How To Create Transgenic Organisms

5. THE PROCESS OF GENETIC MODIFICATION

Production of GMOs is a multistage process which can be summarized as follows:

1. identification of the gene interest;

2. isolation of the gene of interest;

3. amplifying the gene to produce many copies;

4. associating the gene with an appropriate promoter and poly A sequence and insertion into plasmids;

5. multiplying the plasmid in bacteria and recovering the cloned construct for injection;

6. transference of the construct into the recipient tissue, usually fertilized eggs;

7. integration of gene into recipient genome;

8. expression of gene in recipient genome; and

9. inheritance of gene through further generations.

5.1 Choice of target genes

As shown in Table 2 the most popular gene used in aquatic species is growth hormone (GH) for reasons that are obvious. GH has been widely used in terrestrial species and as the gene sequence is highly conserved; the product is readily utilized across species boundaries. It may also be noted that, at least in some cases, enhanced growth is associated with more effective utilization of food.

Cold water temperatures are often a major problem in aquaculture in temperate climates when an unusually cold winter can severely damage both production and brood fish stocks of fish. Some marine teleosts have high levels of serum anti-freeze proteins (AFP) or glycoproteins (AFGP) which reduce the freezing temperature by preventing ice-crystal growth. Fletcher, Hew and Davies (2001) have shown that there is one class of AFGP and four classes of AFP. Most are expressed primarily in the liver and some show clear seasonal changes (Melamed et al., 2002). Work has particularly focussed on the production of AFP from the winter flounder (Pleuronectes americanus), and the gene has been successfully introduced into the genome of Atlantic salmon, integrated into the germ line and passed on to F3 offspring where it was expressed in the liver. However, a number of Ala, Pro-specific endopeptidases are required for production of mature proteins and these are not present in Atlantic salmon. Furthermore, the AFP gene in winter flounder, and possibly other Arctic species, exists in many copies (see Section 7). Thus, much further work is required in order to develop effective antifreeze activity in Atlantic salmon (Hew et al., 1999). Work on AFP has also been conducted in goldfish (Wang et al., 1995) and milkfish (Wu et al., 1998).

Genetic manipulation has also been undertaken in order to increase the resistance of fish to pathogens. This is currently being addressed by the use of DNA vaccines (encoding part of the pathogen genome) and antimicrobial agents such as lysozyme (Demers and Bayne 1997). An example is the injection of Atlantic salmon with a DNA sequence encoding infectious hematopoeitic necrovirus (IHNV) glycoprotein under the control of the cytomegalovirus promoter (pCMV). Challenge with the virus eight weeks later revealed that a significant degree of resistance had been achieved. The fish were still resistant and were shown to have generated antibodies three months later (Traxler et al., 1999). Similar studies have been undertaken for other fish diseases eg. Haemorrhagic septicaemia virus (VHS) (Lorenzen, Olesen and Koch, 1999) and work of this kind appears to have great potential value for fish farms (Melamed et al., 2002). We would also draw attention to the work using a cecropin B gene from the moth Hyaloplova cecropin. When channel catfish transgenic for this gene were challenged with Flavobacterium columnare and Edwardsiella ictaluri survival was better for transgenics than controls (Dunham et al., 2002)

Table 2. Results in induction of GMOs in aquatic species.

Species	Target gene	Int	Exp	Trans	Reference
At. Salmon	GH	+	+	+	Hew & Fletcher, 2001
At. Salmon	AFP	+	+	+	Hew & Fletcher, 2001
Coho salmon	GH	+	+	Nd	Stevens Devlin, 2000
Tilapia	tiGH	+	+	+	Martinez et al., 1999
Tilapia	Fish GH	+	+	Nd	Rahman & Maclean, 1999
Carp	GH	+	+	Nd	Hinits & Moav, 1999
Salmon	Glucose transporter and hexokinase	+	+	+	Pitkanen et al., 1999
Tilapia	GH	+	+	Nd	Rahman et al., 1998
At. salmon	GH	+	+	+	Saunders, Fletcher & Hew, 1998
Carp	HGH	+	+	+	Fu et al., 1998
At. salmon	GH	+	+	+	Stevens, Sutterlin & Cook, 1998
Tilapia	tiGH	+	+	+	de la Fuente et al., 1998
Tilapia	INT-tiGH	+	+	+	de la Fuente et al.,1998
Tilapia	CSGH	+	+	+	Rahman & Maclean, 1998
Tilapia	INT-tiGH	+	+	+	Hernandez et al., 1997
Tilapia	ypGH	+	+	Nd	Chen et al., 1997
Abalone	GH	+	+	Nd	Powers, Kirby & Gomez-Chiarri, 1996
Medaka	CAT	+	+	+	Kinoshita et al., 1996
Tilapia	GH	+	+	+	de la Fuente et al., 1996
At. salmon	GH AFP	+	+	+	Choy et al., 1996
Tilapia	tiGH	+	+	+	Martinez et al., 1996
Tilapia	Lac Z	+	+	+	Alam et al., 1996
Tilapia	tiGH	+	Nd	Nd	Martinez et al., 1996
Coho salmon	GH	+	+	+	Devlin et al., 1995a
Pacific salmon	CSGH	+	+	+	Devlin et al., 1995b
Common carp	RTGH	+	+	Nd	Chatakondi et al., 1995
Common carp	CSGH	+	+	+	Moav et al., 1995
Medaka	Lac Z	+	+	Nd	Tsai, Tseng & Liao, 1995
Brine shrimp	Luciferase reporter gene	+	Nd	Nd	Gendreau et al., 1995
Common carp	RTGH	+	+	Nd	Chatakondi et al., 1995
Common carp	CSGH	+	+	+	Moav et al., 1995
Pacific salmon	CSGH	+	+	+	Devlin et al., 1995b
Rainbow trout	CSGH	+	+	+	Devlin et al., 1995a
Cutthroat trout	CSGH	+	+	+	Devlin et al., 1995b
Chinook salmon	CSGH	+	+	+	Devlin et al., 1995b
Loach	CSGH	+	+	Nd	Tsai, Tseng & Liao, 1995
Salmon	GH	+	+	Nd	Devlin et al., 1994
Chinook salmon		+	Nd	Nd	Sin et al., 1994
Laminaria japonica	Plasmid BI221	+	+	Nd	Qin et al., 1994
Undaria pinnatifida	Plasmid BI221	+	+	Nd	Qin et al., 1994
Nigorobuna	E. coli beta galactosidase	+	+	Nd	Ueno et al., 1994
Blunt snout bream	HGH	+	+	+	Wu et al., 1994
Common carp	HGH	+	+	+	Wu et al., 1994
Oreochromis niloticus	Bacterial lacZ	+	+	+	Mclean, 1994
Zebrafish		+	Nd	Nd	Hackett et al., 1994
African catfish	AFP GH	+	+	Nd	Erdelyi et al., 1994
Common carp	AFP GH	+	+	Nd	Erdelyi et al., 1994
Abalone	GH	+	+	Nd	Powers et al., 1994
Pacific salmon	GH sockeye salmon	+	+	Nd	Devlin et al., 1994
Zebrafish	Firefly luciferase	+	+	-	Patil, Wong & Khoo, 1994
Zebrafish	CSGH	+	+	+	Zhao, Zhang & Wong, 1993
Common carp	RTGH	+	+	+	Chen et al., 1993
Zebrafish	luciferase	+	-	-	Kavumpurath et al., 1993
Common carp	HGH	+	+	+	Cui et al., 1993
Tilapia	RGH	+	Nd	Nd	Rahman & Maclean, 1991
Zebra fish	CAT	+	Nd	Nd	Khoo et al., 1992
Tilapia	HGH	Nd	Nd	Nd	Ber et al., 1992
Zebrafish	CAT	+	+	Nd	Sharps et al., 1992
Goldfish	Neomycin resistance CAT	+	+	Nd	Guise, Hackett & Faras, 1992
Northern Pike	BGH	+	+	Nd	Guise, Hackett & Faras, 1992
Atlantic salmon	Winter flounder AFP	+	+	+	Fletcher, Davies & Hew, 1992
At. salmon	Bacterial CAT Chinook salmon GH	+	Nd	Nd	Jun Du et al., 1992
Common carp	RTGH	+	+	+	Chen et al., 1992
Channel catfish	RTGH	+	+	+	Chen et al., 1992
Northern Pike	Bacterial CAT BGH and CGH	+	Nd	Nd	Moav et al., 1992
Walleye	Bacterial CAT BGH and CGH	+	Nd	Nd	Moav et al., 1992
Zebrafish	Bacterial CAT BGH and CGH	+	Nd	Nd	Moav et al., 1992
Carp	Grass carp GH	+	Nd	Nd	Zhu, 1992
Zebrafish		+	-	Nd	Khoo et al., 1992
Northern pike	BGH CSGH	+	+	Nd	Gross et al., 1992
Channel catfish	Salmon GH	+	+	Nd	Dunham et al., 1992
At. salmon	CSGH	+	+	Nd	Jun Du et al., 1992
Gilthead seabream	BGH and HGH	+	-	Nd	Cavari et al., 1993
Rainbow trout	Carp alpha globin	+	+	+	Yoshizaki et al., 1991
Rainbow trout	BGH	+	+	Nd	Chandler et al., 1990
Tilapia	HGH	+	Nd	Nd	Brem et al., 1988

Abbreviations used in Table 2 above:

At:	Atlantic salmon
GH:	Growth Hormone
AFP:	Anti-freeze Protein
Nd:	Not determined
HGH:	Human Growth Hormone
BGH:	Bovine Growth Hormone
CS:	Coho Salmon Growth Hormone
YP:	Yellowfin Porgy Growth Hormone
CAT:	Chloramphenicol Acetyl Transferase
TiGH:	Tilapia Growth Hormone
RTGH:	Rainbow Trout Growth Hormone
RGH:	Rat Growth Hormone
Int:	Integration
Exp:	Expression
Trans:	Transmission

There appears to be no published evidence for integration of vaccine DNA into the recipient genome. Nevertheless, the persistence of the DNA appears often to be relatively long which suggests some replication (not normally expected with non-chromosomal pieces of DNA). It seems desirable for the moment to regard such treated animals as "transient" GMOs rather than full GMOs.

5.2 Isolation of the gene of interest

Usually the gene of interest will already be available as an element of a "library" of short sections of the total genome of the donor strain or species. If this is the case the procedure followed is to multiply the gene using the PCR reaction. If, however, the gene is to be taken from a genome not previously investigated, a more complex procedure will need to be followed. The use of the technique of the Polymerase Chain Reaction (PCR) enables the gene in both the cases noted above to be multiplied to the level of several million copies needed for the generation of the construct (see Section 5.3).

5.3 Cloning the gene of interest

When many copies of the target gene have been generated, the gene is placed in a "construct" (see Section 5.4). Once the gene of interest has been ligated enzymatically into the construct, this whole complex is ligated into bacterial plasmids (see Figure 3), which act as "production vectors" and enable the gene to be replicated many times within the bacterial cells. The bacteria are then plated out. It is possible to tell from reporter genes (see below) whether the vector has been taken up by the bacterial cells. This usually involves some colour change in the colonies containing inserted DNA. The many times amplified DNA construct is then enzymatically cut out of the plasmids (after these have been removed from the bacterial cells) and it is ready to be used for insertion into eggs of the host species.

A more detailed outline of the technical details of the processes outlined in Sections 5.2 and 5.3 may be found in Maclean (1998).

5.4 The construct

A construct is a piece of DNA which functions as the vehicle or vector carrying the target gene into the recipient organism. It has several different regions as shown in Figure 2. There is a promoter region which controls the activity of the target gene, a region where the target DNA is inserted, usually some type of reporter gene to enable one to ascertain whether the target has combined successfully with the construct and a termination sequence.

Figure 2. Diagram of DNA sequence of a basic plasmid and incorporated construct.

The sources of these several DNA sequences may be different species although promoter and target genes would ideally be derived from the same species

As shown in Table 3, constructs have been reported from 92 studies. The number of different constructs is greater than the number of target genes used in aquaculture and a substantial research effort has been made in this area. From the early 1990s research focussed on developing "all fish" constructs in preference to using mammalian promoters.

The use of all-fish constructs has dramatic effects on expression of transgenes, e.g. Devlin et al. (1994), developed an all salmon gene construct which accelerates the growth of transgenic salmonids by over 11 fold. In tilapia, Maclean (1994) found that using carp beta actin instead of rat beta actin promoter led to a ten fold increase in production of hormone in transgenic animals.

Table 3. Summary of major research effort in inducing GMOs in aquatic species.

Species	Target gene	Typical construct	Typical induction method	Number of studies
Salmon spp.	GH AFP	Ocean pout AFP linearized DNA	Microinjection	17/92
Rainbow Trout	GH	Ocean pout AFP	Microinjection	14/92
Tilapia spp.	GH	Cytomegalovirus (CMV)	Microinjection	12/92
Carp	GH	Rous Sarcoma Virus Long Tandem Repeat	Microinjection	17/92
Zebrafish	Luciferase	pMTL plasmid	Microinjection	16/92
Medaka	CAT	AFP	Microinjection	11/92

Other important work suggested that the optimal stage at which the transgene is introduced might vary between cells and species eg. Garcia del Barco et al. (1994) using Zebrafish showed that there were differences in the regulatory requirements for cells and embryos, and suggested therefore that constructs should be assayed in both cells and embryos.

Other work shows how critical the nature of the gene construct is. Devlin et al. (1995a) showed that using an opAFPGHc gene construct in coho salmon eggs gave rise to some alevins which had the typical brown colouration, while the remainder displayed a distinct green colouration. The results suggest that the green phenotype arose from the presence of the opAFPGHc construct and therefore could be indicative of transgene uptake/transmission. All the offspring were tested by PCR for presence of the transgene and 182 of 184 alevins were correctly assigned on this basis. However, it was found that later in development all fish turned green (the normal colour later in development) and so the transgenic fish were showing accelerated growth. Later in development it was found that most of the transgenic fish showed signs of cranial abnormality probably due to accelerated growth (see Section 9.3). While the onstruct was useful in that transgene uptake could be monitored, further work was needed to ensure that healthy fish could be produced.

5.5 Techniques for inducing transgenics

Transgenic fish have largely been produced through microinjection into fertilised eggs or early embryos (see Table 2). Electroporation of sperm has been shown to be successful in some species eg. Zebrafish (Khoo et al., 1992) Chinook salmon (Sin et al., 1994) and Loach (Tsai, Tseng and Liao, 1995). Liposomes have also been utilized as vectors (Khoo 1995). Ballistic methods using microprojectiles have been investigated in Artemia with a view to their use in generating transgenic crustacea (Gendreau et al., 1995) and also in seaweed species (Qin et al.,1994). "Baekonisation", an electric, flat field type of electroporation was utilized to transfer DNA into Zebrafish embryos (Zhao, Zhang and Wong, 1993), this method appeared to be successful but has not been taken up in the same way as other forms of electroporation and microinjection methods.

More recently the use of embryonic stem cells (ESC) as a method for inducing transgenesis has been advocated. These cells are undifferentiated and remain totipotent, so they can be manipulated in vitro and subsequently reintroduced into early embryos where they can contribute to the germ line of the host. In this way genes could be stably introduced or deleted (Melamed et al., 2002). Despite the early success of ESC technology in mice, the uptake of the technology for fish has been slow, although early precursor cells (Mes 1) have been cultivated from Medaka and show many of the same features as mouse ESC. Studies by Hong, Winkler and Schartl (1996, 1998) and Hong, Chen and Schartl (2000) showed that 90 percent of host cell blastulae transplanted with Mes 1 cells developed into mosaic fry, and these cells became integrated into organs derived from all three germ layers, and differentiated into various types of functional cells.

Another example of new and possibly more efficient ways for gene transfer is the use of pantropic retroviral vectors. These are able to infect a wide range of host cells and have been used to infect newly fertilized Medaka eggs with a reporter gene, which appeared to become integrated into the entire germ line of some of the P1 females (Lu, Burns and Chen, 1997). In Zebrafish when retroviral infection and microinjection were compared, the two methods were equally efficient in passing the transgene into eggs, but there was wider variability in the extent of reporter gene expression among those founders that were microinjected (Linney et al.,1999). However, the use of retroviruses is not without problems (see Section 9.1).

The microinjection method is suitable for relatively small numbers of organisms being manipulated whereas electroporation, sperm/liposome mediation and bombardment methods are more suitable for mass treatments. The most popular method of insertion of transgenes in aquaculture is microinjection; in 92 studies reviewed from 1985 to the present, 68 used microinjection, eleven used sperm mediated methods, six used electroporation and five used both sperm mediation and electroporation. However, the problem of mosaic expression of the transgenes is common, and this gives rise to varying proportions of transgenic genotypes in the progeny.

A useful review of technical details of the techniques mentioned can be found in Sin (1997).

5.6 Integration sites

The factors determining sites of integration are still poorly understood though research in this direction is increasing. It is particularly important to gain greater accuracy in controlled site of integration because of the unpredictable effects of uncontrolled integration on resident genes. Caldovic and Hackett (1995) tested the ability of special sequences called transposable border elements from other species to confer position-independent expression of transgenes or enhance integration of transgenic constructs into fish chromosomes. Early results indicate that such elements from some species do not act as enhancers and do not improve integration frequencies. However, both avian and insect border elements were found to confer position-independent expression as judged from expression of CAT genes in F₁ fish. Hackett et al., (1994) showed that co-transfer of retroviral integrase protein with transgenic DNA can accelerate and enhance the rate of integration. More studies of this type are needed to improve the success and controlled positioning of integration of transgenes in the future.

5.7 Expression of gene

The uptake and integration of a transgene does not guarantee that the gene will express itself in the new genetic environment. Tests must be carried out to determine whether there is expression and if there is expression, at what level this takes place. Clearly, in commercial aquaculture only those transgencs expressing the target gene at a sufficiently high level will be of interest.

5.8 Inheritance of gene

A fish which expresses the target gene at an acceptable level may not be able to transmit the gene to progeny. This is because many transgenics are mosaic individuals and unless the gonads are included in the tissues possessing the transgene the transgenic animals will not breed true. Appropriate breeding tests must, therefore, be carried out.

The high proportion of mosaic individuals is one reason why the proportions of progenies of different genotypes resulting from parents that are putatively hemizygous for a transgene do not necessarily conform to mendelian expectations. Another reason is the integration of two or more copies of the transgene at different sites in the recipient genome. Further breeding tests will be required in order to establish a pure breeding line of transgenic fish.