This invention relates to transgenic fish containing a modified insulin gene which has been altered to secrete humanized insulin. This invention further relates to the xenotransplantation of transgenic islets in the treatment of diabetes and to an improved method for mass isolation of fish islets.

BACKGROUND OF THE INVENTION

Diabetes mellitus is a disease resulting in significant morbidity and mortality. The total annual direct and indirect costs of diabetes in the Unites States exceeds $90 billion dollars. Insulin-dependent diabetes mellitus (IDDM), because it occurs in a younger population than non-IDDM, accounts for a disproportionate percentage of these costs. Although the acute manifestations of IDDM can be controlled with daily insulin injections, most patients eventually develop sequelae such as blindness, nephropathy, neuropathy, microangiopathy, and cardiovascular disease. Substantial evidence suggests that meticulous control of glycemia will prevent or minimize these sequelae.

A more physiological method of treating diabetes would be pancreas or islet transplantation. Whole or segmental pancreas transplantation has been performed successfully in man and some preliminary evidence suggests that this technique will prevent the sequelae of diabetes in man. However, pancreas transplantation is not trivial surgery; it poses problems with drainage for exocrine secretions and requires a lifetime of immunosuppressive therapy. On the other hand, islet transplantation has certain theoretical advantages particularly related to the ease of surgery, the absence of extraneous exocrine tissue, and the cryopreservability of isolated islets. More importantly, islets are more amenable to immunoalteration. Various methods have been developed to prolong allograft survival without continuous immunosuppression in rats and mice. The ability to transplant islets without continuous immunosuppression may eventually prove absolutely

necessary in man because many immunosuppressive drugs are somewhat toxic to islets.

Recent improvements in the methods of mass islet isolation and several recent clinical reports suggest that islet transplantation is on the verge of becoming a feasible treatment for IDDM. However, several obstacles exist. First, islets comprise only 2% of the human pancreas; yields from human "islet isolation" procedures are extremely variable and several human donor pancreases are often required to generate sufficient islets for a single transplant. Second, islet allograft rejection has proven difficult to manage using conventional methods and, unfortunately, the majority of islet allografts are quickly lost. Third, there are insufficient human donor pancreases available to treat the vast numbers of type I diabetic patients. Therefore, it seems likely that widespread implementation of islet transplantation would require the development of clinical islet xenotransplantation.

In response to this eventuality, many biomedical corporations are spending millions of dollars developing and patenting "bio-artificial pancreas" technologies (i.e., microencapsulation or macroencapsulation of islet tissue) . The underlying concept behind these approaches is that the islet tissue is protected from the immune system by a membrane with pore sizes small enough to prevent immunocytes and antibodies from damaging the graft yet large enough for insulin, oxygen, glucose, and nutrients to pass freely.

During the past few years, several clinical islet transplantation centers have devoted extensive effort to develop experimental islet xenotransplantation models using large animals as donors. Most of these studies have centered on porcine, bovine, canine, or non-human primate islets. However, the pancreata in these species, like the human pancreas, are fibrous and do not readily yield large quantities of intact, viable islet tissue. Moreover, generation of islet preparations from large animal donors is expensive and islet yields are variable.

We have developed a unique animal model for islet xenotransplantation utilizing tilapia, a teleost fish, as islet donors (1) . The islet tissue in certain teleost fish, called principal islets or Brockmann bodies (BBs) , is anatomically distinct from their pancreatic exocrine tissue and can be easily identified macroscopically. Expensive islet isolation procedures, such as required when procuring islet tissue from mammalian pancreases, are unnecessary. The BBs can be simply harvested with a scalpel and forceps. We have shown that tilapia islets transplanted into diabetic nude mice will produce long-term normoglycemia and a mammalian-like glucose tolerance curve (2) .

Teleost fish insulin has been used to maintain human diabetics (1) . However, it is likely that the immunogenicity of teleost insulin may prevent clinical application for BB xenotransplantation. On the other hand, the production of transgenic fish whose BBs produce humanized insulin may circumvent this problem. Transgenic fish which produce BBs that physiologically secrete humanized insulin, combined with improvements in bioartifical pancreas technology and encapsulation procedures, would eliminate the need for human pancreatic donors and islet isolation procedures.

Until recently, BBs were harvested manually by microdissection while visualized through a dissecting microscope inside a laminar flow hood (3) . Although this was much easier and less expensive than the standard procedure of harvesting islets from rodents, it was a time consuming and tedious task. Although it was easy to harvest sufficient islets to perform xenografts in mice, this method was not well- suited to harvest large volumes of islet tissue as would be required for clinical use or large animal studies. Furthermore, microdissection allows us to collect less than 50% of the islet tissue per donor fish (i.e., those large BBs that are easily visible with the naked eye) . Therefore, development of a more efficient method of harvesting BBs would be critical for the future application of fish islets as a donor source for

clinical and experimental use. We have recently developed a mass-harvesting method (4) .

To date, transgenic fish technology has been used to produce hardier fish that will grow rapidly and will tolerate adverse environments (5-6) . Most of these efforts have been directed at insertion of growth hormone transgenes. Another approach has been to insert antifreeze genes from species that tolerate very cold waters (i.e., such as winter flounder) into other species so that they will not only survive, but actually thrive in colder water. This approach permits aquaculture in more northerly regions and allows aquaculture stocks to grow year-round, rather than just during the summer growth season.

No previous transgenic fish studies have been directed at the insulin gene. In fish, insulin is primarily a growth hormone whereas, in mammalians, it is primarily a glucostatic hormone (7) . Therefore, it is very likely that altering the expression and/or structure of the fish insulin gene may enhance growth. Consequently, a transgenic fish with altered fish insulin gene expression may demonstrate enhanced growth potential.

SUMMARY OF THE INVENTION

This invention relates to a humanized insulin gene capable of being expressed in a tropical fish islet cell wherein said gene comprises a coding sequence of an insulin gene under the control of regulatory sequences of the tropical fish insulin gene, wherein said humanized insulin gene encodes an insulin protein that is biologically active in humans.

The term "humanized insulin gene" as used herein means a modified fish insulin gene that contains a sufficient portion of the human insulin coding sequence to produce the humanized insulin product. The humanized insulin product should be biologically active when transplanted into a human and should contain sufficient human insulin sequences to avoid destruction of the humanized insulin product by the human immune system. In a preferred embodiment the humanized insulin gene should encode the alpha chain and at least the first 29 amino acids of the beta chain of human insulin.

The humanized or modified insulin gene further contains a sufficient portion of the regulatory sequences of the tropical fish insulin gene to allow expression of the humanized gene in tropical fish islet beta cells.

The present invention also relates to the preparation of a transgenic fish whose islet cells secrete humanized insulin.

This invention also relates to the islets of the transgenic fish and their use in xenotransplantation to treat diabetes.

In a preferred embodiment the tropical fish is a teleost fish such as tilapia.

This invention further relates to an improved method of harvesting principal islets comprising the steps of: a) incubating the islets with collagenase for a period of time sufficient to release the islets from adipose and connective tissue; and b) separating the released islets from adipose and connective tissue.

This invention also relates to the use of the humanized insulin gene to promote growth in fish. BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 - Amino acid sequence of the tilapia insulin hormone determined by automated Edman degradation. The amino acid sequence of the tilapia A chain and B chain is illustrated in SEQ. ID. No. 1 and 2, respectively.

Figure 2 - A comparison of the primary structure of the tilapia insulin with the human insulin. (-) represents sequence identity. The amino acid sequence of the A chain and B chain human insulin sequence is illustrated in SEQ. ID. No. 3 and 4, respectively.

Figure 3 - DNA and the corresponding amino acid sequence of tilapia insulin gene. The DNA sequence is also shown in SEQ. ID. No. 5

Figure 4 - DNA and the corresponding amino acid sequence of a first humanized tilapia insulin gene. The DNA sequence is also shown in SEQ. ID. No. 6.

Figure 5 - DNA and the corresponding amino acid sequence of a second humanized tilapia insulin gene. The DNA sequence is also shown in SEQ.ID. No. 7.

DETAILED DESCRIPTION OF THE INVENTION Selection of Appropriate Donor Species

When selecting a fish donor species for xenotransplantation one should consider the following seven criteria. First, the fish should be readily available locally or easily bred. Second, it should be large enough to work with easily. Third, the fish species should have one or more discrete BBs composed of relatively pure endocrine tissue. Fourth, the BBs must be able to tolerate culture at 37°C, the body temperature of the host, without undergoing necrosis, degranulation, or loss of function. Fifth, the species should maintain fasting plasma glucose levels in a mammalian range. Sixth, insulin secretion must be glucose dependent. And finally, the BBs must be able to maintain long-term normoglycemia after transplantation into diabetic nude mice.

In a preferred embodiment of the present invention, a warm water teleost fish is used as donor species. Phylogenetically, teleosts are a large and diverse infraclass of bony fish containing more than 30,000 species. Brockmann bodies (BBs) tend to occur only in "higher" teleosts; other teleosts tend to have disseminated islets. Tilapia (Oreochromis nilotica) and the giant gorami (Osphronemus gorami) are examples of tropical higher teleosts. Other tropical fish species having anatomically distinct islet tissue such as tambaqui (Colossoma macropomum) and Pacu (Piractus mesopotamicus) may also be used as donor species.

The primary structure of teleost insulins varies from that of man. Several teleost insulins have been purified and their amino acid sequences determined. All have an extra residue at the beginning of the B chain and are missing residue 30 at the end of the B chain; substitutions may occur at various other residues on both the A chain (e.g., 9 aa substitutions between human and cod insulin) and B chain (6 aa

- human and cod) . Most teleost fish insulins exhibit one-third to one-half the biological potency of human insulin (8) . Preparation of Humanized Insulin Gene

Tilapia insulin has been purified and sequenced (8) (Figure 1) . Tilapia insulin differs from human insulin by nine amino acids in the A-chain and eight a ino-acids in the B-chain (Figure 2) .

Using an .in vitro polymerase chain reaction (PCR) cloning strategy (Takara Shuzo Co. LTD) and nested degenerate primers based on conserved amino acids from human and several teleost fish insulins (9-11) , the entire tilapia insulin gene has been cloned, including both 1 kb of 5' untranslated and 1.2 kb of 3' untranslated sequence (Figure 3 and SEQ. ID. No. 5) . The complete coding regions, including the leader sequence, B chain, C-peptide, and A chain have been sequenced. There is 100% identity between the deduced amino acid sequence and the protein sequence that was recently done by us using amino acid sequencing of purified tilapia insulin (Figure 1) . In comparison to human insulin, the tilapia A chain has 12/21 identical amino acids (57.1%) and five of the substituted amino acids are conserved changes. The B chain has 23/30 identical amino acids (76.7%) and three conserved changes when compared to the human protein. As expected, the C-peptide has little identity with the human C-peptide. The genomic structure of all known insulin genes contains an intron in the C chain. As expected, the tilapia insulin gene contains a 316 bp intron after the seventh amino acid in the C chain (phase 1) . This intron has the proper GT/AG splicing donor/acceptor sequences. The 1 kb 5' upstream sequence should contain all the regulatory units so the gene is regulated (i.e. glucose responsive) in a tissue specific manner similar to its native counterpart.

Using site-directed mutagenesis and linker substitutions, the tilapia insulin gene was humanized so that it contains exons which code for human insulin while still maintaining all of the tilapia regulatory (non-coding) sequences. The substituted codons were devised so that they will code for the human insulin chains but will still use

tilapia preference codons in order to get proper protein expression. The tilapia pre-insulin leader and the C chain were not altered. This construct will maintain all the similar recognition sites for the endopeptidases and, therefore, transgenic fish should technically process this modified protein as the native insulin. The first of the humanized tilapia genes lacks the human terminal threonine on the B chain (i.e., like tilapia) (Figure 4 and SEQ. ID. No. 6) ; this will guarantee proper cleavage of the B-C chain. The second construct contains the terminal B chain threonine (Figure 5 and SEQ. ID. No. 7) . Preparation of Transσenic Fish

To create transgenic fish, the humanized or modified gene is inserted into fertilized tilapia ova. First, the modified gene insert is removed from the plasmid DNA. The DNA is suspended in NT buffer (88 mM NaCl, 10 mM Tris-HCl pH 7.5) and 10* - 10 ⁷ copies are injected into each pronucleus (12) . A linear fragment should increase the likelihood of integration and should also decrease extrachromosomal copies of the gene (12) .

Unlike mice, which may produce 10-15 eggs at a time, fish may produce thousands of eggs at once. Tilapia are particularly prolific and are, in many ways, ideal for transgenic studies. Unlike many other teleost fish species which spawn once a year, female tilapia are ready to spawn every two weeks. Tilapia grow quickly and become sexually mature in five months.

Tilapia broodstock and offspring can be pit tagged to enable identification of individual fish by electronic scanning. Broodstock, a male and a female separated by a perforated plastic sheet, are kept in 20 gallon aquariums at 28°C. When the female becomes ripe, her eggs are stripped into a dry petri dish and the male's milt (semen) are stripped into a capillary tube. The tube's contents are mixed with the eggs and then water is added to the dish. After 2-3 minutes, the eggs are washed to remove excess sperm. Eggs are microinjected beginning within 5 minutes of fertilization. Fertilized eggs

are maintained in water at 21°C, which slows down division, so that the fertilized eggs remain unicellular prior to injection. This technique has been reported to permit microinjection of tilapia eggs for 2.5 hours after fertilization (13) .

Teleost fish eggs have micropyles, a single sperm- sized opening in the chorion. The micropyle diameter of a tilapia egg is 6 um. Micropyle injections are performed as described by Rahman and Maclean (13) . Eggs are positioned with the micropyle readily visible and then immobilized in an egg holder designed by the inventors. With the aid of an operating microscope and a micromanipulator, the needle tip is advanced down the micropyle and 250 pi -2nl of DNA solution injected using a "gene pusher". The needles, measuring 3-5 um at the end, are made from borosilicate glass tubes using an microelectrode puller. Following injection, the eggs are kept in plastic hatching funnels with a water flow rate of 0.75 1/minute until 10 days post-fertilization at which time the fry are transferred to an aquarium and allowed to grow (13) .

Once introduced into tilapia by microinjection, the humanized tilapia insulin gene should contain enough tilapia sequences so that homologous recombination should occur in a miniscule percentage of fish. Homologous recombination is presently used in mice to disrupt genes (i.e., knockout experiments) to determine the in vivo function of the non- functioning disrupted gene. Tilapia fin clippings can then be screened using polymerase chain reaction (PCR) (14) using primers that incorporate the human sequences and do not give a product with the fish genome. Because of the great sensitivity of PCR, one can initially screen fin clippings in batches; one would then need to screen individual fish only when a positive batch is identified. Individual fish expressing the transgene would be pit tagged at 3 months of age so that they could be identified through their lives by electronic scanning. Screening for Transgenic Fish

The objective of the breeding program is to develop a genetically stable production strain of transgenic tilapia

for use in tissue transplantation. This line should have the following properties: (1) absence of humanized tilapia DNA sequences other than the insulin locus (extraneous humanized DNA) , (2) homozygosity for the humanized insulin gene at the tilapia insulin locus, (3) genetic and developmental stability and uniformity, (4) good growth and survival characteristics, and (5) genetic identifiability for security against contamination and for protection of the proprietary interests of the developers. Initial Screening Procedures:

The breeding program will be based in part on the screening with four PCR primers: primer (AT) tilapia sequence just upstream of the 1 kb leading strand of the insert, primer (BT) tilapia sequence in the leading strand outside the humanized coding region, primer (CH) humanized tilapia sequence in the insulin coding region, and primer (CT) tilapia sequence homologous to CH. Homozygous Transgenic Founder Population:

The transgenic fish (generation GO) identified by initial screening for AT-CH PCR product will be heterozygotes at the insulin locus and may also contain extraneous humanized inserts elsewhere in the genome. These GO fish will be bred to as many wild-type individuals as possible, to produce a population containing 50% transgenic heterozygous offspring (Gl) Fish containing one copy of the transgene will be selected by screening for the AT-CH PCR product. These heterozygous fish will then be mated amongst themselves to produce 25% homozygous transgenic fish in generation G2. The G2 homozygotes will be selected out from the 50% heterozygotes by simultaneous screening for AT-CH and AT-CT PCR products.

The resulting homozygous transgenic founder population will be expanded by random mating in the Marine Gene Probe Laboratory (MGPL) tilapia facility. We will then produce sufficient offspring so that we can purify and sequence the transgenic insulin to confirm its amino acid sequence.

Purified Transgenic Population:

Extraneous humanized inserts (other than at the insulin locus) in the founder population will be screened and culled by southern blotting restriction digests with a humanized tilapia insulin probe. This will be done in the GO and Gl generation. If the extraneous humanized inserts are rare, they will be selected out of the founder population by screening restriction digests with the humanized tilapia insulin probe. If they are very abundant, they will be diluted out by several generations of recurrent backcross selection to the non-transgenic base population in the MGPL, followed by another round of mating to re-establish homozygosity. The strategy will be to backcross the transgene to several strains of tilapia in the MGPL. When extraneous inserts have been removed, the purified transgenic population will be expanded by random mating.

The possibility that tissue from transgenic tilapia may ultimately be transplanted into humans sets rigid requirements for the stability and predictability of the material. The fish used for transplantation and associated support studies should be genetically uniform. It seems undesirable to aim for an isogenic transgenic strain that is homozygous at all background (non-insulin) loci, because of the generally poor fitness and low developmental stability of highly inbred fish. Instead, the goal should be an isogenic, first generation hybrid of homozygous line crosses. Such hybrids will be genetically identical, heterozygous at many background loci, but homozygous for the humanized tilapia insulin gene. Sib-mating rather than gynogenesis is a preferred procedure for production of homozygous lines. Multiple inbred parental lines will be developed as a security measure against loss of lines through inbreeding depression during sib-mating. Transgenic Transplantation Strain:

Selection for fitness (growth and fecundity) in the inbred parental strains will be performed by within-family selection (15, 16) . Selection will be initially on the strains

themselves to reduce inbreeding depression and then secondarily on their performance in the isogenic transgenic hybrid. It is expected that selection for fitness will retard the approach of background loci towards homozygosity. This process will be controlled and monitored by extensive screening at microsatellite loci distributed throughout the genome. The final transgenic transplantation strain will be isogenic hybrid. The inbred parental lines will be sufficiently viable to create no risk of die-off during long term production of this strain. Genetic Identification and Security:

If necessary (for example for security or proprietary reasons) , the transgenic production strain may be identifiable as such, not only at the strain level but also at the chromosome level. This can be accomplished through the development of unique allele profiles at microsatellite loci on all chromosomes. If deemed necessary fish to be made available for transplantation can be made incapable of reproduction by hormonal sterilization. Non-homologous Recombination:

As detailed above, we will attempt for homologous recombination. However, the humanized tilapia insulin gene will contain its "natural" promoter and other upstream regulatory sequences. Therefore, if integrated into another site (i.e. non-homologous recombination), it should still function. Islet cells from these transgenics would produce regulatable levels of both human and fish insulin.

Fish expressing both human and fish insulin genes may also be selected for production. Islets from such fish should still regulate blood sugar levels and, therefore, be useful as islet donors. Furthermore, fish expressing both genes (or multiple copies of the humanized gene) may demonstrate enhanced growth potential and, therefore, might prove valuable for aquaculture. Harvesting BB's from Transgenic Fish

A new method of harvesting BBs using collagenase type II is described below. The procedure is much easier and less

time consuming, especially when applied for mass isolation; it has more than doubled the yield of islet tissue per fish and, therefore, decreased the need for donor fish by >50%. Histologically, BB preparations harvested in this manner are cleaner than BB preparations acquired by microdissection. The inventors have previously transplanted the BBs from tilapia isolated by this new method into athymic nude mice as well as into euthymic BALB/c mice and compared the results with those of using fish islets harvested by the standard microdissection method. The results indicated that this new isolation method does not alter either long-term function after transplantation in nude mice or mean graft survival time after transplantation in euthymic mice (4) .

Donor fish are housed at 28°C and fasted prior to sacrifice. Donor fish are anesthetized by placing them in a bucket of water containing 2-phenozyethanol (1 ml/L) or ethyl p-aminobenzoate (benzocaine) (200 mg/L) ; the fish are then weighed and given an identifying number. Next, each fish is placed on its left side in a dissecting tray. The donor fish is then killed by cutting the cervical spinal cord with a scalpel. Next, the peritoneal cavity is widely exposed by cutting along the ventral surface from the anus to the pericardial cavity; this cut is extended dorsally caudal to the right operculum (gill flap) and then caudally along the dorsal aspect of the peritoneal cavity to the anus. Finally, the triangular flap of musculointegumentary tissue bounded by these three cuts is removed (4) . The "BB region" adjacent to the bile duct is identified in situ (4) , excised, and placed into a petri dish (4) containing Hanks balanced salt solution (HBSS) containing 27 mM Hepes, 200 U/ml penicillin, and 200 ug/ml streptomycin sulfate (Gibco, Grand Island, NY) ; dish is labelled with the fish identifying number. The dish is then placed in a laminar flow hood and held until any additional donor fish have been processed through these steps.

Microscopically, these BB regions are composed of adipose tissue, blood vessels, nerves, bile ducts, and approximately 15 BBs (4) . The BB regions are collected and

kept in cold Hanks balanced salt solution (HBSS) containing 27 M HEPES, 200 U/ml penicillin, and 200 ug/ml streptomycin sulfate (Gibco, Grand Island, NY) . After collecting the required number of BB regions, they are rinsed and placed in a 50 ml plastic centrifuge tube containing 37°C pre-warmed collagenase solution (2 regions/ml) . The collagenase solution is prepared by dissolving collagenase type II (Sigma C-6885, St. Louis, MO) in HBSS at a concentration of 3 mg/ml. Next, each tube containing BB regions and collegenase solution is placed in a 37°C shaker/water bath. Digestions are carried out for approximately 15 ± 5 minutes. The optimal digestion time is judged visually by periodically stopping the digestion and holding the tube against a light. Digestion is stopped *rhen most of the small and moderate-sized BBs were clearly liberated while the largest BBs are still weakly attached to a few thread-like structures. The digestion is stopped by pouring cold HBSS into each tube. The tubes are centrifuged for 1 minute at 1000 RPM and the supernatant containing a layer of free adipocytes and the diluted collagenase is discarded. The pellet is washed twice with cold HBSS. The digested tissue is then resuspended in HBSS and placed into 100 mm plastic culture dishes. Under the dissecting microscope (lOx magnification) or with the naked eye plus good illumination, the released BBs can be easily identified from the residual debris composed of vessels, bile ducts, and connective tissue. The BBs appear spherical and whitish; they are less translucent than the connective tissue (4) . They are then handpicked using a siliconized pipette. Usually, each donor fish has one or two large BBs and 10-15 small or intermediate-sized BBs; the size of the BBs ranged from 0.3-5mm in maximum dimension (4) . The largest BBs often are intimately associated with a large caliber blood vessel and this is usually still weakly attached after digestion (4) ; however, this is easily separated with the help of microvascular scissors and jeweller's forceps. All collected BBs are then transferred into CMRL-1066 culture medium (Gibco, Grand Island, NY) plus 10% calf serum, 2,5 mg/ml D-glucose, lOOU/ml penicillin, and 100 ug/ml streptomycin

sulfate. The larger BBs are divided with microscissors to achieve a uniform size (< 0.5 mm in maximum dimension) . Then all BB fragments are cultured overnight free-floating at 37°C in 95% air/5% C0 ₂ . BB fragments are cultured overnight before transplantation to allow insulin to leak out of any damaged cells and to permit exocrine tissue attached to the BB fragments to degenerate prior to transplantation. Once all petri dishes are in the incubator, plasma glucose levels for each donor fish are determined and recorded by donor fish identifying number. After overnight culture, the BB fragments are ready for transplantation. Functional Studies

Prior to clinical use, the functional characteristics of the transgenic islets need to be characterized. This is to be done in vitro and in vivo. As a first step, the amino acid sequence of the secreted insulin from the transgenic islets is determined to confirm that it is humanized. The kinetics of tilapia insulin secretion by non-transgenic tilapia islets as well as its secretagogues are determined in vitro and compared to secretion studies using transgenic islets and a radioimmunoassay for human insulin. Because the structure of piscine insulin differs from human, porcine, bovine, and rat insulin amino acids, most antibodies produced against insulin of these species do not crossreact with an affinity sufficient for radioimmunoassay. Although a few antibodies have been raised against teleost fish insulin (17-21) , the insulin of one teleost species often differs from that of other teleost species. Therefore, an antibody to purified tilapia insulin (8) has been raised to permit insulin measurement by standard RIA methodology.

Further, the comparative effects of tilapia, humanized tilapia, and human insulin on mammalian muscles cells, one of the primary targets of insulin, will be studied. IDDM adversely affects glucose transport, glycogen synthesis, glucose oxidation, and glucose transporter GLUT-4 translocation and/or activity; these problems are normalized by appropriate insulin treatment. Therefore, the acute and chronic effects

of treatment with humanized tilapia insulin on glucose transport, glucose metabolism (glycogenesis, glycolysis, and oxidation) , insulin binding, and GLUT-4 translocation will be examined. In addition, some of these chronic studies will be performed on streptozotocin-diabetic athymic nude mice which have been made normoglycemic with BB grafts.

Transgenic islets will function appropriately in a mammalian recipient. This is done as in the inventors previous studies transplanting non-transgenic fish islets into nude mice (1,2) . Diabetes should be induced in recipient mice at least one week prior to transplantation with a single intravenous injection of streptozotocin (175-200 mg/kg) ; all recipient mice should have at least two successive non-fasting plasma glucose measurements between 350 and 500 mg/dl. For all mouse strains the inventors have tested, the optimal volume of tilapia BB fragments to insure normoglycemia and to minimize the probability of hypoglycemia immediately post transplantation is about 0.0025 kg of donor fish body weight/gm of mouse weight (eg., a 24 gm mouse would require about 0.6 kg of donor tilapia) . The total donor fish body weight can generally be provided by one large fish or multiple small fish. Tilapia weighing between 200 and 800 gms each are usually used as BB donors. Interestingly, donor fish body weight does not seem to correlate closely with BB weight, consequently total donor fish weight seems to be the better method for quantifying the amount of tissue required for transplantation into a particular recipient mouse.

At least 30 minutes before transplantation, BB fragments are removed from the incubator, switched into CMRL media containing 4.0 mg/ml D-glucose, and then returned to the incubator for 30 minutes. This steps further degranulates the islets and serves to protect the recipient from acute hypoglycemia during the first 24-48 hours after transplantation (i.e, when insulin is most likely to leak out of any dead or dying cells) . After this 30 minute incubation period, BB fragments are placed in complete media containing 2.5 mg/ml D- glucose and then returned to the incubator until the time they

are transplanted. Immediately prior to transplantation, the BB fragments are washed in incomplete media (i.e., without fetal calf serum) .

The diabetic nude mouse recipient is anesthetized with an i.p. injection of 50-55 mg/kg sodium pentobarbitol, placed on its right side and secured. The left side is shaved and then cleansed with an iodine antiseptic solution followed by alcohol. The left kidney is palpated and then delivered through a 0.75 cm subcostal incision at the costovertebral angle. The externalized kidney is secured and moistened with Hepes-HBSS. While visualized through a dissecting microscope, the renal capsule is gently lifted with jeweller's forceps and a 0.2 cm incision is made with the cutting edge of a 25 gauge needle. A "hockey stick-shaped" glass microspatula is used to separate the capsule from the surface of the kidney -- thus creating a space into which the BB tissue may be transplanted. Great care must be exercised to prevent tearing of the renal capsule with the microspatula.

BB fragments are placed beneath the kidney capsule by gently lifting the capsule with curved jeweller's forceps and then pushing each fragment under the capsule with the curved glass microspatula. The surface of the kidney should be frequently moistened with hepes-HBSS during this process. Once all of the BB fragments.are under the kidney capsule, they should be distributed evenly so that large matted masses of BB tissue are avoided. The incision in the renal capsule is then sealed using a fine tip, low-temperature ophthalmic cautery (Xomed-Treace, Jacksonville, FL) . The kidney is then returned to its normal position and the muscular layer and the skin are closed individually with 5-0 silk suture (2, 22) . Transplant Assessment a. In vivo graft function

Mice generally become normoglycemic (<200 mg/dl) several hours post-transplant. Recipient mice should be observed occasionally in the post-operative period for signs of hypoglycemia. During the first 24-48 hours post transplantation, insulin release is poorly regulated and mice

may fall victim to hypoglycemia. Hypoglycemic mice should be treated immediately with i.p. glucose. If sufficient volume of BB tissue has been transplanted, mice should remain normoglycemic until rejection. Mean non-fasted plasma glucose levels after transplantation of tilapia BB fragments into nude mice are in the range of 70-100 mg/dl (2) ; however, mean plasma glucose levels tend to be somewhat lower during the first week after transplantation. BB grafts function physiologically in nude mice; long-term recipient mice have relatively normal glucose tolerance curves (2) . It is of interest that mean fasting plasma glucose levels (72 mg/dl) in long-term recipient mice are nearly identical to those of the donor fish (75 mg/dl ⁾ (2) . The inventors have found that, if non-fasting plasma glucose levels fluctuate widely, it is likely that insufficient BB tissue was transplanted (23) . b. Histology

Histological assessment is required when each recipient mouse is sacrificed. First of all, the recipient's native pancreas should be examined histologically for evidence of Beta cell regeneration; this is accomplished by staining histologic sections for insulin with aldehyde fuchsin (24) or immunoperoxidase.

Next, each left kidney should be examined for graft viability and the degree of Beta ce.ll granulation. Beta cell granulation can be examined by immunoperoxidase staining using the inventors' antibody to tilapia insulin. Sections of the graft-bearing kidney should also be stained with hematoxylin and eosin. c. Single cell preparations

There may be some advantage to converting the BB fragments to single cell suspensions. This is associated with an additional advantage in that single cells and small cell agglomerates display a much higher surface area to volume ratio than whole BBs or BB fragments and, therefore, are less susceptible to limitations in the diffusion of oxygen and nutrients (25) .

BBs can be easily dispersed into single cells. BBs are placed in a 15 cc plastic conical centrifuge tube and washed twice with calcium-magnesium-free HBSS. This is aspirated and the BB are resuspended in 7 ml of versene (Gibco #670-5040AG, Grand Island, NY) at room temperature for seven minutes. The BBs are sedi ented and the supernatant removed. Next, 4 ml of a trypsin solution (Sigma #T-0646, St. Louis, MO; 1 mg/ml in calcium-magnesium-free HBSS filtered through a 0.22 um syringe filter) is added to the tube and the tube is then placed in a 37°C water bath inside a laminar flow hood; the trypsin/BB solution is then gently pipetted up and down constantly. After two minutes, the larger fragments are allowed to sink to the bottom and the supernatant containing free islet cells is removed and immediately transferred to a tube containing complete tissue culture media which is then placed on ice. An additional 4 ml of the trypsin solution is added to the pellet and this process is repeated three times. Each time the supernatant is pooled and diluted with complete media. Finally, the tube containing the supernatant is centrifuged at 600 g for 5 minutes at 4°C. The supernatant is removed and the cells are resuspended in complete tissue culture media and counted. Viability, as assessed by trypan blue exclusion, is about 95%. Compared to dispersed rodent or human islet cells, dispersed BB cells have less tendency to reaggregate into clumps or chains of cells after overnight culture although some reaggregation will occur as illustrated in Figure 10. Although single cell suspensions immobilized in plasma clots will produce normoglycemia when transplanted under the renal capsules of diabetic nude mice, data suggests that single cell preparations do not function in vivo quite as well as BB fragments -- probably because cell-cell communications are diminished (26) .

As an alternative to single cell methods, mammalian islet-sized BBs ("micro-BBs") can be produced with a Cellector Tissue Sieve (Bellco Glass, Vineland, NJ) simply by pushing BB fragments sequentially through a sterile stainless steel #60 mesh (opening size 230 um) filter and then through a sterile

#80 mesh (opening size 190 um) filter. Both filters should be carefully washed with excess media. All filtrates should be collected in plastic petri dishes and cultured overnight. Because BBs are composed of almost pure endocrine cells and have minimal connective tissue stroma, very little tissue should remain on the filters after several washes. Although the "micro-BBs" initially appear ragged when the dishes are examined with an inverted microscope, they "round up" nicely after overnight culture and appear remarkably similar to cultured mammalian islets. After culture, "micro-BBs" can be hand-picked with a Pasteur pipette and counted while visualized through a dissecting microscope housed within a laminar flow work station. It seems likely that "micro-BBs" may function better than single cells because some cell-cell communication remains intact. Unlike BB fragments, "micro-BBs" are small enough that they can be transplanted into sites other than under the kidney capsule. Alternatively, the BBs can simply be teased apart and then cultured which will produce similar mammalian islet-sized BBs. We have previously shown that micro-BBs prepared in this fashion from non-transgenic tilapia can be successfully transplanted into the testes of diabetic nude mice and will produce long-term normoglycemia. Transplantation Into Humans

BBs from transgenic fish of the present invention, once characterized, can be harvested, encapsulated, and surgically transplanted into diabetic patients. The BBs can be encapsulated according to known techniques in the art for encapsulating islet cells (27-29) . We have previously shown that macroencapsulated BBs from non-transgenic tilapia will provide long-term normoglycenia and mammalian-like glucose tolerance profiles after intraperitoneal transplantation in mice (30) .

One problem that has plagued investigators working with the various artificial pancreas technologies is that much of initial islet volume is lost after encapsulation because of central necrosis secondary to hypoxia in the centre of the islets. It is very likely that piscine islet tissue would be

superior to mammalian islet tissue for encapsulation because fish tissues, including islets, can tolerate lower oxygen tensions (31) . Even if glucose-stimulated insulin secretion by the transgenic fish islets proved to be somewhat less efficient or slightly slower than that of human islets, they still might prove exceedingly useful for artificial pancreases with a large number of fish islets and a much smaller number of human islets; conceivably the fish islets could provide the bulk of the insulin production and the human islets could "fine tune" the insulin response just as short-acting and long-acting insulins are now mixed to optimize insulin therapy in diabetics.

Transgenic fish islets described herein represent a nearly unlimited supply of human insulin-producing islet tissue that would not require expensive enzymatic islet isolation procedures for procurement. The utility of these islets is likely to be further increased in that we have shown that tilapia BBs can be cryopreserved in liquid nitrogen (using standard technique developed for cryopreserving mammalian islets) , thawed, and then transplanted into diabetic nude mice confirming islet function. This would facilitate shipping and long-term storage of the transgenic islets. Other Applications of the Invention

As described in detail above, the transgenic fish of the present invention are useful in providing large amounts of humanized insulin for the treatment of diabetes. This occurs through the homologous recombination of the humanized tilapia insulin gene. However, if non-homologous recombination occurs, the transgenic fish may express both human and fish insulin genes. Such fish may be useful in aquaculture as fish expressing both genes (or multiple copies of the human gene) may demonstrate enhanced growth potential.

One skilled in the art can readily appreciate that various modifications can be made to the invention without departing from the scope and spirit of invention. Modifications intended to be within the scope of the invention include modifications or substitutions to the sequence of the

humanized insulin gene which do not substantially affect the biological activity or immunogenicity of the insulin product in humans.

REFERENCES

1. Wright, JR Jr: Experimental transplantation using principal islets of teleost fish (Brockmann bodies) . In Pancreatic Islet Cell Transplantation: 1892 - 1992 - One Century of Transplantation for diabetes, edited by C. Ricordi, R.G. Landes Co., Austin, 1992, p.336-351.

2. Wright JR Jr., Polvi S and Maclean H: Experimental transplantation using principal islets of teleost fish (Brockmann bodies) : Long-term function of tilapia islet tissue in diabetic nude mice. Diabetes 41:1528-32.1992.

3. Wright JR Jr. Preparation of Fish Islets (Brockmann bodies) In: Lanza RP; WL Chick, eds. Pancreatic Islet Transplantation genes Vol. 1. Procurement of Pancreatic Islets. Austin: RG Landes co. ; (1994:123-32) .

4. Yang H and Wright JR Jr: A method for mass harvesting islets (Brockmann bodies) from teleost fish. Cell transplant. 4:621-8, 1995.

5. Hackett PB: The molecular biology of transgenic fish. In: Biochemistry and molecular Biology of Fishes, 2, Hochachka P and Mommesen T (eds.) Amsterdam: Elsevier, 1993.

6. Hew CL and Fletcher G, eds. : Transgenic Fish. New Jersey: World Scientific, 1992.

7. Mommesen TP and Plisetskaya EM: Insulin in fishes and agnathans : history, structure, and metabolic regulation. Rev. Aquatic Sci. 4:225-59, 1991.

8. Nguyen TM, Wright JR Jr Nielson PF, and Conlon JM: Characterization of the pancreatic hormones from the Brockmann body of the tilapia-implications for islet xenograft studies. Comp. Biochem. and Physiol. lllC:33-44, 1995.

9. Hobart PM, Shen L-P, Crawford R. Pictet RL and Ruttner WJ: Comparison of the nucleic acid sequence of anglerfish and mammalian insulin mRNA's from cloned CDNA's. Science 210: 1360- 3, 1980.

10. Chan SJ, Cao Q-P, Nagamatsu S, and Steiner DF: Insulin and insulin-like growth factor genes in fishes and other primitive chordates. In: Biochemistry and molecular Biology of Fishes, 2. Hochachka P. and Mommesen T. (eds.) Amsterdam: Elsevier, 1993, pp. 407-17.

11. Steiner DF: Structure and evolution of the insulin gene. Ann. Rev. Genet. 19: 463-84, 1985.

12. Penman DJ, Beeching AG, Penn S, and MacLean N: Factors affecting survival and integration following

microinjection of novel DNA into rainbow trout eggs. Aquaculture 85: 35-50, 1990.

13. Rahman MA and Maclean N: Production of transgenic tilapia (Oreochromis niloticus) by one-cell-stage microinjection. Aquaculture 105: 219-32, 1992.

14. Saiki RK, Gelfand DH, Stoffel S, Scharf SJ, Higuchi R, Horn GT, Mullis KB, and Erlich HA: Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239: 487-91, 1988.

15. Uraiwan S and Doyle RW: Replicate variance and the choice of selection procedure for tilapia (Oreochromis nilotic) stock improvement in Thailand. Aquaculture 57: 93-8, 1986.

16. Doyle RW and Herbinger C: The use of DNA fingerprinting for high-intensity, with-in family selection of fish breeding. 5th World Congress, Genetics Applied to Livestock Production. Guelph, Ontario, Canada, pp.364-371.

17. Furuchi M. Nakamura Y. Yone Y. A radioimmunoassay method for determination of fish plasma insulin. Bull Jap Soc. Sci Fish 1980; 46:1177-81.

18. Patent TN, Foa PP. Radioimmunoassay of insulin in fishes: Experiments in vivo and in vitro. Gen Comp Endocrinol 1971: 16:41-6.

19. Plisetskaya EM, Leibush BN. Radioimmunological determination of insulin in lower vetebrates. Zh Evol Biokhim Fiziol 1974; 10:623-5.

20. Thorpe A. Ince BW. Plasma insulin levels in teleosts determined by a charcoal-separation radioimmunoassay technique. Gen Comp Endocrinol 1976: 30:332-9.

21. Tilzey JF, Waights V, Holmes R. The development of a homologous teleost insulin radioimmunoassay and its use in the study of adrenaline on insulin secretion from isolated pancreatic islet tissue of the rainbow trout. Salmo gairdnerii

(R.) . Comp Biochem Physiol. 1985: 81A-821-5.

22. Wright JR Jr, Kearns II, Polvi S et al . Experimental xenotransplantation using principal islets of teleost fish

(Brockmann bodies) : Graft survival in selected strains of inbred mice. Transplant Proc. 1994; 26:770.

23. Wright JR Jr, Polvi S, Schrezen eir J., Al-Abdullah I. Longterm function of teleost fish principal islets (Brockmann bodies) after transplantation under the renal capsule in diabetic nude mice. Transplant Proc. 1992: 24:3029- 30.

24. Culling CFA, Allison RT, Barr WT, Aldehyde fuchsin (Halmi, 1952) , In: Cellular Pathology Technique 4th ed. ,

London: Buttherworths, 1985: 47808.

25. Schrezenmeir J, Laue Ch, Sternheim ET et al . , Long- term function of single-cell preparations of piscine principal islets in hollow fibers. Transplant Proc. 1992; 24:2941-5.

26. Pipeleers D, In't Veld P, Maes E, and Van de Winkel M: Glucose-induced insulin release depends on functional cooperation between islet cells. Proc. Natl. Acad. Sci. USA 79: 7322-5, 1982.

27. Iwata H, Kobayashi K, Takagi T, Oka T, Yang H, and Amemiya H: Feasibility of agarose microbeads with xenogeneic islets as a bioartifical pancreas. J. Biomed. Mater, Res. 28: 1003-11, 1994.

28. Iwata H, Takagi T, Kobayashi K, Oka T, Tsuji T, and Ito F: Strategy for developing microbeads applicable to islet xenotransplantation into a spontaneous diabetic NOD mouse. J. Biomed. Materials Res, 28: 1201-7, 1994.

29. Lanza RP, Ecker D, Kuntreiber WH, Staruk JE, Marsh J, and Chick WL: A simple method for transplantating discordant islets into rats using alginate gel spheres. Transplantation 59:1485-9, 1995.

30. Yang H & Wright JR Jr: Long-term function of fish islet xenografts in mice by alginate macroencapsulation. Abstract submitted for International Congress of the Transplantation Society, Barcelona.

31. Schrezenmeir J, Kirchgessner H, Gero L, Kunz LA, Beyer J, and Mueller-Klieser W: Effect of microencapsulation on oxygen distribution in islets organs. Transplantation 57: 1308-14, 1994.

SEQUENCE LISTING

(1) GENERAL INFORMATION:

(i) APPLICANT:

(A) NAME: DALHOUSIE UNIVERSITY

(B) STREET: TECHNOLOGY TRANSFER OFFICE

(C) CITY: HALIFAX

(D) STATE: NOVA SCOTIA

(E) COUNTRY: CANADA

(F) POSTAL CODE (ZIP) : B3H 3J5

(G) TELEPHONE: 902-494-1648 (H) TELEFAX: 902-494-5189

(ii) TITLE OF INVENTION: TRANSGENIC FISH IN THE TREATMENT OF DIABETES

(iii) NUMBER OF SEQUENCES: 7

(iv) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Floppy disk

(B) COMPUTER: IBM PC compatible

(C) OPERATING SYSTEM: PC-DOS/MS-DOS

(D) SOFTWARE: Patentin Release #1.0, Version #1.30 (EPO)

(vi) PRIOR APPLICATION DATA:

(A) APPLICATION NUMBER: US 08/417,866

(B) FILING DATE: 06-APR-1995

(2) INFORMATION FOR SEQ ID NO: 1:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 21 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(v) FRAGMENT TYPE: N-terminal

(vi) ORIGINAL SOURCE:

(A) ORGANISM: TILAPIA INSULIN A CHAIN

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 1:

Gly He Val Glu Glu Cys Cys His Lys Pro Cys Thr He Phe Asp Leu 1 5 10 15

Gin Asn Tyr Cys Asn 20

(2) INFORMATION FOR SEQ ID NO: 2:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 30 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(vi) ORIGINAL SOURCE:

(A) ORGANISM: TILAPIA INSULIN B CHAIN

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 2:

Val Gly Gly Pro Gin His Leu Cys Gly Ser His Leu Val Asp Ala Leu 1 5 10 ¹⁵

Tyr Leu Val Cys Gly Asp Arg Gly Phe Phe Tyr Asn Pro Arg 20 25 30

(2) INFORMATION FOR SEQ ID NO: 3:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 21 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(vi) ORIGINAL SOURCE:

(A) ORGANISM: HUMAN INSULIN A CHAIN

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 3:

Gly He Val Glu Gin Cys Cys Thr Ser He Cys Ser Leu Tyr Gin Leu 1 5 10 15

Glu Asn Tyr Cys Asn 20

(2) INFORMATION FOR SEQ ID NO: 4:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 30 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(vi) ORIGINAL SOURCE:

(A) ORGANISM: HUMAN INSULIN B CHAIN

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 4:

Phe Val Asn Gin His Leu Cys Gly Ser His Leu Val Glu Ala Leu Tyr 1 5 10 15

Leu Val Cys Gly Glu Arg Gly Phe Phe Tyr Thr Pro Lys Thr 20 25 30

(2) INFORMATION FOR SEQ ID NO: 5:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 2078 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

( ii ) MOLECULE TYPE : DNA ( genomic )

(vi ) ORIGINAL SOURCE :

(A) ORGANISM : TILAPIA INSULIN GENE

(xi ) SEQUENCE DESCRIPTION : SEQ ID NO : 5 :

GTCCCCATAA TCGCACACAA GTCCCCACAA TGTAGGTGAA ATAGGTTCCA CGGAAACACG 60

TGGAACAGGG GGTGTGTCCA GGTGGTGCTG GTGGAGTATA AATGGAGAGA AGGCTCTTGG 120

TTCTGCCTCA CACAGAAAAG CTGCTCCTGC CCTTCATCTC AGAGTTACCT CCTCCTCTCT 180

GTCTGTGCAG GTGAGTGCTG GCTGTAGGTT TGGTTGTGAG GACAGTGACT GTGATGCTAA 240

CGTGAATGTG CTTTTGTGTT CAGCTCTTTT CCAGCATGGC AGCGCTCTGG CTCCAGGCCT 300

TCTCCCTGCT CGTCTTAATG ATGGTTTCGT GGCCGGGCTC CCAGGCCGTC GGTGGGCCAC 360

AGCACCTGTG CGGCTCCCAC CTGGTGGATG CCCTGTACCT GGTCTGTGGG GACAGAGGCT 420

TCTTCTACAA CCCCAGGAGA GATGTGGACC CTCTGCTTGG TGAGACCACC AACCACAAAC 480

AGAAACACTA GACAAACTAT TTGAGGGCAG CTTTTCTTTC TCTGAGTTCA CTTTAAATCA 540

GCTTTCATGT TGGAAACATG GTAATAGTAA TTTTCCATAT CTTTATGGAC CCTACATGAT 600

TAGTTTACAT CTATTGCCAT TTGTCTCAAC ACCTGCATCA TATAATAGCG CTGATTTTGT 660

AACACTTTGT GTTAGAATTA GTAATTTATG TTCTAATAAT GTTTGATATG TATTCTTTAA 720

TATAAATGAC CAGAATTTTT AGATCTGAAC ATTCACCTGC TCTTCATCCC ATCAGGTTTC 780

CTCCCTCCAA AGGCAGGTGG TGCTGTGGTG CAAGGTGGTG AGAATGAAGT GACCTTCAAA ^' 840

GACCAGATGG AAATGATGGT GAAGCGAGGC ATTGTGGAGG AATGCTGTCA CAAACCCTGT 900

ACCATCTTCG ACCTGCAGAA CTACTGCAAC TGAACTGCTC TGCTGGACTT TGTTTAGTCG 960

AGCCAGGCTC GGCTATTCAG GTCTGAGTCC CAGCCCCACC TCGCTCCCTG CTTCAGAGGA 1020

GAGCCACAGC TGTCCTCTCT CTGAAAACCA ACTGCTGTCA AATGAAGTGC TGAGAAATGG 1080

ATAAAATTAA TTTTCCAAGA AATAAAAATG CAAAATGTGA CAACGTGAGG CAAAAAAGTG 1140

TGTTCTTTTG TTGTGATGAA TTCAGTTAAT TGATTAAAGT GAAAACTCGA ACATGTTAGG 1200

TACCTGCTGC TATCCAGCAC AAACTGCTGA GCTTTCACTT TCCAAAGCTT TGTGTTTAGC 1260

TTATAGTGTC TCTGAACAGG ATATAAACAC ATCATGCACT CTGACATGAT GTCCTTTTCA 1320

AACAATCCCT TGTCATCTTC ATTTCAGCAG GTCAGTGTTT TTTATTCAGG TCCTCGTGAT 1380

GACACAGAAG ATAAAAACAC CAAGTATTCT AAAAATTATC AAATTGAATT TTAAGTTCAA 1440

AAGCATTCTT CCATCACAGT CAACAGAACC CCAAGACCTG AAGTTCCAAA GGCCTGTGGT 1500

GTTACCACTA TGCTATCTAC ATATGTTACC TGCTTTTAAC TATTAAACGG AGCAGATGGA 1560

TCAGAAGGTT AATAGCTGAT CAGATCATGT CAGCTCATTA GCTTCAGTTT GTTTTACTGA 1620

GTGCTGTAAC CACTCAATCA GAAACACACT GTTACTTAAT CTGAGTACAT ACTTGTATAC 1680

ATTAAACTTG GAAAAAGATA GATGTGAAAT GTAAAAGTGA GGTCAATCGT CAAATGTGAC 17 0

ACGATATTTG GATCTGTTTA TCCTTCCAGG ATCACGGGGC GGGGAGAGCC TATCCCAGCT 1800

ACCATAGTGC GAGAGGTCAG GTACACCCTG GACAGGTCAA CAGCCTGTCA CAGGGTTTAC 1860

GCTCAGAGAC AAGCAACCAC GCCACTTCGG CTTATACCAG TATTATTTCC TAAATGTTGC 1920

CAATAAAAAA CAAAATCAGT AGAATTTTAA GCAGTTTTCA TTTTAATTTA ACCTCATTTG 1980

AAGAAGAAGT CAGAGGTCCA AAGTATGGGA ATATTTATAA TTCCAATGTT GTCAATTCAA 2040

ATAATGGCAA TAAGAAACAT AGTTTGAAAT AGAATAAG 2078

( 2 ) INFORMATION FOR SEQ ID NO : 6 :

( i ) SEQUENCE CHARACTERISTICS :

(A) LENGTH : 2072 base pairs

(B ) TYPE : nucleic acid

( C) STRANDEDNESS : single

(D) TOPOLOGY : linear

( ii ) MOLECULE TYPE : DNA ( genomic )

(vi ) ORIGINAL SOURCE :

(A) ORGANISM : HUMANIZED TILAPIA INSULIN NO . l

(xi ) SEQUENCE DESCRIPTION : SEQ ID NO : 6 :

GTCCCCATAA TCGCACACAA GTCCCCACAA TGTAGGTGAA ATAGGTTCCA CGGAAACACG 60

TGGAACAGGG GGTGTGTCCA GGTGGTGCTG GTGGAGTATA AATGGAGAGA AGGCTCTTGG 120

TTCTGCCTCA CACAGAAAAG CTGCTCCTGC CCTTCATCTC AGAGTTACCT CCTCCTCTCT 180

GTCTGTGCAG GTGAGTGCTG GCTGTAGGTT TGGTTGTGAG GACAGTGACT GTGATGCTAA 240

CGTGAATGTG CTTTTGTGTT CAGCTCTTTT CCAGCATGGC AGCGCTCTGG CTCCAGGCCT 300

TCTCCCTGCT CGTCTTAATG ATGGTTTCGT GGCCGGGCTC CCAGGCCTTC GTGCAGCAGC 360

ACCTGTGCGG ATCCCACCTG GTGGAGGCCC TGTACCTGGT CTGTGGGGAG AGAGGCTTCT 420

TCTACACCCC CAAGAGAGAT GTGGACCCTC TGCTTGGTGA GACCACCAAC CACAAACAGA 480

AACACTAGAC AAACTATTTG AGGGCAGCTT TTCTTTCTCT GAGTTCACTT TAAATCAGCT 540

TTCATGTTGG AAACATGGTA ATAGTAATTT TCCATATCTT TATGGACCCT ACATGATTAG 600

TTTACATCTA TTGCCATTTG TCTCAACACC TGCATCATAT AATAGCGCTG ATTTTGTAAC 660

ACTTTGTGTT AGAATTAGTA ATTTATGTTC TAATAATGTT TGATATGTAT TCTTTAATAT 720

AAATGACCAG AATTTTTAGA TCTGAACATT CACCTGCTCT TCATCCCATC AGGTTTCCTC 780

CCTCCAAAGG CAGGTGGTGC TGTGGTGCAA GGTGGTGAGA ATGAAGTGAC CTTCAAAGAC 840

CAGATGGAAA TGATGGTGAA GCGAGGCATT GTGGAGCAAT GCTGTACCTC CATTTGTTCC 900

CTGTACCAGC TGGAGAACTA CTGCAACTGA ACTGCTCTGC TGGACTTTGT TTAGTCGAGC 960

CAGGCTCGGC TATTCAGGTC TGAGTCCCAG CCCCACCTCG CTCCCTGCTT CAGAGGAGAG 1020

CCACAGCTGT CCTCTCTCTG AAAACCAACT GCTGTCAAAT GAAGTGCTGA GAAATGGATA 1080

AAATTAATTT TCCAAGAAAT AAAAATGCAA AATGTGACAA CGTGAGGCAA AAAAGTGTGT 1140

TCTTTTGTTG TGATGAATTC AGTTAATTGA TTAAAGTGAA AACTCGAACA TGTTAGGTAC 1200

CTGCTGCTAT CCAGCACAAA CTGCTGAGCT TTCACTTTCC AAAGCTTTGT GTTTAGCTTA 1260

TAGTGTCTCT GAACAGGATA TAAACACATC ATGCACTCTG ACATGATGTC CTTTTCAAAC 1320

AATCCCTTGT CATCTTCATT TCAGCAGGTC AGTGTTTTTT ATTCAGGTCC TCGTGATGAC 1380

ACAGAAGATA AAAACACCAA GTATTCTAAA AATTATCAAA TTGAATTTTA AGTTCAAAAG 1440

CATTCTTCCA TCACAGTCAA CAGAACCCCA AGACCTGAAG TTCCAAAGGC CTGTGGTGTT 1500

ACCACTATGC TATCTACATA TGTTACCTGC TTTTAACTAT TAAACGGAGC AGATGGATCA 1560

GAAGGTTAAT AGCTGATCAG ATCATGTCAG CTCATTAGCT TCAGTTTGTT TTACTGAGTG 1620

CTGTAACCAC TCAATCAGAA ACACACTGTT ACTTAATCTG AGTACATACT TGTATACATT 1680

AAACTTGGAA AAAGATAGAT GTGAAATGTA AAAGTGAGGT CAATCGTCAA ATGTGACACG 1740

ATATTTGGAT CTGTTTATCC TTCCAGGATC ACGGGGCGGG GAGAGCCTAT CCCAGCTACC 1800

ATAGTGCGAG AGGTCAGGTA CACCCTGGAC AGGTCAACAG CCTGTCACAG GGTTTACGCT 1860

CAGAGACAAG CAACCACGCC ACTTCGGCTT ATACCAGTAT TATTTCCTAA ATGTTGCCAA 1920

TAAAAAACAA AATCAGTAGA ATTTTAAGCA GTTTTCATTT TAATTTAACC TCATTTGAAG 1980

AAGAAGTCAG AGGTCCAAAG TATGGGAATA TTTATAATTC CAATGTTGTC AATTCAAATA 2040

ATGGCAATAA GAAACATAGT TTGAAATAGA AA 2072 ( 2 ) INFORMATION FOR SEQ ID NO : 7 :

( i ) SEQUENCE CHARACTERISTICS :

(A) LENGTH : 2075 base pairs

(B ) TYPE : nucleic acid

( C) STRANDEDNESS : single

(D) TOPOLOGY : linear

(ii) MOLECULE TYPE: DNA (genomic)

(vi) ORIGINAL SOURCE:

(A) ORGANISM: HUMANIZED TILAPIA INSULIN NO.2

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 7:

GTCCCCATAA TCGCACACAA GTCCCCACAA TGTAGGTGAA ATAGGTTCCA CGGAAACACG 60

TGGAACAGGG GGTGTGTCCA GGTGGTGCTG GTGGAGTATA AATGGAGAGA AGGCTCTTGG 120

TTCTGCCTCA CACAGAAAAG CTGCTCCTGC CCTTCATCTC AGAGTTACCT CCTCCTCTCT 180

GTCTGTGCAG GTGAGTGCTG GCTGTAGGTT TGGTTGTGAG GACAGTGACT GTGATGCTAA 240

CGTGAATGTG CTTTTGTGTT CAGCTCTTTT CCAGCATGGC AGCGCTCTGG CTCCAGGCCT 300

TCTCCCTGCT CGTCTTAATG ATGGTTTCGT GGCCGGGCTC CCAGGCCTTC GTGCAGCAGC 360

ACCTGTGCGG ATCCCACCTG GTGGAGGCCC TGTACCTGGT CTGTGGGGAG AGAGGCTTCT 420

TCTACACCCC CAAGACCAGA GATGTGGACC CTCTGCTTGG TGAGACCACC AACCACAAAC 480

AGAAACACTA GACAAACTAT TTGAGGGCAG CTTTTCTTTC TCTGAGTTCA CTTTAAATCA 540

GCTTTCATGT TGGAAACATG GTAATAGTAA TTTTCCATAT CTTTATGGAC CCTACATGAT 600

TAGTTTACAT CTATTGCCAT TTGTCTCAAC ACCTGCATCA TATAATAGCG CTGATTTTGT 660

AACACTTTGT GTTAGAATTA GTAATTTATG TTCTAATAAT GTTTGATATG TATTCTTTAA 720

TATAAATGAC CAGAATTTTT AGATCTGAAC ATTCACCTGC TCTTCATCCC ATCAGGTTTC 780

CTCCCTCCAA AGGCAGGTGG TGCTGTGGTG CAAGGTGGTG AGAATGAAGT GACCTTCAAA 840

GACCAGATGG AAATGATGGT GAAGCGAGGC ATTGTGGAGC AATGCTGTAC CTCCATTTGT 900

TCCCTGTACC AGCTGGAGAA CTACTGCAAC TGAACTGCTC TGCTGGACTT TGTTTAGTCG 960

AGCCAGGCTC GGCTATTCAG GTCTGAGTCC CAGCCCCACC TCGCTCCCTG CTTCAGAGGA 1020

GAGCCACAGC TGTCCTCTCT CTGAAAACCA ACTGCTGTCA AATGAAGTGC TGAGAAATGG 1080

ATAAAATTAA TTTTCCAAGA AATAAAAATG CAAAATGTGA CAACGTGAGG CAAAAAAGTG 1140

TGTTCTTTTG TTGTGATGAA TTCAGTTAAT TGATTAAAGT GAAAACTCGA ACATGTTAGG 1200

TACCTGCTGC TATCCAGCAC AAACTGCTGA GCTTTCACTT TCCAAAGCTT TGTGTTTAGC 1260

TTATAGTGTC TCTGAACAGG ATATAAACAC ATCATGCACT CTGACATGAT GTCCTTTTCA 1320

AACAATCCCT TGTCATCTTC ATTTCAGCAG GTCAGTGTTT TTTATTCAGG TCCTCGTGAT 1380

GACACAGAAG ATAAAAACAC CAAGTATTCT AAAAATTATC AAATTGAATT TTAAGTTCAA 1440

AAGCATTCTT CCATCACAGT CAACAGAACC CCAAGACCTG AAGTTCCAAA GGCCTGTGGT 1500

GTTACCACTA TGCTATCTAC ATATGTTACC TGCTTTTAAC TATTAAACGG AGCAGATGGA 1560

TCAGAAGGTT AATAGCTGAT CAGATCATGT CAGCTCATTA GCTTCAGTTT GTTTTACTGA 1620

GTGCTGTAAC CACTCAATCA GAAACACACT GTTACTTAAT CTGAGTACAT ACTTGTATAC 1680

ATTAAACTTG GAAAAAGATA GATGTGAAAT GTAAAAGTGA GGTCAATCGT CAAATGTGAC 1740

ACGATATTTG GATCTGTTTA TCCTTCCAGG ATCACGGGGC GGGGAGAGCC TATCCCAGCT 1800

ACCATAGTGC GAGAGGTCAG GTACACCCTG GACAGGTCAA CAGCCTGTCA CAGGGTTTAC 1860

GCTCAGAGAC AAGCAACCAC GCCACTTCGG CTTATACCAG TATTATTTCC TAAATGTTGC 1920

CAATAAAAAA CAAAATCAGT AGAATTTTAA GCAGTTTTCA TTTTAATTTA ACCTCATTTG 1980

AAGAAGAAGT CAGAGGTCCA AAGTATGGGA ATATTTATAA TTCCAATGTT GTCAATTCAA 2040

ATAATGGCAA TAAGAAACAT AGTTTGAAAT AGAAT 2075