Overcoming barriers to the oral delivery of peptide and protein therapeutics to aquacultured organisms

Introduction

The United Nations (UN) estimates that by 2037 the world’s population will have exceeded 10 billion.

This rapid growth in humankind will bring with it severe difficulties, not only in terms of providing shelter, education, and health care, but also in maintaining current levels of protein nutrition. The Food and Agricultural Organization (FAO) of the UN acknowledge that global fishery output must be increased by at least 50% to offset projected shortfalls in dietary protein by 2030. At present, production by traditional fisheries and aquaculture has reached approximately 150 million tons. Of this total, commercial and artisanal fisheries account for around 94 million tons of seafood. Aquaculture production exceeded 30 million tons in 2000. An estimated 30 million tons of seafood was discarded at sea. The latter comprised species of little or no current commercial value and represented approximately 20% of total fishery production.

Accordingly, if a food use can be found for presentday discards, a further 30-40 million tons must be produced in order to achieve a 50% increase in global fishery production. Realistically, this shortfall can only be met by the exploitation of new fisheries resources and by emplacement of enhanced fisheries management practices, positions that at present would appear untenable, since most of the world’s fisheries have already exceeded maximum sustainable yields or are being fished at optimal levels. Aquaculture, therefore, presents the only method of offsetting predicted fishery shortfalls but the industry will have to more than double present day harvests. Although such a prospect would seem improbable, it is noteworthy that for almost a quarter of a century, aquaculture has represented the fastest growing sector of the agricultural industry and, at times, has recorded annual increases in harvest close to 20%. Annual production of four species of fish (silver, grass, common and bighead carp) exceeds 1 million metric tons (mmt) while combined production of all tilapias and salmonids likewise achieve 1 mmt each, yearly.

The sustained growth of aquaculture has been achieved mainly by bringing into play untapped aquatic resources and by intensifying production per unit area of existing operations. Nevertheless, the major proportion of global aquaculture remains in the domain of extensive and semi-intensive farming systems. Realization of a doubling in productivity by aquaculture will demand an even greater understanding of animal life cycles and how these may be beneficially manipulated to increase yields.

Incontestable is that the continued success of the industry, on a global basis, will be heavily dependent upon the emplacement of novel biotechnologies and the development of innovative engineering systems that will permit the environmentally sustainable cultivation of food fish at high densities (Mayer and McLean, 1995). Critical to farming of all species will be an advanced understanding and control over reproduction, larval rearing, nutrition, growth, product quality, disease, and other production-related issues.

Many of the latter may be gainfully influenced by an increasing variety of therapeutic molecules (peptides, proteins, oligonucleotides/nucleic acids, antibacterial, antifungal and other chemotherapeutic agents) produced by fermentation, combinatorial chemistry, and recombinant DNA methods. The exploitation of these new generation products by aquaculture however, has been hindered due to the lack of appropriate delivery systems. The preferred route of bioactive delivery to aquacultured organisms is oral. Due to various physical, chemical, and immunological factors the gastrointestinal (GI) administration of therapeutic molecules however, is inherently difficult. Nevertheless, because oral drug delivery systems offer substantial advantages over traditional routes of administration, interest in the development of aquaculture-specific oral formulations has increased dramatically.

Recognizing that the GI tract represents an intrinsically hostile environment for many potentially useful therapeutic molecules, it is not surprising that emphasis has been placed on developing methods that protect such drugs from the degrading action of the gut. The following provides brief consideration of strategies that might be employed in designing oral drug formulations for fish. Prominence is given to examples that employ bioactive peptides and proteins, since many production-related processes may be profitably manipulated by the same. Success in the design of efficacious oral formulations for cultivated fish clearly demands a thorough appreciation of GI anatomy and physiology, since such information provides insight into tactics that may be used to protect or enhance the absorption of specific molecules.


Anatomy and physiology of the teleost gastrointestinal tract

Teleosts present a high level of morphological variation in the digestive tract and, indeed, express greater diversity and plasticity in this organ than any other vertebrate group. The teleost gut extends from simple tube-like structures, void of loops or folds, and unvarying in diameter, through to more fully differentiated tracts reminiscent of higher, monogastric plans. Between these two extremes are systems that are agastric in form, and intestines that have developed a variety of extensions, express intricate patterns of crenellation and include intense folds that project into the gut lumen, similar to the mammalian villus (Figure 1). Villi however, do not exist in fish: the discrete intestinal folds of teleosts do not contain the internal blood supply, lymph duct, or crypts of Lieberkuhn that characterize these structures in mammals. The relative length of the fish intestine generally correlates well with diet: carnivores have shorter intestines than herbivores, although exceptions to this rule occur (see Harder, 1975).

A regional functional differentiation of the fish gut has been described. Stroband et al. (1979) examined the absorption of reporter proteins by the gut of adult grass carp and observed that absorption was localized in the second or distal segment of the gut, but not in the rostral (first 68%) or caudal (last 9%) part. Thus, based upon micro- and macroanatomical features, the intestine may be divided into fore-, mid-, and hindgut. The presence of pyloric cecae in some carnivores (Figure 1) significantly increases gut surface area. Cytologically, the stomach, when present, mimics that of mammals, with the gastric mucosa presenting three cell types viz. oxyntic (production of HCl and pepsin), endocrine (gastrin, somatostatin, pancreatic polypeptide) and surface mucus, the latter of which produce sialomucins, sulfomucins and neutral mucosubstances (Smith, 1989). The intestinal epithelium is basic in form. With some exceptions, for example the presence of ciliated cells in certain fish larvae, the principal cell types are absorptive and mucoid (Pedini et al., 2002; McLean et al., 2002b; Figure 1). The presence of enteroendocrine and pear-shaped cells, and the attendance of plasma, and other immunocytes within the gut’s architecture, have also been noted (Smith, 1989). Like the mammalian gut (Daugherty and Mrsny, 1999), the fish intestine harbors the body’s greatest number of lymphoid cells and produces more antibodies than any other organ. Teleosts, as opposed to elasmobranchs, do not appear to possess discrete aggregations of gut-associated lymphoid tissue (GALT) corresponding to the bursa of Fabricius of birds or Peyer’s patches of higher vertebrates.

Rather, teleost GALT is generally described as diffuse although concentrations of lymphoid cells have been observed in the midgut of certain species.

Teleosts thus represent excellent models for examining various aspects of GI physiology, and drug absorption phenomena in particular. The fish gut presents the identical barriers to drug absorption as seen in higher vertebrates as well as providing certain advantages unmatched by mammalian or avian models. The lack of a functional stomach in certain species bestows the formulation or pharmaceutical chemist with a model for dissecting out the effects of gastric digestion, a condition that can only be challenged by the gastric brooding frog.

The lack of discrete areas of GALT may also be considered an advantage in designing delivery systems that are non-GALT dependent. Thirdly, since fish are poikilothermic, surviving temperatures between -2 and 44 oC, the impact of temperature upon absorptive phenomena may be readily examined. These apparent advantages however, also represent complications to the development of oral drug delivery systems for use in aquaculture, since the wide variety of GI architectures and physiologies may impede the development of universal formulations.




Figure 1.A) The gastrointestinal tract of teleosts varies considerably in form, being relatively short in carnivores, such as flounder, or extremely long, as seen in phytoplanktivores such as tilapia. Some 18% of all known teleosts do not possess a functional stomach(s). Instead, these species exhibit an intestinal bulb or swelling which serves to control the entry of ingested material into the intestine proper. Especially in carnivores, the intestine may express extensions of the gut (c), which are termed pyloric cecae. Pyloric cecae may vary in number and size and increase the surface area available for absorption. The intestine proper has often been sub-divided into fore-, mid- and hind-gut, distal, rostral or caudal intestine or anterior (a), median and posterior (p) segments. B) The surface area of the fish intestine is increased by a series of intestinal folds, which vary in height and complexity depending upon ontogeny and species. The complexity and height of folds sometimes decreases in the posterior section of the gut. Gut lumen = l; fold = f. C) Intestinal folds incorporate various cell types, the population of which changes with age. For example, at weaning, the numbers of goblet cells (g) increases rapidly, as too do absorptive enterocytes (a). Below the epithelial layer lie the basal lamina (l), strata compactum (sc) and granulosum (sg) and muscularis circularis and longditudinalis (ml). D) Absorptive enterocytes, which are blanketed by a mucus coat and unstirred water layer (m), possess, at their apical surface, a complex comb-like border comprising microvilli (mv), below which lies the terminal web (t). The terminal web area may exhibit a large number of pinocytotic invaginations and vesicles (pv), which, upon migrating towards the basal membrane of the cell may coalesce forming larger vacuoles that may ultimately form a supranuclear vacuole (s). Incoming vesicles may unite with lysosomal compartments of the cell forming primary lysosomes (l1). Unification of primary lysosomes creates secondary (l2) or tertiary membrane-bound cellular components. The patency of the epithelial layer is maintained by socalled tight junctions or the junctional complex (tj), which fuses adjoining cells together. Lamellar or tubular infoldings (l), are believed to provide an exit line for the pinocytosed contents of vacuoles and vesicles into the intracellular space, where such potentially antigenic material may be sampled by intraepithelial lymphocytes (il). The precise location of the absorptive enterocytes nucleus (n) may vary (see Figure 4).


The fish gut as a barrier

The intestinal epithelium represents the largest frontier between external environment and internal milieu. The major role of the GI interface is to digest and selectively absorb essential nutrients, vitamins, minerals, and co-factors, while at the same time limiting the entry of pathogens, toxins, and undigested macromolecules into the internal milieu.

This complex multi-functional role has resulted in the evolution of a variety of mechanisms that act either alone or in concert to minimize the absorption of non-nutrient material and undigested macromolecules.

The fish gut, like that of higher vertebrates, presents, in principal, three barriers to the absorption of intact peptides, proteins and other macromolecules: physical, (bio)chemical and immunological (McLean and Donaldson, 1990).

Even given this challenging environment however, incontrovertible evidence has accumulated to support the opinion that the barrier function of the teleost GI tract is incomplete. Conceptually, two routes are available for the apical to basolateral transfer of macromolecules across intestinal epithelia viz. between adjacent cells (paracellular or intercellular uptake or persorption), or through cells (transcellular uptake). Specific material may not be limited to either pathway, but most molecules are preferentially transported via the transcellular or paracellular route due to their chemical and physical properties (Pade and Stavchansky, 1997). In fish, evidence to support the existence of a paracellular route for macromolecular absorption is sparse (see below).

However, transcellular transport mechanisms have been well characterized and it is now clear that molecules in excess of a million Daltons can transfer from the gut lumen and gain access to the circulation (review: McLean et al., 1999; 2002a). Despite the apparently natural permeability of the gut, oral availability of most peptide, protein, and similar therapeutic compounds is generally considered inadequate as a means of manipulating production-related processes in aquaculture. Unlike the clinical situation, a much narrower range of bioactives are of interest to aquaculture production so the development of efficacious oral formulations may be less demanding.


Application of oral delivery systems to aquaculture

In the immediate term, three important productionrelated metabolic processes offer potential for manipulation in aquaculture through the oral delivery of various therapeutic agents: reproduction, growth, and health. Other areas that may draw attention in the future include enhancing end product quality, including flesh pigmentation, increasing the uptake of antimicrobials and poorly absorbed nutrients and protecting valuable feed additives from digestion.

The ability to induce ovulation and spermiation in aquacultured organisms is of practical importance because many species fail to spawn in captivity, or captive males and females express asynchronous maturation. Often, spawning does not take place even though gonads express normal growth and development, which indicates that spawning failure is due to the lack, or inhibition, of those factors which induce final maturation, ovulation or spermiation.

Furthermore, the ability to induce spawning outside the normal season for some species permits the production of hybrids that may express superior production characteristics. As well, the control of spawning provides the industry with the means to lengthen grow-out periods. Total control over reproduction has added benefits in that it also maximizes larval production due to enhanced fertilization and egg survival.

A number of candidate and high-value aquaculture organisms express slow growth, which represents a severe handicap to bringing such species into industrial production. Recombinant DNA technologies however provide the means to produce an almost inexhaustible supply of growth regulating peptides such as growth hormone releasing factor (GRF) and growth hormone (GH). Growth hormone treatment of fish is known to stimulate appetite, enhance growth rates, benefit feed conversion efficiencies, modify composition to favor leanness while also stimulating the immune system and salt water transfer (smoltification), the latter of which is so important to the production of salmonids (McLean and Devlin, 2000). Moreover, by its action, GH reduces waste output during cultivation and thus its application represents a method of reducing the environmental impact of intensive operations.

Finally, and no less important are issues relating to the health and welfare of cultured fish. Disease represents a major impediment to the expansion of aquaculture in various regions of the world and continues to hamper even well established operations. The advent of commercially available vaccines has substantially reduced the industry’s reliance upon antimicrobials. Vaccination has become of increasing importance because many important pathogens now express resistance to traditional treatments. In attempts to enhance disease resistance in cultured animals, the aquaculture industry has recently commenced the use of diets fortified with micronutrients and immunostimulants.

The major technical constraint to applying reproductive, growth factor, and health technologies to aquaculture however, is the lack of effective and economic methods of administration. For many commercially important species, handling stress can result in failure to reproduce and, in severe circumstances, may cause death of highly valuable broodstock. The ability to control the reproductive event through the oral administration of appropriate peptides (e.g., LH-RH) and proteins (e.g., human chorionic gonadotropin; HCG), thereby alleviating handling and visual stress, would offer clear advantages to industry.

Oral vaccination offers many potential rewards in that it would reduce labor costs, save time, decrease possibilities for crosscontamination with needles, enhance employee safety, eliminate handling stress of treated animals and abolish the need to dispose of treatment waters and needles. New technologies have produced several synthetic peptide-based viral vaccines for clinical trials, and similar expertise could transfer to the aquaculture industry. Furthermore, third generation antimicrobials, as illustrated by the peptide antibiotics and a wide range of immunostimulants, have been shown to enhance disease resistance in larval fish even before they have an ability to produce specific antibodies. Similar effects are seen in juvenile and adult fish and shellfish. Of particular interest are a number of complex proteins that enhance non-specific and humoral immune defense mechanisms. Many other candidate macromolecules offer similar immunoadvantages but their use would increase if oral formulations were to become available.

Intensive aquaculture operations may maintain hundreds of thousands of individuals, such that in general, the only relevant method for delivering production-related, bioactive compounds is the oral route. Unlike vaccines and ovulation-inducing drugs however, growth factor technologies present a different set of circumstances when oral formulations are considered. In general, vaccines and reproductive peptides and proteins are only given to fish on two occasions at specific times during a production cycle. In contrast, growth factors must be delivered on a sustained basis throughout growout if significant advantages are to be accrued.

Different delivery strategies may be contemplated therefore, for the various treatment regimes needed.
Irrespective of the potential advantages that oral formulations may offer aquaculture, if this route of delivery is to be gainfully employed, formulations must be devised to surmount the various barrier functions of the GI tract. A number of tactics have been examined in attempts to protect therapeutics from GI secretions or to enhance their absorption by the gut. In general, most of the strategies inspected in designing oral delivery systems for fish have their roots in clinical drug delivery research, although the unique nature of aquatic organisms has also resulted in the development of novel technologies. To date, research has considered methods of modifying or avoiding physical and chemical barriers to absorption.


AVOIDING PHYSICAL BARRIERS

The fish gut presents both mucus and epithelial barriers to the absorption of orally-presented macromolecules, both of which may be considered as physical obstructions. Gastrointestinal mucus is a heavy, gelatinous, continually shifting matrix that lines the entirety of the tract (Sood and Panchagnula, 2001). Predominantly composed of water, with approximately 5% comprising glycoproteins, enzymes, electrolytes, lymphocytes, bacteria, and sloughed epithelial cells and, in disease states, DNA (MacAdam, 1993), mucus protects the underlying epithelia from pathogens, autodigestion, abrasives, and potential noxious chemicals, while maintaining differential pH between gut lumen and cell surface (Norris et al., 1998). Mucus is continually secreted, thereby countering losses due to enzymatic degradation, physical erosion, and ejection with fecal material. Steady-state secretion is maintained by exocytosis of mucus from goblet cells (Figure 1), the mechanism of which may be controlled by neurohormonal transmitters. The viscoelastic nature of mucus facilitates the passage of food along the gut by allowing food and fecal boli to slide over the underlying epithelia without causing damage (MacAdam, 1993). The mucus lining is thus a significant barrier to the oral delivery of macromolecules and several studies have illustrated that mucus impedes the transfer of proteins of >5 kDa from the gut lumen to epithelial surface. Gut mucus also acts to agglomerate microparticulates, resulting in increased size and reduced diffusion coefficients and absorption potential. The diffusion of microparticulates through mucin follows the Stokes-Einstein equation and size, hydrophobicity and molecule charge each play a role in determining diffusibility. In mammals, the thickness of the mucin gel layer varies regionally and similar observations have been made with chinook salmon (Schep et al., 1998). However, it is notoriously difficult to measure the rate of mucus secretion since it adheres strongly to the underlying mucosa and its physical removal may cause the release of intracellular mucus. The regional differentiation in mucin production obviously has consequences in the targeting and design of specific oral formulations for use with cultured fish and higher vertebrates alike.

The mucus barrier to drug delivery can be substantially reduced by so-called mucolytic agents.

These compounds may be separated into three types based upon their presumed mechanism of action viz. proteases that act upon the glycoprotein component of mucus, sulphydryl complexes, which act on disulfide links, and detergents that break the non-covalent bonds in mucus. However, while several compounds have been examined for efficacy as mucolytic agents for increasing the absorption of peptide drugs in mammals, relatively few studies have been undertaken with fish.

Nevertheless, Schep et al. (1997) examined the effect of dithiothreitol (DTT) and sodium deoxycholate as mucolytic agents in chinook salmon and reported that either alone or in combination, each reduced mucus secretion in the proximal and distal intestine of chinook salmon. DTT is known to reduce the viscosity of intestinal mucus and thereby diminish the barrier to diffusion and uptake of macromolecules. Suzuki et al. (1988) used the mucoadhesive polymer polyacrylic acid as a means of increasing the absorption of salmon pituitary extract to goldfish. Treatment resulted in significant changes in circulating gonadotropin and, interestingly, with increasing dose of polyacrylic acid, an extended time course for the appearance of the protein was observed. This study illustrates the potential for employing mucoadhesives as a means of enhancing the gut residency time of specific proteins and suggests that targeted delivery to the specific area of GI absorption may provide significant benefit in terms of reducing the amount of bioactive protein delivered. Certain surfactants also appear to principally act upon the mucus coat.

Thus, in studies that examined the uptake of HRP in rainbow trout, co-administration of a synthetic detergent Mega-9 (Figure 2) caused gelatinization of the mucus lining of the gut (McLean and Ash, 1990). The authors speculated that this gelling reduced the physical barrier presented by mucus, resulting in enhanced enterocyte-protein interactions and absorption of the reporter protein. The detergent properties of lysophosphatidylcholine have also been employed as a means of enhancing the absorption of peptides in fish. Thus, Solar et al. (1990) coadministered the ovulation-inducing decapeptide LHRH with lysophosphatidylcholine and sodium bicarbonate to sablefish, which resulted in successful induction of spawning. However, lysophosphatidylcholine is known to severely damage a variety of epithelia following exposure, often causing irreparable damage (Martin et al., 1996). This compound would have only restrictive use, perhaps where treatments were given once or twice, for example, during the induction of spawning.

Lying below the protective mucus coat of the intestine are various surface irregularities that in unison augment the potential surface area for the absorption of nutrients and other lumenal content (Figure 1). Intestinal folds may increase the gut’s surface area over 30-times, whereas the microvilli blanket of absorptive enterocytes further amplifies the area for absorption and digestion by around 600- fold. Integral membrane components of the so-called brush border include amino-oligopeptidases and endopeptidases that act upon peptides of ten residues (decapeptides) and below.

The apical surfaces of the cells of the intestinal epithelia are sealed by tight junctions (Figure 1) that act to regulate the passage of ions and macromolecules. Tight junctions are complex, dynamic structures. Their configuration requires the participation of a variety of proteins, two key elements of which are ZO-1 and occludin. Occludin is believed to be a transmembrane protein strand that completes the seal between cells. Other structural components of the tight junction and their regulation still require complete characterization (Fasano, 1998). A number of studies have examined the possibility of artificially loosening the junctional complex as a means of increasing paracellular absorption of peptides, proteins, and vaccines from the gut lumen. Often, however, the compounds employed have expressed unacceptable side effects.

Many cationic surfactant calcium chelators, which deplete Ca2+, relax the integrity of tight junctions, induce disruptions of actin filaments and decrease cell adhesion (Citi, 1992), while severely compromising barrier functions of the intestinal epithelium due to cell lysis and exfoliation (Hochman and Artursson, 1994). Thus, sodium caprate, a wellestablished epithelial tight junction opener (Lindmark et al., 1998), at high concentrations may cause epithelial damage which limits its potential as an oral formulation component for fish, since control over delivery becomes problematic. A more elegant method for increasing tight junction permeability is the use of zona occludens toxin (ZOT), derived from Vibrio cholerae. ZOT is believed to interfere with the polymerization process that leads to the formation of occludin strands and, through this intervention, increases intercellular permeability. Administration of ZOT (1.1 x 10^-10 M) with insulin to rats increased absorption by up to 72% (Fasano, 1998). While no studies have evaluated ZOT as a permeability enhancer for inclusion in oral formulations for fish, it is likely that it would find application only with small peptides, (<12000 D), such as presented by LH-RH and its analogues.




Figure 2. Different methods have been examined in attempts to enhance the absorption of orally administered peptides and proteins to fish. Strategies include the use of enzyme inhibitors and the use of detergents to reduce the barrier functions of gut mucus. Co-administration of the artificial detergent Mega-9 (M-9; 5% w/v) or soybean trypsin inhibitor (SBTI; 5% w/v) with the reporter protein horseradish peroxidase (HRP; 50 μg/g BW) to rainbow trout resulted in significant increases (*P<0.05) in HRP tissue accumulation 45 min. post oral intubation. An apparent synergistic effect upon HRP absorption was observed when the reporter protein was delivered with a mixture of M-9 and SBTI (1% w/v) (from McLean and Ash, 1990).


A wide range of epithelial enhancers have been examined for their efficacy in increasing the uptake of bioactives in mammalian models and these include a number of non-ionic, cationic, anionic, and zwitterionic surfactants, various phospholipids, etc. Their mechanisms of action and experimental applications have been extensively reviewed (e.g., Swenson and Curatolo, 1992; LeCluyse and Sutton, 1997; Sood and Panchagnula, 2001). Hyperabsorption of hGH by the gut of carp has been attained following its co-administration with the bile salt, sodium deoxycholate (SDA; Hertz et al., 1991). However, studies that examined the degree of epithelial damage by evaluating the release of cytosolic lactate dehydrogenase (Uchiama et al., 1999) indicate that the concentration of SDA employed (50 mM) during the carp studies would likely cause significant, perhaps even irreversible damage, to the intestinal mucosa. Nevertheless, as with many initial studies, the findings of Hertz and colleagues have provoked further examination of SDA as a potential candidate as an absorption enhancer for application in fish.

Thus, Schep et al. (1997) explored the use of SDA in assessing the permeability of the chinook salmon intestine to mannitol. These authors reported a significant increase in gut permeability when employing SDA concentrations of 5 mM, with no apparent damage to the epithelia following single treatments. A variety of Tweens have also been studied for their enhancing effects for orally delivered GnRH, with Tween 20 and 80 providing significant increases in uptake in the agastric carp (Mikolajczyk et al., 2002). Clearly, greater ranges of surfactants are deserved of further investigation in fish models as putative absorption enhancers.


EVADING BIOCHEMICAL BARRIERS

The low bioavailability of orally administered peptide and protein drugs, vaccines, and other macromolecules, is generally accredited to their enzymatic degradation. The molecular dissection of such substances occurs in the gut lumen, at the brush border, and by membrane-bound and cytosolic proteases and peptidases stationed along the length of the intestine. In teleosts, identified lumenal proteases include, but are not limited to, pepsin, trypsin, chymotrypsin, elastase, carboxypeptidase A and B, collagenase, and carboxylesterase. Several studies have demonstrated membrane-bound enzymes in fish including leucine aminopeptidase, acid and alkaline phosphatase and γ-glutamyl transferase, although their activities are generally much lower than recorded for higher vertebrates. Further, regional distributions of membrane-bound enzymes have been reported both along the gut and on the sides of intestinal folds. Comparatively few studies have examined cytosolic enzymes in fish although the incidence of various di-, and tripeptidases has been recorded for several species.

Clearly, this battery of enzymes represents a major impediment to oral delivery systems that must be defeated if success is to be achieved. An obvious strategy to protect the stability of administered therapeutics is the use of enzyme inhibitors or compounds that profitably modify (decrease) the activity of specific enzymes. A broad range of inhibitors of gastric, pancreatic (lumenal) and brush border membrane-bound enzymes are commercially available and some of these have been successfully employed to enhance the absorption of orally delivered peptides and proteins to fish.

For example, in attempts to eliminate peptic digestion, studies have examined the utility of co-delivering the reporter protein HRP with pepstatin as a means of increasing the absorption (bioavailability) of HRP to rainbow trout (Figure 3). This strategy resulted in enhanced levels of the indicator protein in the liver of treated fish. An alternative to comparatively expensive enzyme inhibitors, as a means of protecting orally administered compounds from gastric digestion, is the use of antacids. Antacids reduce not only the protein hydrolyzing effect of HCl but also the activation of pepsinogen. Moreover, elevated pH weakens the activity of pepsin.

One study has employed antacids in attempts to protect recombinant bovine (rb)GH from gastric digestion in coho salmon (McLean et al., 1990). Co-delivery of 12.5 μg rbGH/g BW/week for 6 weeks with 1% w/v sodium bicarbonate resulted in accelerated growth when compared to fish administered rbGH alone and controls. Similar results have also been achieved using aluminum and magnesium hydroxides and calcium carbonate.

Gastric secretions also restrict the transfer of oral vaccines, although the early success of Duff (1942) in administering vaccines in feeds has met with variable results since. These disparities in response likely occur due to differences in gastric evacuation rates and intestinal transit times, as well as the form in which the antigenic component is presented. The prevailing temperature, vaccine dose, variability in challenge type (cohabitation, injection, etc.) and challenge organism virulence, together with past nutritional and immune status of test animals, also influence the success or failure of oral vaccination studies. The restrictive nature of the gastric barrier to the oral delivery of vaccines is best illustrated by comparing the immune response of fish following oral and rectal administrations of vaccine. For example, Johnston and Amend (1983) reported inducing immunity to Vibrio anguillarum in sockeye salmon following rectal, but limited response after oral, bacterin administrations.

Similar studies to those of Johnston and Amend have since been executed with a variety of fish and pathogens and it is evident that differences exist in terms of bacterin uptake and responsiveness to vaccination between species (e.g., Vigneulle and Baudin-Laurencin, 1991). The use of enteric coated microspheres and vaccines, encapsulated antigens and the bioencapsulation of vaccines in live feeds and biofilms have all been evaluated as means to protect antigenic components from the degrading action of the stomach (Lavelle et al., 1997; Azad et al., 1999).

Protection of vaccine from gastric digestion enhances immune response and survival to subsequent exposure to pathogens. This was eloquently demonstrated by Anders (1978), who codelivered a Vibrio bacterin and antacid to European eel and trout. In these experiments, a 10-fold increase in protection to challenge was recorded for antacid-protected vaccine when compared to fish orally treated with vaccine alone, which, the author concluded, resulted due to the inhibition of pepsin and a reduced impact of the acidic environment. Of particular note have been studies that employed a novel proprietary oral delivery platform (Oralject) suitable for use with a variety of vaccine preparations and production-related proteins and peptides. This system, which includes enzyme inhibitors, an antacid and an uptakeincreasing agent (Vandenberg, 2001), has been successfully employed to vaccinate salmonids against Aeromonas salmonicida, the causative agent of furunculosis (Vandenberg et al., 2002, 2003).




Figure 3. Strategies of enhancing the gastrointestinal uptake of marker proteins in fish have included an examination of inhibitors of pepsin (pepstatin; 0.05% w/v), trypsin (leupeptin; 0.005% w/v) and chymotrypsin (aprotinin; 0.05% w/v). Co-administration of these anti-enzymes with HRP (50 μg/g BW) to rainbow trout resulted in significant increases (*P<0.05) in HRP tissue accumulations 45 min. post oral intubation with pepstatin and aprotinin, but leupeptin was without effect.


Numerous naturally occurring inhibitors to lumenal proteases are commercially available. Many of these products are derived from the pulses, the most notable of which is soybean trypsin inhibitor (SBTI).

Inhibition of tryptic and catheptic activity can also be achieved using leupeptin. Although the first record of absorptive modification of antigenic material by the teleost gut used SBTI (Figure 2), surprisingly few studies have examined the benefits of co-administering SBTI with bioactive molecules any further. This scant interest may be because, when administered for extended duration, enzyme inhibitors interfere with the normal processes of protein digestion, sometimes resulting in decreased growth performance. Nevertheless, Breton and colleagues (1995; 1998; Ollevier et al., 1997) reported that the forced delivery of D-Arg6 Pro9 salmon GnRH to African catfish, common carp and rainbow trout, when combined with various enhancing agents, including chicken egg-white trypsin inhibitor, resulted in significant increases in circulating levels of maturational gonadotropin (GtH II). However, since the study did not include a trypsin inhibitor-GnRH group, it remains difficult to dissect out whether observed improvements in endocrine response were due to addition of trypsin inhibitor or otherwise.

In later studies however (Lescroart et al., 2000; Mikolajczyk et al., 2002), the positive benefits of adding chicken egg white trypsin inhibitor to GnRH absorption to trout and carp respectively were more clear-cut. The impact of coadministering leupeptin and aprotinin, both potent inhibitors of tryptic and catheptic activities, upon gut absorptive processes in fish are illustrated in Figure 3. Co-delivery of leupeptin with the tracer HRP to rainbow trout was without effect upon tissue or plasma presence. This may have resulted due to the relatively low dose employed. In stark contrast, delivery of HRP with aprotinin significantly enhanced hepatic and renal levels of the reporter protein (Figure 3).

Other technologies have also been considered as a means of enhancing the absorption of orally administered drugs through defeat of the enzymatic barrier. These include the application of supplementary agents, such as bioadhesives (Sakuma et al., 2001; Ahmed et al., 2002) codelivered with enzyme inhibitors and/or adjustment of intestinal pH to abolish lumenal protease action.

It has been reported that a variety of mucoadhesive polymers improves the absorption of poorly available bioactives, especially when delivered as microparticulate formulations. While the mode of action of these compounds remains to be fully elucidated, it has been suggested that they act by binding to the intestine’s mucus layer (Chen and Langer, 1998). In doing so, muco-, or bioadhesives increase intestinal residency of pharmaceuticals, resulting in increased absorption efficiency. No reports exist in the public domain to describe specific experiments with this technology as a means of enhancing the absorption of bioactives or vaccines in aquacultured fish although compounds employed for other reasons (epithelial penetration) may express bioadhesive characteristics. Similarly, the use of citric acid as a means of inducing a transient and regionalized decrease in intestinal pH, thereby lowering enzymatic activities and enhancing drug absorption have yet to be examined in fish models.

No reports upon methods of inhibiting brush border or cellular enzymes have been researched although the use of monensin as a putative inhibitor of the transfer of internalized proteins between pinocytotic vesicles and lysosomes (Schlegel et al., 1981) has been examined. When HRP was administered with various concentrations of monensin however, no enhanced uptake was observed.

Another method of avoiding gastric, lumenal, and potentially cellular hydrolysis is to employ peptides and proteins that either express natural resistance to degradation or have been chemically modified to enhance survival in the gut and/or increase bioactivity once absorbed. One study has examined this aspect in fish. Thus, McLean et al. (1991) reported that the superactive analogue of LHRH (des-Gly10 {D-Ala6} LHRH ethylamide), when orally delivered at 20 μg LHRHa/g body weight of 17ß-oestradiol-primed coho was absorbed at 100 times greater levels than that of native LHRH.

Moreover, both the analogue and native forms of LHRH induced a GtH response in treated fish. The authors speculated that the increased absorption of the analogue resulted due to its ability to resist acid and enzymatic hydrolysis. Although not tested in oral formulations with fish, an analogue form of rbGH, when injected into coho salmon, returned approximately nine times the growth-promoting potency of the native form of bGH (Down et al., 1989). The enhanced bioactivity of such proteins may provide a means of reducing the amount of substance delivered by oral methodologies or express greater resistance to hydrolysis in the gut lumen. These possibilities, however, remain to be further explored.


THE IMMUNOLOGICAL BARRIER

Some authors speculate that the pinocytotically active absorptive cells of the fish gut perform a function similar to that of the more highly differentiated mammalian M-cell. In the mammalian intestine, M-cells overlie lymphoid aggregations (Peyer’s patches) (review Neutra et al., 1996).

These cells are specialized in sampling and presenting incoming antigenic materials to Tlymphocytes which may ultimately stimulate B-cell differentiation into plasma cells that are able to produce secretory immunoglobulin. Secretory immunoglobulins prevent microbial attachment and colonization of the epithelial surface and bind specific macromolecules, thereby hindering passage to the internal milieu.

Several studies with fish have determined that following oral vaccination, the number of intraepithelial lymphoid cells associated with the second gut segment increases (e.g., Davina et al., 1982). Moreover, histological investigations have illustrated the transfer of incoming intact proteins, including GH, into macrophages and granule cells located within the basal lamina (e.g., Rombout and van den Berg, 1989; Le Bail et al., 1989). Clearly, increased research effort is warranted to further characterize the role of immunocytes and the mucosal immune system in inhibiting the delivery of orally administered bioactives to aquacultured organisms. While development of methods that avoid the action of the immune barrier is improbable, and perhaps even undesirable, a more thorough understanding of its role may pave the way to enhanced oral platform designs. Delivery of therapeutics in microparticulate form may represent one such strategy.


ABSORPTION OF MICROPARTICULATES

It has been estimated that humans ingest 1012 ultrafine particles per day (Powell et al., 2000). It is likely that cultured fish are also exposed to a similar loading (108 particles per day for a platesized animal). It is not too surprising to find therefore, that the vertebrate gut absorbs microparticulates. Indeed, this ability has been established for well over a century (Herbst, 1844).

Nevertheless, it has only been over the last decade that interest in this phenomenon, particularly as a means of delivering bioactive compounds, has increased (reviews: Yeh et al., 1998; Delie, 1998; Hussain et al., 2001; Florence and Hussain, 2001).

Three potential routes for microparticulate uptake by the gut are available. These include transcellular passage across absorptive enterocytes, paracellular routes between enterocytes and, in higher vertebrates, via specialized (M or microfold) cells of gut-associated lymphoid tissue. The latter is unavailable to fish although evidence for the paracellular transport of microparticulates has been presented. Thus, the ability of the rainbow trout gut to absorb microparticulate material was examined following intubation (3.5 g/kg) of yeast (appr. 5.6 μm diameter), commercial grade cornstarch (appr. 21 μm diameter), or potato starch (appr. 43 μm diameter). Both yeast and cornstarch particles (8-14 μm), were observed to pass from gut lumen to the lamina propria via a paracellular or persorptive route only.

No evidence for the like passage of potato starch was found. (McLean et al., 2000; Figure 4). The authors concluded that microparticulate transport was size limited in fish and correlated with particle shape (angular versus smoothed). Clearly, the potential exists to employ persorption for the delivery of micro-, or bio-encapsulated vaccines, growth promotants, reproductive peptides and similar materials to fish or potentially as a means of providing supplemental nutrition for larval animals.

Indeed, Tsai et al. (1994) reported that striped mullet fed with yeast recombinant for a fish growth hormone, yielded significant growth acceleration over a 6 week trial period when compared to control groups. Although the latter study did not examine the route of uptake of the yeast, the persorptive pathway may have been the preferential route of absorption, providing the GH with a protective barrier against all comers. This area of research remains fertile ground for study. Specifically there is a need to establish whether the phenomenon expresses a dose-dependency and if the mechanism imparts a viable option for controlled oral delivery of bioactives to aquatic organisms.


Perspective

Even though a considerable number of methods have been evaluated in attempts to increase permeability of the fish gut to orally administered productionrelated drugs, this discipline remains in its infancy when compared to mammalian research. Many techniques of enhancing the uptake of orally delivered bioactives remain untested in fish, including the use of prodrugs and chimeras that exploit receptor-mediated endocytosis.

Nevertheless, significant strides have been taken towards commercializing oral dosage forms for fish and a number of companies have recently brought their technologies to the marketplace. Clearly, oral drug delivery systems for aquaculture will continue to evolve to service the increased desire for such formulations, especially as these relate to vaccination. Equally lucid is that a greater understanding of teleost GI physiology will lead to the development of more efficient delivery platforms.

Increased intelligence concerning the impact of gut transit and residency times upon drug dissolution and uptake and the impact of rearing conditions (temperature, salinity, feeding practices, etc.) upon absorption efficiency will undoubtedly lead to a greater level of sophistication in the design of carrier systems. Augmented knowledge relating to the histological development of the fish gut, its lumenal and brush border enzyme systems, and the control of mucus secretion will no doubt provide additional insights for formulation chemists to more precisely develop age-specific delivery systems. Contemporary research upon optimizing the concentrations of cocktail-based approaches to drug delivery is poor and this demands immediate redress if economic therapies are to become a solid reality.




Figure 4.Photomicrographs of cornstarch stationed between absorptive enterocytes of the anterior hind gut of rainbow trout 18 hrs post-lumenal introduction. The starch granules, indicated by dotted arrows, were approximately 13 x13 μm (Figure 4 A, B). Solid arrows mark the point of entry of the starch granules. Figure 4 C illustrates the migration route of a persorbed cornstarch granule (two-dimensional dashed arrow). Granule dimensions were 7x7 μm. The presumed entry point of the granule is marked by the solid arrow. The epithelial layer of the gut was apparently separated from the lamina propria during the
passage of cornstarch from the gut lumen. H&E x400. (From McLean et al., 2000; with permission).


References

Ahmed, A., C. Bonner and T.A. Desai. 2002. Bioadhesive microdevices with multiple reservoirs: a new platform for oral drug delivery. J. Cont. Release 81:291-306.

Anders, E. 1978. Versuche zur oralen Immunisierung Regenbogenforellen und Aalen gegen Vibrio anguillarum. Fisch. Forsch. Rostock 16:129-134.

Azad, I.S., K.M. Shankar, C.V. Mohan and B. Kalita. 1999. Biofilm vaccine of Aeromonas hydrophila – standardization of dose and duration of oral vaccination of carps. Fish Shellf. Immunol. 9:519-528.

Breton, B., I. Roelants, T. Mikolajczyk, P. Elper and F. Ollevier. 1995. Induced spawning in teleost fish after oral administration of GnRH-A. In: Fifth International Symposium on the Reproductive Physiology of Fish (F.W. Goetz and P. Thomas, eds). University of Texas at Austin, pp. 102-104.

Breton, B., Y. Roelants, F. Ollevier, P. Epler and T. Mikolajczyk. 1998. Improved bioavailability of orally delivered peptides and polypeptides in teleost fish. J. Appl. Ichthyol. 14:251-257.

Chen, H. and L. Langer. 1998. Oral particulate delivery: status and future trends. Adv. Drug Deliv. Rev. 34:339-350.

Citi, S. 1992. Protein kinase inhibitors prevent junction dissociation induced by low extracellular calcium in MDCK epithelial cells. J. Cell Biol. 117:169-178.

Daugherty, A.L. and R.J. Mrsny. 1999. Transcellular uptake mechanisms of the intestinal epithelial barrier Part one. Pharm. Sci. Tech. Today 2:144-151.

Davina, J.H.M., H.K. Parmentier and L.P.M. Timmermans. 1982. Effect of oral administration of Vibrio bacteria on the intestine of cyprinid fish. Dev. Comp. Immunol. 2:157-166.

Delie, F. 1998. Evaluation of nano- and microparticulate uptake by the gastrointestinal tract. Adv. Drug Delivery Rev. 34:221-233.

Down, N.E., E.M. Donaldson, H.M. Dye, T.C. Boone, K.E. Langley and L.M. Souza. 1989. A potent analog of recombinant bovine somatotropin accelerates growth in juvenile coho salmon (Oncorhynchus kisutch). Can. J. Fish. Aquat. Sci. 46:178-183.

Duff, D.C.B. 1942. The oral immunization of trout against Bacterium salmonicida. J. Immunol. 44:87-94.

Fasano, A. 1998. Modulation of intestinal permeability: an innovative method of oral drug delivery for the treatment of inherited and acquired human diseases. Mol. Gen. Metab. 64:12-18.

Florence, A.T. and N. Hussain. 2001. Transcytosis of nanoparticle and dendrimer delivery systems: evolving vistas. Adv. Drug Delivery Rev. 50:S69- S89.

Harder, W. 1975. Anatomy of Fishes. E. Schweizerbart’sche Verlagsbuchhandlung, Stuttgart. Herbst, G. 1844. Das lymphgefäbsystem und seine verrichtung. Vandenhoeck and Ruprecht, Göttingen.

Hertz, Y., A. Tchelet, Z. Madar and A. Gertler. 1991. Absorption of bioactive human growth hormone after oral administration in the common carp (Cyprinus carpio) and its enhancement by deoxycholate. J. Comp. Physiol. 161B:159-163.

Hochman, J. and P. Artursson. 1994. Mechanisms of absorption enhancement and tight junction regulation. J. Cont. Release 29:253-267.

Hussain, N., V. Jaitley and A.T. Florence. 2001. recent advances in the understanding of uptake of microparticulates across the gastrointestinal lymphatics. Adv. Drug Delivery Rev. 50:107-142.

Johnston, K.A. and D.F. Amend. 1983. Efficacy of Vibrio anguillarum and Yersinia ruckeri bacterins applied by oral and anal intubation of salmonids. J. Fish Dis. 6:473-476.

Lavelle, E.C., P.G. Jenkins and J.E. Harris. 1997. Oral immunization of rainbow trout with antigen microencapsulated in poly(DL-lactide-coglycolide) microparticles. Vaccine 15:1070-1078.

Le Bail, P.Y., M.F. Sire and J-M. Vernier. 1989. Intestinal transfer of growth hormone into the circulatory system of rainbow trout, Salmo gairdneri. Interference by granule cells. J. Exp. Zool. 251:101–107.

LeCluyse, E.L. and S.C. Sutton. 1997. In vitro models for selection of development candidates. Permeability studies to define mechanisms of absorption enhancement. Adv. Drug Delivery Rev. 23:163-183.

Lescroart, O., T. Mikolajczyk, W.J. Enright, F. Ollevier and B. Breton. 2000. Oral administration of a GnRH analogue (GnRHa) using a microgranulate delivery form: effect of the dose of GnRHa and the inclusion of trypsin inhibitor (TI) on plasma GTH-II concentration and ovulation in rainbow trout. Fourth International Symposium on Fish Endocrinology, Seattle, WA. W-1347.

Lindmark, T., N. Schipper, L. Lazorova, A.G. DeBoer and P. Artursson. 1998. Absorption enhancement in intestinal epithelial Caco-2 monolayers by sodium caprate: Assessment of molecular weight dependence and demonstration of transport routes. J. Drug Targeting 5:215-223.

MacAdam, A. 1993. The effect of gastrointestinal mucus on drug absorption. Adv. Drug Deliv. Rev. 11:201-220.

Martin, E., J.C. Verhoef, S.G. Romeijn, P. Zwart and F.W.H.M. Merkus. 1996. Acute histopathological effects of benzalkonium chloride and absorption enhancers on rat nasal epithelium in vivo. Int. J. Pharmaceut. 141:151-160.

Mayer, I. and E. McLean. 1995. Biotechnological and bioengineering strategies for reduced waste aquaculture. Water Sci. Technol. 31:85–102.

McLean, E. and R. Ash. 1990. Modified uptake of the soluble protein antigen, horseradish peroxidase HRP, following oral delivery to rainbow trout Oncorhynchus mykiss. Aquaculture 87:373–379.

McLean, E. and R.H. Devlin. 2000. Application of biotechnology to enhance growth of salmonids and other fish. In: Recent Advances in Marine Biotechnology (M. Fingerman and R. Nagabhushnam, eds). Science Publishers Incorporated, Enfield, New Hampshire, USA, pp. 17-55.

McLean, E. and E.M. Donaldson. 1990. The absorption of bioactive proteins by the fish gastrointestinal tract: a review. J. Aquat. Anim. Health 2:1–11.

McLean, E., E.M. Donaldson, H.M. Dye and L.M. Souza. 1990. Growth acceleration of coho salmon Oncorhynchus kisutch following oral administration of recombinant bovine somatotropin. Aquaculture 91:197–203.

McLean, E., D. Parker, C.M. Warby, N.M Sherwood and E.M.Donaldson. 1991. Gonadotropin release following oral administration of LHRH and its superactive analogue des Gly DAla LHRH ethylamide to 17ß-estradiol primed coho salmon. J. Fish Biol. 38:851–858.

McLean, E., B. Ronsholdt, C. Sten and Najamuddin. 1999. Gastrointestinal delivery of peptide and protein drugs to aquacultured teleosts. Aquaculture 177:231-247.

McLean, E., J.S. Goddard, M.R.G. Claereboudt, H.S. Al-Oufi and J-Y. Mevel. 2000. The teleost gut persorbs microparticles. Ribarstvo 59:47-56.

McLean, E., J.-Y. Mevel and R. Ash. 2002a. Intestinal uptake of macromolecules and microparticulates. In: Contemporary Issues in Marine Science and Fisheries. Hasanuddin University Press, Makassar, Indonesia, pp. 207- 242.

McLean, E., B. Vestbö and S.D. Dunn. 2002b. Transitions in intestinal cytology of normally growing and growth-retarded European eels (Anguilla anguilla, L., 1758) maintained in a commercial aquaculture recirculation facility. In: Proceedings of the 4th International Conference on Recirculating Aquaculture, Roanoke, VA, USA, July 18-21, 2002, pp. 143- 151.

Mikolajczyk., T., Y. Roelants, P. Epler, F. Ollevier, J Chyb and B. Breton. 2002. Modified absorption of sGNRH-a following rectal and oral delivery to common carp, Cyprinus carpio L. Aquaculture 203:375-388.

Neutra, M.R., E. Pringault and J.P. Kraehenbuhl. 1996. Antigen sampling across epithelial barriers and induction of mucosal immune responses. Ann. Rev. Immunol. 14:275-300.

Norris, D.A., N. Puri and P.J. Sinko. 1998. The effect of physical barriers and properties on the oral absorption of particulates. Adv. Drug Del. Rev. 34:135-154.

Ollevier, F., I.V.O. Roelants, B. Breton, P. Epler and T. Mikolajczyk. 1997. Oral delivery form having a high absorption efficiency and method for making the same. World Patent Office 9740702.

Pade, V. and S. Stavchansky. 1997. Estimation of the relative contribution of the transcellular and paracellular pathway to the transport of passively absorbed drugs in the Caco-2 cell culture model. Pharm. Res. 14:1210-1215.

Pedini, V., P. Scocco, A.M. Gargiulo, P. Ceccarelli and S. Lorvik. 2002. Glycoconjugate characterization in the intestine of Umbrina cirrosa by means of lectin histochemistry. J. Fish Biol. 61:1363-1372.

Powell, J.J., R.R. Harvey, P. Ashwood, R. Wolstencrift, M.E. Gershwin and R.P. Thompson. 2000. Immune potentiation of ultrafine dietary particles in normal subjects and patients with inflammatory bowel disease. J. Autoimmun. 14:99-105.

Rombout, J.H.W.M. and A.A. van den Berg. 1989. Immunological importance of the second gut segment of carp: 1. Uptake and processing of antigens by epithelial cells and macrophages. J. Fish Biol. 35:13–22.

Sakuma, S., M. Hayashi and M. Akashi. 2001. Design of nanoparticles composed of graft copolymers for oral peptide delivery. Adv. Drug Deliv. Rev. 47:21-37.

Schep, L.J., I.T. Tucker, G. Young and G. Butt. 1997. Regional permeability differences between the proximal and distal portions of the isolated salmonids posterior intestine. J. Comp. Physiol. 167B:370-377.

Schep, L.J., I.T. Tucker, G. Young, R. Ledger and G. Butt. 1998. Permeability of the salmon (Oncorhynchus tshawytscha) posterior intestine in vivo to two hydrophilic markers. J. Comp. Physiol. 168B:562-568.

Schlegel, R., M. Willingham and I. Pastan. 1981. Monensin blocks endocytosis of vesicular stomatitis virus. Biochem. Biophys. Res. Comm. 102:992-998.

Smith, L.S. 1989. Digestive functions in teleost fishes. In: Fish Nutrition (J.E. Halver, ed). Academic Press, San Diego, CA, pp. 331-421.

Sood, A. and R. Panchagnula. 2001. Peroral route: An opportunity for peptide and protein delivery. Chem. Rev. 101:3275-3303.

Stroband, H.W.J., F.H. van der Veen and L.P.M. Timmermans. 1979. Regional functional differentiation in the gut of the grass carp (Ctenopharyngodon idella Val.). Histochemistry 64:235-249.

Suzuki, Y., M. Kobayashi, K. Aida and I. Hanu. 1988. Transport of physiologically active salmon Gonadotropin into the circulation of goldfish, following oral administration of salmon pituitary extract. J. Comp. Physiol. B 157:753–758.

Swenson, E.S. and W.J. Curatolo. 1992. Means to enhance penetration. Adv. Drug Deliery Rev. 8:39- 92.

Tsai, H.-J., J-C. Kuo, S-W. Lou and T.T. Kuo. 1994. Growth enhancement of juvenile striped mullet by feeding recombinant yeasts containing fish growth hormone. Prog. Fish-Cult. 56:7-12.

Uchiama, T., T. Sugiyama, Y.S. Quan, A. Kotani, N. Okada, T. Fujita, S. Muranishi and A.

Yamamoto. 1999. Enhanced permeability of insulin across the rat intestinal membrane by various absorption enhancers: their intestinal mucosal toxicity and absorption-enhancing mechanism of n-lauryl-beta-D-maltopyranoside. J. Pharm. Pharmacol. 51:1241-1250.

Vandenberg, G.W. 2001. Composition for intestinal delivery. Patent number: CA 2396711

Vandenberg, G.W., C. Gaudreault, B. Chabot, G. Munger and D. Bernier. 2002. Oral immunization of brook trout (Savelinus fontinalis) against furunculosis using a novel oral delivery system: influence of application frequency and antigen dose. Proceedings of the 19th Annual Meeting of the Aquaculture Association of Canada September 17-20, 2002, Charlottetown, PEI.

Vandenberg, G.W., C. Gaudreault, B. Chabot, G. Munger and V. Dallaire. 2003. Oral immunization of brook trout (Savelinus fontinalis) against furunculosis using a novel oral delivery system. Proceedings from the Annual Meeting of the US Aquaculture Association. Aquaculture America 2003. February 18-21, 2003 Louisville, Kentucky.

Vigneulle, M. and F. Baudin Laurencin. 1991. Uptake of Vibrio anguillarum bacterin in the posterior intestine of rainbow trout Onchorhynchus mykiss, sea bass Dicentrarchus labrax and turbot Scophthalmus maximus after oral or anal intubtaion. Dis. Aquat. Org. 11:85- 92.

Yeh, P.-Y., H. Ellens and P.L. Smith. 1998. Physiological considerations in the design of particulate dosage forms for oral vaccine delivery. Adv. Drug Delivery Rev. 34:123-133.