RatHexokinase Type II Complementary DNAs in RainbowTrout Embryos: Effects on Glucose Metabolism

Aleksei Krasnov,* Tiina I. Pitkanen, Mika Reinisalo, and Hannu Molsa

Institute of Applied Biotechnology, University of Kuopio, P.O. Box 1627, 70211 Kuopio, Finland

Abstract: Sugars are utilized poorly in fish mainly because of low rates of transport across plasma membrane

and phosphorylation. To evaluate whether it is possible to augment carbohydrate metabolism in fish using

heterologous genes, expression of human glucose transporter type 1 (hGLUT1) and rat hexokinase type II

(rHKII) complementary DNAs cloned with cytomegalovirus promoter was followed in rainbow trout embryos.

Both genes were transcribed. Hexokinase activity, undetectable in control, was found in transformed blastulas.

Increased rates of 14C-methylglucose uptake and sensitivity to cytochalasin B indicated the presence of facili-

tative hexose transport due to hGLUT1 expression. Effect of hGLUT1 on production of 14CO2 from glucose was

greater than that of rHKII. Coexpression of the genes did not increase the rate of glucose oxidation compared

with expression of hGLUT1 alone.

Key words: rainbow trout, glucose transporter, hexokinase, expression, metabolism

INTRODUCTION

Limited ability to utilize carbohydrates is a well-known

metabolic feature of carnivorous fish. Thus, for example,

fish react to oral administration or increased dietary levels

of carbohydrate with persistent hyperglycemia and abnor-

mal deposition of glycogen in liver; rates of glucose turn-

over determined in various teleost species are manyfold

lower than those in omnivorous animals and birds (for

review, see Cowey and Walton, 1989).

In animal tissues the transport of glucose across plasma

membrane and subsequent phosphorylation, catalyzed with

glucose transporters (GLUT) and hexokinase (HK), respec-

tively, are believed to play a rate-limiting role. The GLUT

family includes proteins with similar structure, containing

12 putative membrane-spanning domains, that are charac-

terized with different tissue specificity, kinetic parameters,

and sensitivity to hormones (Burant et al., 1991; Olson and

Pessin, 1996). The HK family in animals is represented with

four isoforms with high affinity (types I, II, III) and low

affinity to glucose (type IV, glucokinase) (Katzen et al.,

1968; Thelen and Wilson, 1991). To elucidate the role of

GLUTs and HKs in glucose metabolism, transgenic mice

were constructed, that overexpressed HKII in striated

muscles (Chang et al., 1996), HKIV (glucokinase) in livers

(Hariharan et al., 1997), yeast HK in pancreatic b cells

(Epstein et al., 1992), GLUT1 (Ren et al., 1993; Marshall et

al., 1993) as well as GLUT4 in skeletal muscles (Gibbs et al.,Received; accepted June 30, 1998.

*Corresponding author. Fax: +358-17-163148; e-mail: krasnov@uku.fi

Mar. Biotechnol. 1, 25–32, 1999

© 1999 Springer-Verlag New York Inc.

1995; Hansen et al., 1995; Tsao et al., 1996) and in whole

body (Voss-McCowan, 1994; Ikemoto et al., 1995). Though

the effects of HK and GLUT genes varied substantially in

magnitude, significant changes in parameters of carbohy-

drate metabolism have been revealed in all transgenic lines.

In contrast to higher vertebrates and lower fish (lam-

preys), facilitative glucose transport has not been found in

red blood cells of rainbow trout (Tse and Young, 1990) or

of some other teleost species (Bolis et al., 1971; Kim and

Isaacks, 1978; Ingermann et al., 1984; Tiihonen and Nikin-

maa, 1991). Furthermore, HK activity in salmonid fish is

the lowest among glycolytic enzymes (Walton and Cowey,

1982).

This study was aimed at examining the potential of

gene transfer to augment carbohydrate metabolism in fish.

Given that no piscine GLUT or HK genes have been cloned,

functionality of heterologous genes in fish cells was assessed.

Transient expression of human glucose transporter type 1

(hGLUT1) and rat hexokinase type II (rHKII) cDNAs, with

human cytomegalovirus (CMV) promoter and its effect on

glucose metabolism were followed in rainbow trout (On-

corhynchus mykiss) embryos.

MATERIALS AND METHODS

Constructs

RSV-luciferase was used to study temporal patterns of tran-

sient expression in rainbow trout embryos. Rat hexokinase

type II cDNA (rHKII) was provided by Dr. Wilson (Michi-

gan State University, U.S.A.) as a 3.64-kb fragment in

pUC18. hGLUT1 cDNA (2.47 kb) in Bluescript KS− back-

bone was received from Dr. Soeller (Pfizer Central Re-

search, Groton, U.S.A.). The rHKII fragment was digested

with EcoRI and ligated to partially digested expression vec-

tor pRc/CMV (Invitrogen). The hGLUT1 cDNA fragment

was digested with BamHI, blunt-ended, and ligated to

HindIII-digested, blunt-ended pRc/CMV vector. Con-

structs of CMV-rHKII (5.61 kb) and CMV-hGLUT1 (3.66

kb) were linearized with NruI–SmaI and NruI–HaeII diges-

tion, respectively. For delivery into rainbow trout embryos,

RSV-luciferase was used in circular form; CMV-rHKII and

CMV-hGLUT1 were applied as a mixture of linear and

circular forms.

Microinjections

From 40 to 80 pg of DNA was microinjected into fertilized

rainbow trout eggs as described elsewhere (Pitkanen et al.,

1998). A total of 1426 microinjections were made. Injected

eggs were incubated in tap water at 10°C.

Preparation of Embryos

Embryos were excised manually from chorions and sepa-

rated from yolk under a dissecting microscope. For RNA

isolation, embryos were frozen at −70°C in TRIzol reagent

(GibcoBRL). For studies on hexose metabolism, embryos

were transferred into physiologic saline (128 mM NaCl, 3

mM KCl, 1.5 mM MgCl2, 1.5 mM CaCl2 10 mM HEPES,

pH 7.4–7.6) and sedimented at 1300 g for 5 minutes to

remove yolk. Washing was repeated twice, and finally the

embryos were lysed in 50 mM sodium phosphate, pH 7.3 (5

µl per embryo), to measure luciferase or hexokinase activity,

or resuspended in saline (10 µl per embryo), to analyze

hexose transport or breakdown of 14C-glucose.

Isolation of RNA

Total RNA was isolated from embryos (10–20 per sample)

using TRIzol reagent and dissolved in RNase-free water

(RNasin 1 U/µl, Promega). Contamination with microin-

jection DNA was degraded with RQ1 RNase-free DNase (2

U/µg RNA) by incubating in 1× reverse transcription buffer

(Promega) for 60 minutes in 37°C. The RNA was extracted

once with 1 vol phenol (water saturated) and 1/5 vol chlo-

roform–isoamyl alcohol mixture (49:1), and precipitated

with 1 vol isopropanol. After washing in 75% ethanol, pel-

lets were dissolved in sterile water containing ribonuclease

inhibitor (1 U/µl, Promega).

Synthesis of cDNA and Reverse Transcription–Polymerase Chain Reaction

The first-strand cDNA was synthetized from 1 µg of total

RNA using 8 units of AMV reverse transcriptase (Promega),

10 pmol oligo(dT)15 primers (Promega), and a mixture of

dNTPs (10 mM each, Finnzymes) in 1× reaction buffer for

30 minutes in 45°C. After denaturation (98°C, 5 minutes),

1.2 µl of cDNA solution was used as a template in a volume

of 30 µl of reaction mixture in polymerase chain reaction

(PCR) amplification (Dynazyme DNA Polymerase Kit,

Finnzymes). The 602-bp fragment located in rHKII cDNA

and the 361-bp fragment located in hGLUT1 cDNA was

amplified using upper primer 58-GCTACCACGCACCCTACA-38

and lower primer 58-CCCTCATCTCCCTCCAC-38, and upper

primer 58-CCGAGAGTCCCCGCTTCC-38 and lower primer 58-

GTCACAAACAGCGACACGACAG-38, respectively. After denatur-

26 Aleksei Krasnov et al.

ation at 94°C for 6 minutes, samples were cycled 39 to 42

times at 96°C for 15 seconds, 50°C for 15 seconds, and 72°C

for 60 seconds, and further extended at 72°C for 5 minutes.

Amplification products were analyzed on 1.0% agarose-

TAE gel, stained with ethidium bromide, and photographed

under ultraviolet light.

Hexokinase Assay

Suspensions of embryos lysed in 50 mM sodium phosphate

buffer were centrifuged for 5 minutes at 2000 g. We added

5 µl of supernatant to 800 µl of substrate (40 mM Tris HCl

[pH 7.6], 4 mM MgCl2, 2 mM EDTA, 10 mM glucose, 1

mM NADP, 3 mM ATP, 0.2 IU glucose-6-phosphate dehy-

drogenase). Increase of NADPH concentration was mea-

sured at 340 nm.

Luciferase Assay

Luciferase activity was measured with a luminometer (Bio-

Orbit) using standard reagent (20 mM Tricine, 1.07 mM

(MgCO3)4Mg (OH)2, 2.67 mM MgSO4, 0.1 mM EDTA,

33.3 mM dithiothreitol [pH 7.8], 0.002 mM coenzyme A,

0.47 mM luciferin, 0.53 mM ATP).

Uptake of 14C-O-methyl-D-glucose

Embryos (20 per sample) were resuspended in saline con-

taining 3-O-methylglucose (OMG) with tracer (14C-OMG,

55.2 mCi/mmol) added to 0.5 µCi/ml. Cytochalasin B,

when used, was applied at concentration 0.01 mM. After

incubation for 45 minutes in 10°C, ice-cold stop solution

(150 mM NaCl, 1.25 mM KI, 0.001 mM HgCl2, 0.1 mM

phloretin, 10 mM HEPES, pH 7.4–7.6) (Weiser et al., 1983)

was added, and embryos were immediately sedimented at

1300 g for 1 minute. Washings were repeated 4 times, and

pellets were lysed in distilled water with 1% Triton X-100.

For controls the tracer was added after the stop-solution.

Breakdown of 14C-Glucose

Embryos were resuspended in saline containing 4 mM glu-

cose and preincubated at 10°C for 30 minutes. After addi-

tion of 14C-glucose (7.1 mCi/mmol) to 0.5 µCi/ml, tubes

were closed with rubber stoppers supplied with central

wells. Released 14CO2 was entrapped in paper filters soaked

with hyamine. The reaction was stopped after 8 hours by

injection of equal volume of 12% TCA, and tubes were

shaken vigorously for 30 minutes. Lactate was analyzed as in

Roca et al. (1985).

Lysed embryos or paper filters were transferred to 5 ml

of HiSafe Optiphase scintillation cocktail (Wallac), and ra-

dioactivity was measured in a 1214 Rackbeta (LKB-Wallac)

liquid scintillation counter. Unless indicated otherwise, all

chemicals were purchased from Sigma.

Statistical Analyses

Significance of difference (P < .05) between means for

transformed and control embryos was assessed using non-

parametric Mann-Whitney U test.

RESULTS

Temporal Pattern of Transient Expression inRainbow Trout Embryos

Experiments on transient expression of CMV-rHKII and

CMV-hGLUT1 in rainbow trout embryos consisted of three

consecutive series, the first of which was designed to cor-

roborate transcription of the constructs. In addition, the

reporter gene RSV-luciferase was used to quantify the tran-

sient transgene expression and to find the developmental

stage with the highest level. Daily sampling of embryos for

the reverse transcription–PCR analyses and luciferase assay

was carried out within 2 to 5 days after microinjections.

Transcripts of CMV-hGLUT1 and CMV-rHKII were

found respectively 3 and 4 days after the injections, and the

expression was further confirmed up to the last sampling

(Figure 1). Luciferase activity was detected during the whole

study period with a clear peak value on the fourth day

(Figure 2). Incubated at 10°C, embryos at this time have

reached the late blastula stage (Ignatieva, 1991) character-

ized by flattening of the germinal disc and dramatic increase

of zygotic gene expression. Assuming that temporal patterns

of transient expression are similar for constructs containing

strong viral constitutive promoters, the subsequent analyses

were made using 4-day-old embryos.

Activities of rHKII and hGLUT1 Found inRainbow Trout Embryos

After obtaining the evidence for transcription of the trans-

genes in rainbow trout embryos, the activity of rHKII and

hGLUT1 was analyzed. Using spectrophotometric assay,

HK activity was found in transformed embryos at the level

GLUT and HK Expression in Trout Embryo 27

of 20.8 nmol NADPH per embryo per hour, whereas no

glucose phosphorylation was detected in control embryos.14C-O-methylglucose (14C-OMG) uptake was measured in

control and CMV-hGLUT1-transformed embryos at con-

centrations from 0.9 to 10 mM, with and without added

cytochalasin B, the specific inhibitor of facilitative glucose

transport. In the control embryos the relation between 14C-

OMG concentration and uptake was linear and no effect of

cytochalasin B was found. In transformed embryos the rate

of hexose uptake deviated from linearity at medium con-

centrations (2.7 and 4.5 mM). In this range it was 1.65-fold

higher than in control, and inhibition of uptake with cyto-

chalasin B (56% to 62%) indicated the considerable contri-

bution of hGLUT1. At higher and lower hexose concentra-

tions, facilitative transport in transformed embryos was

probably masked with passive diffusion. To corroborate the

finding of hGLUT1 activity in trout embryos, further analy-

ses were done using a hexose concentration of 4 mM. As in

the previous test, 14C-OMG uptake in transformed embryos

was 1.53-fold higher than in control; transport was inhib-

ited with cytochalasin B, respectively, to 39.3 ± 1.8% and

5.2 ± 1.4% (Figure 3). In the presence of the inhibitor, the

rates of 14C-OMG uptake were equal in transformed and

control embryos; therefore the observed differences be-

tween groups must be attributed mainly to the facilitative

hexose transport activity due to hGLUT1 expression.

hGLUT1 and rHKII Enhance CO2 Productionfrom 14C-glucose

Next, to evaluate the effect of hGLUT1 and rHKII on car-

bohydrate metabolism, the utilization of 14C-glucose in em-

Figure 1. Reverse transcription-PCR-amplified samples showing

expression of CMV-rHKII and CMV-hGLUT1 cDNAs in rainbow

trout embryos between 2 and 5 days after microinjections. Lanes 1,

3, 5, and 7: Dnase-treated RNA 2, 3, 4, and 5 days after injections,

respectively. Treatment proved unsufficient to remove CMV-

rHKII plasmid residues in RNA extracted from 2-day-old em-

bryos, but no contamination was found further. Lanes 2, 4, 6, and

8: Embryonic cDNA 2, 3, 4, and 5 days after injections, respec-

tively. Lanes 9 and 10: DNase-treated RNA and cDNA, respec-

tively, from uninjected embryos. Lanes a and b: amplification

buffer containing rHKII- and GLUT1-specific primers, respec-

tively. Lane +: rHKII (3 ng) and hGLUT1 (1.5 ng) microinjection

DNA. Lane M: 100-bp DNA ladders (Promega). Each sample lane

represents results of two separate PCR amplifications; rHKII- and

hGLUT1-specific products are seen as 602-bp and 361-bp bands,

respectively.

Figure 2. RSV-luciferase was applied to analyze the temporal pat-

tern of transient expression in rainbow trout embryos. Assays were

done in duplicates; 15 embryos were pooled in every sample. Ac-

tivity of enzyme is expressed in relative light units.

Figure 3. Effects of CMV-hGLUT1 and CMV-rHKII, applied

separately and in combination, on carbohydrate metabolism in

rainbow trout embryos at the stage of late blastula (4 days at

10°C). 14C-OMG uptake was measured at the concentration of 4

mM; cytochalasin B was added to 0.01 mM. Production of 14CO2

was followed in the presence of glucose (4 mM). All analyses were

done in 5 replicates; 20 and 40 embryos were pooled in every

sample to measure, respectively, 14C-OMG uptake and oxidation

of 14C-glucose. Significant differences between control and test

groups (p < .05) are indicated with asterisks.

28 Aleksei Krasnov et al.

bryos was studied. Genes were injected into rainbow trout

eggs separately and as a mixture; resulting test groups are

referred to, respectively, as CMV-rHKII, CMV-hGLUT1,

and CMV-rHKII/CMV-hGLUT1. The last group was pro-

duced taking into account that GLUT and HK catalyze con-

secutive steps of glucose metabolism and that synergism

might be expected from coexpression of two genes. For

analyses embryos were incubated for 8 hours in physiologic

saline with 4 mM 14C-glucose, which was the only exo-

genous source of energy.

Production of 14CO2 in hGLUT1, rHKII, and rHKII/

hGLUT1 groups was respectively, 2.35, 1.39, and 2.72 times

greater than in controls, though in the rHKII group the

level of significance was only 0.10 (Figure 3). The effect of

hGLUT1 on the utilization of glucose in rainbow trout em-

bryos was much stronger than that of rHKII. Coexpression

of both genes did not increase significantly the production

of 14CO2 compared with expression of hGLUT1 alone, sug-

gesting that the utilization of glucose in early rainbow trout

embryos is limited mainly by the rate of transport across

plasma membrane. No accumulation of lactate was found

in the incubation media: under these experimental condi-

tions embryos utilized glucose mainly via the aerobic oxi-

dation pathway.

DISCUSSION

Besides the enhanced growth and resistance to diseases,

modification of fish metabolism will become an important

area for the application of gene transfer technology to aqua-

culture. Thus augmented carbohydrate utilization, which

has been confirmed in animals transgenic for GLUT or HK

genes, would be beneficial for fish farming.

Modification of metabolism by gene transfer, defined

as “metabolic engineering” (Bailey, 1991), implies overex-

pression of homologous genes or application of heterolo-

gous genes to increase the rate-limiting transporter or en-

zyme activities. Given that piscine HK had not been cloned

yet, and that the presence of facilitative glucose transport in

teleosts remains questionable, we used genes of mammalian

origin in our experiments. Application of genes isolated

from phylogenetically remote species to modify metabolism

in fish has associated risks, which can be illustrated by the

following examples. In our parallel studies, rat cDNA for

L-gulono-g-lactone oxidase (GLO) enzyme, which is lack-

ing in teleosts as well as in scurvy-prone animals, was trans-

ferred into rainbow trout to compensate for an absence of

L-ascorbate biosynthesis. Despite high levels of transgene

transcription in embryos and fry, no enzyme was found,

indicating the absence of conditions required for GLO

translation or stability in rainbow trout cells (Krasnov et al.,

1998). In another study, foreign GLO activity was detected

in transgenic medaka (Oryzias latipes), transformed with

similar construct; nonetheless, no L-ascorbate biosynthesis

was found (Toyohara et al., 1996). Heterologous genes can

be nonfunctional in piscine cells, or the activity of related

enzymes may be not sufficient to achieve the desired modi-

fication of metabolic functions.

Unlike proteins acting by endocrine mechanism, GLUT

and HK activity is restricted to transformed cells, which

hampers the evaluation of phenotypic changes in highly

mosaic founder transgenic fish. In fish overexpressing so-

matotropin, growth enhancement can be clearly manifested

already in the first generation of transgenics (Du et al.,

1992; Devlin et al., 1994). However, in our earlier studies

with rainbow trout fingerlings bearing rHKII and hGLUT1,

slight divergence in the parameters of carbohydrate me-

tabolism was revealed in only some of the analyzed trans-

genic individuals. At the same time we were not able to

demonstrate activity of foreign proteins using direct meth-

ods (Pitkanen et al., 1998). Thus conclusions about the

effects of foreign genes on fish metabolism are best made

not earlier than in the second generation, which means at

least 2 years for rainbow trout.

Both uncertainties about the performance of heterolo-

gous genes in fish cells and problems connected with the

detection of phenotypic effects of foreign genes in highly

mosaic transgenic founder fish imply the importance of

model experiments for metabolic engineering. In this study,

to evaluate the compatibility of mammalian GLUT and HK

genes with rainbow trout cells, embryos were used as a

model. Both rHKII and hGLUT1 cDNAs were applied with

strong viral promoter to facilitate high levels of transient

expression. In our earlier experiments (Pitkanen et al.,

1998), the expression of rHKII was shown to be driven

efficiently in Arctic charr (Salvelinus alpinus) embryos with

sockeye salmon metallothionein-B and histone 3 promoters

(Chan and Devlin, 1993). The present study confirms ex-

pression of genes both at the mRNA and activity levels. The

presence of hGLUT1 and rHKII transcripts in rainbow

trout embryos became evident 3 and 4 days after injections.

Hexokinase activity, which was too low in control blastulas

to be detected with the applied assay, proved measurable in

CMV-rHKII-transformed embryos. Further, the increased

uptake of labeled hexose and the sensitivity to specific in-

GLUT and HK Expression in Trout Embryo 29

hibitor cytochalasin B unambiguously indicated the pres-

ence of facilitative transport following the expression of

CMV-hGLUT1 in embryos. Earlier, using LacZ reporter, we

estimated that in rainbow trout blastulas the foreign gene,

injected at dose of 40 to 80 pg, is expressed in about 2% to

5% of cells. Given that the difference between control and

hGLUT1-injected embryos was within 39% to 62%, we can

expect that in transformed cells the rate of glucose uptake

was increased 30 to 80 times.

The expression of mammalian GLUTs has been studied

in various heterologous systems. Several GLUT isoforms

increased rates of glucose transport in Xenopus oocytes, and

the kinetic parameters were similar to those observed in

mammalian cells (Burant and Bell, 1992; Thomas et al.,

1993). GLUT1 also was active in Dictyostelium discoideum,

although mutation of the first 10 amino acids was needed to

adapt the gene to the preferred Dictyostelium codon usage

(Cohen et al., 1996). In yeast, Saccharomyces cerevisae, the

expression of GLUT1 was followed with production of pro-

tein, but because it was retained in intracellular structures,

only a minor increase in glucose transport was found (Kasa-

hara and Kasahara, 1996). No results are available on the

performance of the mammalian HKs in heterologous cells,

although the activity of the yeast HK gene in transgenic

mice was reported (Epstein et al., 1992).

To evaluate the effects of hGLUT1 and rHKII expres-

sion on metabolism in rainbow trout embryos, the oxida-

tion of glucose was examined. The genes were applied sepa-

rately and as a mixture to find out whether transport across

plasma membrane or phosphorylation or both were rate

limiting in rainbow trout embryonic cells. No accumulation

of lactate was found in the media during the incubation of

embryos with glucose. Production of 14CO2 in the hGLUT1

group was considerably higher than in the rHKII group,

and the coexpression of two genes did not add significantly

to the effect. This result suggested that the activity of en-

dogenous embryonic HK, though its level was too low to be

detected with the applied spectrophotometric procedure,

still exceeded the rate of glucose transport enhanced by the

expression of hGLUT1. However, it is possible that the ex-

pression of more potent GLUT isoforms would require an

additional phosphorylation capacity. Thus in Xenopus oo-

cytes the expression of GLUT2 and GLUT3 increased glu-

cose utilization only at concentrations below 1 mM, while

coexpression of a low-affinity HK, glucokinase, produced

an added effect at physiologic concentrations of 5 to 20 mM

(Morita et al., 1994). Functionality of mammalian GLUT

and HK genes in rainbow trout embryos and their effect on

glucose utilization indicate that heterologous genes prob-

ably can be used for the modification of carbohydrate me-

tabolism in fish.

Transient expression in fish embryos has been used

mainly to evaluate novel techniques of DNA delivery and

the performance of various reporter genes and promoter

sequences. In our studies rainbow trout embryos were used

for the first time to examine the physiologic consequences

of the expression of foreign genes. Advantages of this model

are connected mainly with the application of microinjec-

tions as the most straightforward and reliable method of

DNA delivery and the high levels of transient expression,

which ensure sensitivity and reproducibility of analyses.

Mainly for the same reasons, Xenopus oocytes are among

the most preferred subjects in studies on gene expression.

The extent to which preliminary conclusions based on tran-

sient expression in embryos are valid for stable transformed

fish remains an open question and requires further research

in the similarity and distinction between fish embryonic

and differentiated cells.

ACKNOWLEDGMENTS

We express gratitude to Pelagis Ltd. and Tervo Fisheries and

Aquaculture (Finland) for the supply of rainbow trout eggs.

We thank Dr. Wilson (Michigan State University, U.S.A.)

and Dr. Soeller (Pfizer Central Research, U.S.A.) for pro-

viding the plasmids, and Dr. Azhayeva (University of Kuo-

pio) for luciferase assay. We thank Dr. Chourrout (SARS

Research Centre, Norway) and Dr. Devlin (West Vancouver

Laboratory, Canada) for reading the manuscript and dis-

cussion. The study was funded by the Ministry of Agricul-

ture and Forestry of Finland and by a grant from the Mi-

nistry of Education to T.I. Pitkanen.

REFERENCES

Bailey, J.E. (1991). Toward a science of metabolic engineering.

Science 252:1668–1675.

Bolis, L., Luly, P., and Baroncelli, V. (1971). D(+)-glucose perme-

ability in brown trout Salmo trutta L. erythrocytes. J Fish Biol

3:273–275.

Burant, C.F., and Bell, G.I. (1992). Mammalian facilitative trans-

porters: evidence for similar substrate recognition sites in func-

tionally monomeric proteins. Biochemistry 31:10414–10420.

30 Aleksei Krasnov et al.

Burant, C.F., Sivitz, W.I., Fukumoto, H., Kayano, T., Nagamatsu,

S., Seino, S., Pessin, J., and Bell, G.I. (1991). Mammalian glucose

transporters: structure and molecular regulation. Recent Prog

Horm Res 47:349–388.

Chan, W.-K., and Devlin, R.H. (1993). Polymerase chain reaction

amplification and functional characterization of sockeye salmon

histone H3, metallothionein-B, and protamine promoters. Mol

Mar Biol Biotechnol 2:308–318.

Chang, P.Y., Jensen, J., Printz, R.L., Granner, D.K., Ivy, J.L., and

Moller, D.E. (1996). Overexpression of hexokinase II in transgenic

mice: evidence that increased phosphorylation augments muscle

glucose uptake. J Biol Chem 271:14834–14839.

Cohen, N.R., Knecht, D.A., and Lodish, H.F. (1996). Functional

expression of rat GLUT1 glucose transporter in Dictyostelium dis-

coideum. Biochem J 315:971–975.

Cowey, C.B., and Walton, M.J. (1989). Intermediary metabolism.

In: Fish Nutrition, 2nd ed. Halver, J.E. (ed.) New York, N.Y.: Aca-

demic Press, 259–329.

Devlin, R.H., Yesaki, T.Y., Biagi, C.A., Donaldson, E.M., Swanson,

P., and Chan, W.-K. (1994). Extraordinary salmon growth. Nature

371:209–210.

Du, S.J., Gong, Z, Fletcher, G.L., Shears, M.A., King, M.J., Idler,

D.R., and Hew, C.L. (1992). Growth enhancement in transgenic

Atlantic salmon by the use of an “all fish” chimeric growth hor-

mone gene construct. Biotechnology 10:176–181.

Epstein, P.N., Boschero, A.C., Atwater, I., Cai, X., and Overbeek,

P.A. (1992). Expression of yeast hexokinase in pancreatic beta cells

of transgenic mice reduces blood glucose, enhances insulin secre-

tion, and decreases diabetes. Proc Natl Acad Sci USA 89:12038–

12042.

Gibbs, E.M., Stock, J.L., McCoid, S.C., Stukenbrok, H.A., Pessin,

J.E., Stevenson, R.W., Milici, A.J., and McNeish, J.D. (1995). Gly-

cemic improvement in diabetic db/db mice by overexpression of

the human insulin-regulatable glucose transporter (GLUT4). J

Clin Invest 95:1512–1518.

Hansen, P.A., Gulve, E.A., Marshall, B.A., Gao, J., Pessin, J.E.,

Holloszy, J.O., and Mueckler, M. (1995). Skeletal muscle glucose

transport and metabolism are enhanced in transgenic mice over-

expressing the GLUT4 glucose transporter. J Biol Chem 270:1679–

1684.

Hariharan, N., Farrelly, D., Hagan, D., Hillyer, D., Arbeeny, C.,

Sabrah, T., Treloar, A., Brown, K., Kalinowski, S., and Mookhtiar,

K. (1997). Expression of human hepatic glucokinase in transgenic

mice liver results in decreased glucose levels and reduced body

weight. Diabetes 46:11–16.

Ignatieva, G.M. (1991). The rainbow trout Salmo gairdneri. In:

Animal Species for Developmental Studies, vol. 2 Vertebrates, Detlaff,

T.A., and Vassetzky, S.G. (eds.) New York, N.Y.: Consultants Bu-

reau, 89–124.

Ikemoto, S., Thompson, K.S., Itakura, H., Lane, M.D., and Ezaki,

O. (1995). Expression of an insulin-responsive glucose transporter

(GLUT4) minigene in transgenic mice: effect of exercise and role

in glucose homeostasis. Proc Natl Acad Sci USA 92:865–869.

Ingermann, R.L., Hall, R.E., Bissonette, J.M., and Terwilliger, R.C.

(1984). Monosaccharide transport into erythrocytes of the Pacific

hagfish, Eptatretus stouti. Mol Physiol 6:311–320.

Kasahara, T., and Kasahara, M. (1996). Expression of the rat

GLUT1 glucose transporter in the yeast Saccharomyces cerevisae.

Biochem J 315:177–182.

Katzen, H.M., Soderman, D.D., and Cirillo, V.J. (1968). Tissue

distribution and physiological significance of multiple forms of

hexokinase. Ann NY Acad Sci USA 151:351–358.

Kim, H.D., and Isaacks, R.E. (1978). The membrane permeability

of nonelectrolytes and carbohydrate metabolism of Amazon fish

red cells. Can J Zool 56:863–869.

Krasnov, A., Reinisalo, M., Pitkanen, T.I., Nishikimi, M., and

Molsa, H. (1998). Expression of rat gene for L-gulono-g-lactone

oxidase, the key enzyme of L-ascorbic acid biosynthesis, in guinea

pig cells and in teleost fish rainbow trout (Oncorhynchus mykiss).

Biochim Biophys Acta 1381:241–248.

Marshall, B.A., Ren, J.M., Johnson, D.W., Gibbs, E.M., Lillquist,

J.S., Soeller, W.C., Holloszy, J.O., and Mueckler, M. (1993). Germ-

line manipulation of glucose homeostasis via alteration of glucose

transporter levels in skeletal muscle. J Biol Chem 268(25):18442–

18445.

Morita, H., Yano, Y., Niswender, K.D., May, J.M., Whitesell, R.R.,

Wu, L., Printz, R.L., Granner, D.K., Magnuson, M.A., and Powers,

A.C. (1994). Coexpression of glucose transporters and glucokinase

in Xenopus oocytes indicates that both glucose transport and phos-

phorylation determine glucose utilization. J Clin Invest 94:1373–

1382.

Olson, A.L., and Pessin, J.E. (1996) Structure, function, and regu-

lation of the mammalian facilitative glucose transporter gene fam-

ily. Annu Rev Nutr 16:235–256.

Pitkanen, T.I., Krasnov, A., Reinisalo, M., and Molsa, H. (1998).

Transfer and expression of glucose transporter and hexokinase

genes in salmonid fish. Aquaculture (in press).

Ren, J.M., Marshall, B.A., Gulve, E.A., Gao, J., Johnson, D.W.,

Holloszy, J.O., and Mueckler, M. (1993). Evidence from transgenic

mice that glucose transport is rate-limiting for glycogen deposition

and glycolysis in skeletal muscle. J Biol Chem 268:16113–16115.

GLUT and HK Expression in Trout Embryo 31

Roca, P., Gianotti, M., and Palou, A. (1985). Enzymatic lactate

specific radioactivity determination in biological specimens. Anal

Biochem 148:190–193.

Thelen, A.P., and Wilson, J.E. (1991). Complete amino acid se-

quence of the type II isozyme of rat hexokinase, deduced from the

cloned cDNA: comparison with a hexokinase from a Novikoff

ascites tumor. Arch Biochim Biophys 286:645–651.

Thomas, H.M., Takeda, J., and Gould, G.W. (1993). Different

targeting of glucose transporter isoforms heterologously expressed

in Xenopus oocytes. Biochem J 290:707–715.

Tiihonen, K., and Nikinmaa, M. (1991). D-glucose permeability in

river lamprey (Lampetra fluviatilis) and carp (Cyprinus carpio)

erythrocytes. Comp Biochem Physiol 100A:581–584.

Toyohara, H., Nakata, T., Touhata, K., Hashimoto, H., Kinoshita,

M., Sakaguchi, M., Nishikimi, M., Yagi, K., Wakamatsu, Y., and

Ozato, K. (1996). Transgenic expression of L-gulono-g-lactone

oxidase in medaka (Oryzias latipes), a teleost fish that lacks this

enzyme necessary for L-ascorbic acid biosynthesis. Biochem Bio-

phys Res Commun 223:650–653.

Tsao, T.S., Burcelin, R., Katz, E.B., Huang, L., and Charron, M.J.

(1996). Enhanced insulin action due to targeted GLUT4 overex-

pression in muscle. Diabetes 45:28–36.

Tse, C.-M., and Young, J.D. (1990). Glucose transport in fish

erythrocytes: variable cytochalasin-B-sensitive hexose transport

activity in the common eel (Anguilla japonica) and transport de-

ficiency in the paddyfish eel (Monopterus albus) and rainbow trout

(Salmo gairdneri). J Exp Biol 148:367–383.

Voss-McCowan, M.E., Xu, B., and Epstein, P.N. (1994). Insulin

synthesis, secretory competence, and glucose utilization are sensi-

tized by transgenic yeast hexokinase. J Biol Chem 269:15814–

15818.

Walton, M.J., and Cowey, C.B. (1982). Aspects of intermediary

metabolism in salmonid fish. Comp Biochem Physiol 73B:59–79.

Weiser, M.B., Razin, M., and Stein, W.D. (1983). Kinetic tests of

models for sugar transport in human erythrocytes and a compari-

son of fresh and cold-stored cells. Biochim Biophys Acta 727:379–

388.

32 Aleksei Krasnov et al.