Physiological considerations in the design of particulate dosage forms for oral vaccine delivery
Introduction
For chronic therapies, oral delivery continues to be the preferred route of administration. This preference results from enhanced patient compliance and access to the greatest number of markets. However, the gastrointestinal tract possesses a variety of morphological and physiological barriers which limit intestinal absorption. Morphological barriers include the mucus layer, villi and microvilli, the unstirred water layer and the epithelial cell layer which is composed of a number of cell types including villus and crypt cells, goblet cells and M cells of the Peyer's patch. Physiological factors include the range of luminal and microclimate pH, enzymatic activities found both in the lumen and within the enterocytes, specific transport mechanisms which limit absorption (e.g. P-glycoprotein), intestinal transit time and agents such as bile salts secreted into the intestinal lumen. There are a number of approaches which have been explored in an attempt to deliver therapeutic agents by the oral route 1, 2, 3, 4, 5. Despite the intense efforts which have been directed at identifying approaches for efficient oral delivery of peptides and proteins, to date these efforts have met with limited success.
Although the oral route is not amenable to the delivery of many proteins and peptides, the situation for vaccine delivery is quite different. The gut associated lymphoid tissue which is found throughout the mammalian intestine and has membranous (M) cells on its luminal surface provides a portal of entry for antigens. This review will focus on the physiological and morphological factors which regulate the absorption of particulates across the gastrointestinal tract and will compare different animal models for application to the study of oral vaccine delivery.
The major functions of the gastrointestinal tract are to provide a selective barrier between the environment and the systemic circulation and to digest and absorb nutrients, water and electrolytes. Morphological features which contribute to these functions include the structure of the gastrointestinal tract which is similar from the stomach to the anus. A single layer of columnar epithelial cells covers the luminal surface and is interspersed with a variety of specialized cells (e.g. goblet cells, endocrine cells, M cells). Coordination of absorption and digestion are accomplished through unique structural features found along the gastrointestinal tract including gastric glands, villi, microvilli and crypts. The epithelial cell layer covers a region of loose connective tissue, the lamina propria, containing blood and lymph vessels and a variety of cell types (e.g., lymphocytes and eosinophils). Transit along the gastrointestinal tract and mixing within intestinal segments is accomplished by contractions of separate groups of muscle layers found beneath the lamina propria, the muscularis mucosa, the circular and the longitudinal muscle layers. Within the gastrointestinal tract, digestion and absorption are influenced by both the intrinsic (submucosal or Meissner's plexus, myenteric or Auberach's plexus, Henle's plexus) and extrinsic (sympathetic and parasympathetic) neuronal elements. Neurotransmitters (e.g., acetylcholine and norepinephrine), biogenic amines and peptides (e.g., vasoactive intestinal peptide, calcitonin gene related peptide, substance P, 5-hydroxytryptamine, and neuropeptide Y) all act to regulate gastrointestinal function for optimal digestion and absorption.
Transport of macromolecules across the gastrointestinal tract can occur by a variety of pathways including both the paracellular and transcellular routes [1]. Epithelial cells comprise the majority of cells lining the gastrointestinal tract. The spaces between adjacent cells, the `tight junctions' account for less than 1% of the surface area of the intestine [6]. Transport between cells or paracellular transport occurs via aqueous channels. In man, the equivalent pore diameter has been estimated to be between 4–8 Å and in rat and rabbit, a pore diameter of ∼10–15 Å has been reported [6]. An alternate explanation for paracellular transport is movement through the voids produced at the villus tip during extrusion of mature enterocytes. However, this process appears to be too transient to account for the magnitude of paracellular flux observed in intact tissue [7]. Thus, passage through the junctional complexes appears to account for paracellular transport. A variety of approaches for enhancing absorption of peptides and proteins have focused on the junctional pathway. Recent evidence has shown that these junctional complexes can be altered by a variety of cellular regulator mechanisms 8, 9, 10that may be responsible for the increase in pore diameter (up to 10–30 Å) associated with an increase in transport of macromolecules with molecular weight ranging from the size of insulin [8]to dextran (molecular weight ranging from 10 to 2000 kD) [11]. Mullin and Snock [12]have found that tumor necrosis factor alpha (TNFα) may selectively alter tight junctional permeability based on the findings that it produces a decrease in transepithelial electrical resistance and increase in transport of the paracellular marker, mannitol, with no increase in short-circuit current (suggesting no transcellular transport effects). Enhancement of intestinal permeability 13, 14, 15has also been investigated with a variety of agents which disrupt membranes including medium chain glycerides [13]. Whether effects of these agents are confined to the paracellular pathway is poorly understood and will require further investigation to identify approaches with adequate safety. In a series of studies with polypeptides possessing positive, negative or no charge, Tzan and co-workers 16, 17demonstrated that polycations could induce increases in tight junctional permeability but negative or neutral polypeptides had no effect on paracellular permeability. Alternatively, Fasano and co-workers 8, 18have reported that tight junctional permeability can be modulated transiently by exposure to a cholera derived protein, Zonula occludens toxin (ZOT). However, as discussed previously, particles with diameters greater than 15–25 Å will not be able to traverse the paracellular pathway and therefore, physiological regulation of this pathway does not represent an approach for enhancing particulate passage across the intestinal epithelium [19].
Transport of proteins and peptides across the intestinal epithelium can occur by a number of different transcellular pathways, including passive transcellular transport, carrier-mediated transport and endocytosis/transcytosis [1]. For particulates, only transcytosis appears to be involved in passage across the intestinal epithelium.
Although there has been some controversy in the literature on the extent of particle absorption, generally, only a small fraction of the total particle dose appears in the systemic circulation 20, 21, 22. There is evidence in the literature showing that some particle translocation can occur via translocation across enterocytes in the villus part of the intestine 23, 24and through the paracellular pathway [25](Fig. 1). However, because of the low endocytic activity of enterocytes and the presence of tight junctions between them, the amount of particle translocation via these routes is usually very low. It is generally believed that the bulk of particle translocation occurs in the follicle associated epithelium (FAE) 21, 22, 27, 28, 29.
The FAE is a specialized epithelium covering mucosal lymphoid tissue [30]. This epithelium contains M cells that are specialized for endocytosis/transcytosis of antigens and microorganisms to the organized lymphoid tissue within the mucosa. It is believed that transcytosis across M cells is the most efficient pathway for particle translocation on a per cell basis. After translocation across the M cells, particles still need to traverse the underlying basement membrane. The basement membrane beneath the M cells contains pores of 3 μm or larger [31], and is not believed to significantly impede movement of particulates in the nanometer range. Pores in the basement membrane of the adjacent villus core are considerably smaller. The unique pore structure of the M cell basement membrane appears to play an important role in facilitating antigen-to-cell and cell-to-cell interactions during an immune response [31]. M cells contain a special pocket into which transcytosed particles are released after crossing the basement membrane. This pocket is filled with lymphocytes and some macrophages, forming an additional barrier to particle access to the circulation. The capillary network underlying follicular epithelium is considerably less dense than that underlying the villus epithelium 32, 33. Allan and co-workers [34]demonstrated that there are twice as many FAE epithelial cells per subepithelial capillary in follicle domes than in villi. They further observed that the endothelial wall of the follicle capillary had fewer fenestrae than the endothelial wall of the villus capillary. Following intravenous injection, accumulation of horseradish peroxidase (mw 40 kD) and hemoglobin (mw 64.5 kD) in the interstitial space in the subepithelial villus area was found to be 4 and 48 times, respectively, greater than in the subepithelial dome area (e.g., Peyer's patch) [34]. These results demonstrate that in addition to being less dense, the follicle capillary network is also considerably less permeable than the villus capillary network. It has been suggested that because of these morphological characteristics antigens are retained in the Peyer's patch to elicit an optimal mucosal immune response [34]. Generally, larger particles are removed via the lymphatics [35], and these particles may be trapped within macrophages in mesenteric lymph nodes 36, 37, 38.
Although M cells may provide a rather efficient route for the translocation of certain particles, FAE surface area overall still accounts for only a small fraction of the total intestinal surface area. In addition, because of the non-fenestrated capillary endothelium in Peyers patches, direct access of particulates to the circulation is impeded, and those particles removed by lymphatic drainage may not have access to the circulation due to trapping in lymph nodes. Thus, generally, only a small fraction of the total particle dose appears in the systemic circulation 20, 21, 22. Low absorption of particles obviously limits their potential for systemic drug delivery of therapeutic compounds. However, because of M cell transcytosis, particulate delivery holds promise especially for the delivery of vaccines. Antigens delivered via particulate carriers may be protected from degradation and will be presented directly to the mucosal immune system following transcytosis by the M cells.
In the following sections we will review the anatomy and physiology of the Peyer's patch, with particular emphasis on Peyer's patch distribution in the gastrointestinal tract in different animals, M cell differentiation, surface properties of M cells, and transcytosis of particulates by M cells.
Section snippets
General properties
Peyer's patches are collections of lymphoid follicles, which are separated from the intestinal lumen by a single layer of specialized epithelium containing M cells and enterocytes, i.e. FAE. This epithelium is different from the villus epithelium in that the enterocytes are more cuboidal, it contains fewer goblet cells, there is no secretory component [39], and it has reduced activity for some hydrolases in the apical membrane [40]. Peyer's patches play a central role in antigen uptake and
M cell differentiation
Peyer's patch epithelial cells, like villus epithelial cells derive from the surrounding crypts. Follicle associated crypts contain two distinct axes of migration and differentiation. Cells on one axis differentiate into absorptive enterocytes, goblet cells and endocrine cells that migrate up to the villus, while cells on the other axis migrate onto the follicular dome, where they differentiate into mature M cells and distinct follicle-associated enterocytes [56]. For recent reviews on M cells
Conclusions
The utility of particulate dosage forms for the oral delivery of peptides and proteins is currently limited. The only intestinal epithelial cells that are capable of efficient particle translocation are the M cells in Peyer's patches. M cell morphology and Peyer's patch anatomy allows for efficient particle translocation across the epithelium and retention of particles in the follicle, thus ensuring optimal interaction with Peyer's patch lymphocytes. Identification of M cell surface-specific
Acknowledgements
The authors would like to express their gratitude to Dr. Doris Wall for critically reading the manuscript and Thomas Covatta for providing the Peyer's patch micrograph.
References (91)
- et al.
Oral absorption of peptides and proteins
Adv. Drug Del. Rev.
(1992) Paracellular intestinal transport: Modulation of absorption
Adv. Drug. Del. Rev.
(1991)- et al.
Molecular structure of the apical junction complex and its contribution to the paracellular barrier
J. Pharm. Sci.
(1997) - et al.
The enterotoxic effect of Zonula occludens toxin on rabbit small intestine involves the paracellular pathway
Gastroenterology
(1997) - et al.
The importance of gastrointestinal uptake of particles in the design of oral delivery systems
Adv. Drug Del. Rev.
(1995) Intestinal translocation of particles – implications for drug and antigen delivery
Adv. Drug. Del. Rev.
(1990)- et al.
Further histological evidence of the gastrointestinal absorption of polystyrene nanospheres in the rat
Int. J. Pharm.
(1992) - et al.
Current options for vaccine delivery systems by mucosal routes
J. Control. Rel.
(1997) - et al.
Porosity of the basement membrane overlying Peyer's patches in rats and monkeys
Gastroenterology
(1986) - et al.
Microcirculation of intestinal lymphoid follicles in rat Peyer's patches
Gastroenterology
(1981)