Structural requirements for intestinal absorption of peptide drugs

ELSEVIER Journal of Controlled Release 41 (1996) 3-17

journal of control led

release

Structural requirements for intestinal absorption of peptide drugs

Giovanni M. Pauletti, Sanjeev Gangwar, Gregory T. Knipp, Manoj M. Nerurkar, Franklin W. Okumu, Kiyoshi Tamura, Teruna J. Siahaan, Ronald T. Borchardt*

Department g( Pharmaceutical Chemistry. 2095 Constant Ave., The University of Kansas, Lawrence, KS 66047, USA

Received 20 May 1995; revised 3 January 1996; accepted 27 January 1996

Abstract

The clinical development of orally active peptide drugs has been restricted by their unfavorable physicochemical characteristics, which limit their membrane permeation, and their lack of stability against enzymatic degradation. Successful oral delivery of peptides will depend, therefore, on strategies designed to alter the physicochemical properties of these potential drugs, without changing their biological activity, in order to circumvent the barrier properties of the intestinal epithelial cells. This manuscript will focus on the physical and metabolic barrier functions of the intestinal mucosa, the structural features of peptides which influence their passive diffusion and carrier-mediated transport, including efflux mechanisms, and various approaches used to prevent enzymatic degradation of the peptides and increase their permeability across the intestinal mucosa.

Keywords: Peptide delivery; Intestinal barrier functions; Peptide modifications; Prodrugs; Peptidase inhibitors

1. Introduct ion

Recent dramatic advances in recombinant DNA technology and modern synthetic methodologies allow the production of vast quantities of various peptides possessing a diverse array of pharmacological effects. The clinical development of these peptide drugs, however, has been restricted due to very poor absorption across cell membranes and rapid degradation leading to oral bioavailabilities typically less than 1-2% and short in vivo half-lives [1-6]. The successful design of such molecules as

Co~espondlng author. Tel: +1-913-8644820; Fax: +1-913- 8645736.

0168-3659/96/$15.00 © 1996 Elsevier ScienceIreland Ltd. Allr ights PII S 0 1 6 8 - 3 6 5 9 ( 9 6 ) 0 1 3 5 2 - 1

orally available drugs will be a major challenge confronting pharmaceutical scientists in the future. The choice of strategies employed in designing a peptide drug requires careful consideration of the demands imposed by the interactions of the peptide with its pharmacological receptor, its site of action, its in vivo potency and specificity, and its desired route of administration [7]. Designing a suitable structure for a given molecule frequently necessitates a compromise concerning structural features in order to achieve a balance between the optimal pharmacological properties and the optimal pharmaceutical properties (e.g. permeability). This is further compli- cated by the need to protect against enzymatic degradation. This manuscript will focus on the

reserved

4 G.M. Pauletti et al. / Journal of Controlled Release 41 (1996) 3 -17

barrier properties of the intestinal mucosa and how the structural features of a peptide influence its ability to permeate this biological barrier.

2. Barrier properties of the intestinal mucosa

The epithelium lining the gastrointestinal tract clearly acts as a strategic interface between the external environment and the internal milieu of the body. This strategic interface is both a physical barrier, which restricts peptide flux to paracellular and transcellular pathways (Fig. 1A), and a biochemical barrier, which includes pathways of metab-

A Pe

II II H li H

,tide

"~ Peptide

7 I!

Apica l

N

Basolateral

B Metabolism

N N

m

)" M Em

II 1 H P

Eff lux System

P P

II ] H II

P

Apica l

Basolatera[

P = peptide Ebb = brush-border enzymes M = metabolite E~ = intraceUular enzymes

• = apically polarized efflux system

Fig. 1. Barrier properties of the intestinal mucosa. (A) The physical barrier limits peptide flux to (a) paracellular and (b) transcellular pathways. (B) The biochemical barrier reduces transport of peptides through the intestinal mucosa by brush- border and/or intracellular metabolism and by apically polarized efflux systems.

olism and apically polarized efflux systems (Fig. 1B). Over the past two decades, however, it has been recognized that the intestinal mucosa is not an absolute barrier that totally prevents peptide permeation. Various examples of biologically active peptides, including thyrotropin-releasing hormone (TRH) [8] octreotide [9], 1-deamino-8-o-arginine vasopressin (DDAVP) [10,11], calcitonin [12] and delta sleep-inducing peptide (DSIP) [13] have been reported to permeate from the intestinal lumen into the systemic circulation. Nevertheless, with peptides and peptidomimetics, one typically observes poor and variable oral bioavailability [1,3-6,14]. The following section describes the multiple barrier functions of the intestinal mucosa that may be responsible for low peptide oral bioavailability.

2.1. Metabolic barrier

Physiologically, the gastrointestinal tract is designed to break down dietary proteins into subunits sufficiently small to be absorbed. Digestive processes for proteins are catalyzed by a variety of enzymes specialized in the hydrolysis of peptide bonds. Due to the wide substrate specificity of these peptidases, it is not surprising that the metabolic barrier is considered to be the most important of the multitude of barriers limiting the absorption of peptide drugs. Much of the information on proteolytic enzymes in the gastrointestinal tract was obtained in the 1960s and 1970s, when researchers were primarily interested in dietary protein/peptide assimilation [15- 17]. Recently, several reviews were published which focused more on the role of the enzymatic barrier to biologically active peptides [2,14,18].

Upon reaching the duodenum, peptide degradation can be mediated by pancreatic proteases. The relative importance of this luminal hydrolysis in the overall degradation is dependent on the size and the respective amino acid composition of the peptide [17]. However, even when luminal peptide degradation occurs as observed for glycine-containing di- and tripeptides, it constitutes at best 20% of the total degradation in a given intestinal segment. This implies that significant degradation of the peptide requires at least contact with the brush-border membrane or uptake into the cell.

Although lysosomes and other organelles may act

G.M. Pauletti et al. / Journal of Controlled Release 41 (1996) 3 - 1 7

as potential sites of peptide metabolism, peptidases in the brush-border membrane are probably the biggest deterrent to the absorption of small peptides across the intestinal mucosa [17,19]. In general, it appears that brush-border peptidases are active mainly against tri-, tetra-, and higher peptides up to ten amino acid residues [20,21], while intracellular peptidases are active predominantly against dipeptides [22]. Today, a variety of intestinal peptide hydrolases are characterized and listed under the formal enzyme classification (EC) system based on their site of action in a susceptible substrate [23,24]. The properties and the physiological role of peptidases in the body have been extensively reviewed by Bond and Beynon [25].

The regional distribution of intestinal peptidases, including their activity, has been studied recently. In rat and rabbit intestine, the activity of the brush- border exopeptidases, aminopeptidase P (EC 3.4.11.9) and aminopeptidase W (EC 3.4.11.16), increases distally and reaches its highest level in the ileum [26]. The lowest activities, however, were measured in the ileo-caecal junction. Similar results were found in rabbit as well as in human for aminopeptidase N (3.4.11.2) and dipeptidyl peptidase IV (3.4.14.5) [27,28]. In the rat, the activity of aminopeptidase N increases toward the mid-intestine and then decreases distally, whereas dipeptidyl peptidase IV activity decreases slightly toward the distal end [29]. In contrast, the activities of cytosolic peptidases do not show any regional variation [30]. Lysosomal peptidases, on the other hand, seem to have their highest activities predominantly in the caecum [31].

Peptidase activity in the gastrointestinal tract is reported to be sensitive to various factors including age [32], diet [33,34], and administration of drugs [35]. However, their impact on the absorption of biologically active peptides has not been extensively explored.

2.2. Physical barrier

In addition to the metabolic barrier, the intestinal epithelium certainly represents an important physical barrier. The organization and architecture of the intestinal mucosa, which have been extensively reviewed elsewhere [36,37], allow peptides to

traverse the cell barrier via the paracellular and/or the transcellular route (Fig. 1A). The paracellular pathway is an aqueous, extracellular route across the epithelia that is followed by molecules according to their size and charge [38]. Translocation through the paracellular pathway is primarily passive.

The main barrier to the paracellular diffusion of molecules and ions across the epithelial cell layer is primarily the region of the tight junction or zonula occludens. It is one of three distinct morphological elements of the epithelial junctional complexes, the other two being the zonula adherens and the macula adherens or spot desmosomes [39]. In cases where tight junctions are leaky, the rate of permeation through the paracellular pathway may also be restricted by the narrowness and tortuosity of the intercellular space, as well as by the composition of the extracellular matrix [40]. Although the degree of permeability at the tight junctions varies significantly within different epithelia, tight junctions are generally reported to be impermeable to molecules with radii larger than 11-15 ]~ [41-44]. The pore size in the colon, which is assessable for the paracellular route, has been estimated to be between 8 and 9 [45]. These numbers may also represent the maxi- mum hydrodynamic radius of a spherical rigid molecule, which is small enough to diffuse into the intercellular space. However, for peptide drugs possessing a high degree of conformational flexibility, it might be possible that even larger molecules can permeate the tight junctions. Permeability at the tight junctions can be modulated by various endogenous [46-48] and exogenous factors [49,50]. In the future, this may provide a useful approach to enhance peptide permeability via the paracellular route.

The transcellular pathway involves movement of the solute across the apical cell membrane, through the cell interior and across the basolateral membrane by either active or passive processes (Fig. 2). It is well known that di- and tripeptides are absorbed by both a carrier-mediated process and simple diffusion. The latter, however, appears to be of minimal importance [51]. In general, active processes are fairly substrate specific and do not contribute to the transport of non-substrates, although exceptions to this have been found [52]. Cellular internalization of polypeptides by endocytosis is another important biological process, whereby peptides too large to be

6 G.M. Pauletti et al. / Journal of Controlled Release 41 (1996) 3 - 1 7

P P P • Apical

. ill.iLl-il. H II 1 HV~ II H~ II Basolateral

P P P

P '= peptide

Fig. 2. Transce]lular pathways for intestinal absorption of peptides: (a) passive (b) cartier-mediated, (c) vesicular.

absorbed by the di- and tripeptide transport systems may be taken up. The fluid-phase endocytosis (pinocytosis) does not require any interaction between the polypeptide and the apical membrane. In contrast, the receptor-mediated oral absorptive endocytosis involves binding to the plasma membrane before being incorporated into endocytic vesicles. Finally, polypeptides can be carried in endosomes directly to the basolateral side (i.e. bypassing the lysosomes), where they are released into the extracellular space. This process is known as transcytosis. Although there is some evidence that mucosal peptide/protein uptake is mediated by endocytic processes [53,54], in most instances this does not lead to transcytosis.

In contrast, a specialized type of undifferentiated crypt cell, called M-cells in the Peyer's patches, have significant ability to transcytose polypeptides. Their function and potential role in oral polypeptide absorption have been described recently [55].

Passive diffusion within the lipid bilayer from the apical to the basolateral side is for many biologically active peptides inaccessible due to their relatively high hydrophilicity. However, for more hydrophobic derivatives, including peptide-mimetics and peptide prodrugs, the lipid-soluble pathway may be a possible route for oral delivery.

Although the cell layer contributes predominantly to the physical barrier of the intestinal mucosa, additional factors may hinder the passage of pep-

tides. Like most epithelial surfaces, the intestinal mucosa is also coated with a layer of mucus that serves as a lubricant and protective barrier. Mucus, in reality, is a constantly changing mix of many secretions, including exfoliated epithelial cells [56]. The main determinants of the physical and functional properties of mucus secretions are high molecular weight glycoproteins, termed mucins. Much research has been done in understanding the regulatory mechanisms of mucin secretion and its role in the modulation of tissue function [57]. However, the role of the mucus layer as a physical barrier in the absorption of peptides from the gastrointestinal tract has not been well established.

2.3. Efflux systems

In recent years, the barrier function of the intestinal mucosa could not be adequately described by a combination of the metabolic and physical barriers alone. It has long been recognized that polarized efflux systems are present in cancer cells that pose a major barrier to the absorption of a wide variety of chemotherapeutic agents. Although these efflux systems are most commonly observed in tumor cells, they are also known to be present in normal intestinal cells [58]. These efflux systems are related to P-glycoprotein, the principle component of multidrug resistance in a variety of cell types. P-glycoprotein is a 170-180 kDa membrane glycoprotein that acts as an ATP-dependent efflux pump, thereby reducing the intracellular accumulation or the transcellular flux of a wide variety of drugs, including peptides (e.g. gramicidin D, valinomycin) [59]. The polarized expression of these efflux systems suggests that their physiological role is to restrict transcellular flux of some molecules. Thus, in the gastrointestinal epithelia, they serve as a major barrier by limiting the absorption of drugs. Recent studies with inhibitors suggested that, in addition to a verapamil-sensitive efflux system, there is at least one other polarized efflux system present in the apical membrane of intestinal cells, which in MDCK cells, could be inhibited by indomethacin but not by verapamil [60]. These observations indicate that there may exist a family of efflux systems in the apical region of the intestinal mucosa.

Several models have been proposed to explain the

G.M. Pauletti et al. / Journal of Controlled Release 41 (1996) 3 - 1 7 7

mechanism of multidrug resistance through polarized efflux systems, and most of them support the conclusion that drugs are removed directly from the phospholipid membrane. The idea is that the efflux pump is essentially a 'Flippase' that detects drugs in the inner leaflet of the phospholipid membrane and 'flips' it into the outer leaflet of the membrane where it can diffuse away from the cell [59].

In order to increase the efficacy of antitumor agents in cancer therapy, medicinal chemists have become interested in the design and synthesis of potential inhibitors of these efflux systems. In addition, it has been shown that commonly used formulation additives, like surfactants, can be used to inhibit these efflux systems, thereby enhancing the permeability of peptides that are substrates for the efflux systems [61].

3. Structural features that influence the passive diffusion of peptides

3.1. Paracellular

The paracellular route has gained interest for the delivery of peptides because of the perception that it is deficient in proteolytic activity [14]. Several peptide drugs such as octreotide [62,63], potent analogs of vasopressin (e.g. DDAVP) [64], and TRH [65] are assumed to permeate the intestinal mucosa mainly via this route. Since the paracellular pathway is an aqueous extracellular route across the epithelia, structural features that infuence the paracellular absorption of peptide drugs are size, charge and hydrophilicity. Of these structural features, molecular size probably plays the most significant role due to the size restriction of the junctional complexes (see Physical barriers). This sieving effect of the tight junctions may explain the results obtained in a study where a series of model peptides, DPheGly, DPhezGly, and DPhe3Gly, showed decreasing flux with increasing molecular radius [66]. Recently, our laboratory focused on the effect of conformation on the permeation of model peptides through Caco-2 cell monolayers, an in vitro model for the intestinal mucosa [67]. We compared linear, metabolically stabilized hexapeptides, Ac-Trp-Ala-Gly-Gly-X-Ala- NH 2 (e.g. X=Asp) , with cyclic analogs, cyclo[Trp-

AIa-Gly-Gly-X-Ala] (e.g. X=Asp), which were co- valently linked by the N-terminal and C-terminal ends. Solution structure analysis by means of two- dimensional NMR revealed that for the linear hexapeptides significant amounts of secondary structure (e.g. flI turns) exist in a dynamic equilibrium with unfolded solution structures. In comparison, the cyclic analogs predominantly existed in a well-defined conformation containing two flII turns. When the permeability of the Asp-containing peptides was determined across Caco-2 cell monolayers, cyclo- [Trp-Ala-Gly-Gly-Asp-Ala] was shown to be two to three times more permeable than Ac-Trp-Ala-Gly- Gly-Asp-Ala-NH 2. This implies that cyclization may have decreased the average molecular size, account- ing for the observed increase in flux.

Proteins within the junctional complexes consist of polar amino acids with ionizable side chains. As a consequence, the junctional space exhibits an electrostatic field with a negative net charge that may affect the paracellular flux of molecules due to charge-charge interactions [49,68-70]. Differences in the flux of positively and negatively charged peptides have been reported [43,44]. However, the most favorable charge of a peptide for the paracellular passage is still controversial. Experiments performed with a series of cyclic RGD peptides indicate that peptides with a net charge of - 1 to - 2 exhibit optimum permeability for the paracellular route [71,72]. In contrast, a study with small anionic, neutral, and cationic non-peptide markers revealed higher permeability coefficients for the cationic than anionic species [66]. Our laboratory recently demonstrated that the flux of several model peptides, Ac- Trp-Ala-Gly-Gly-X-Ala-NH 2 (X=Asp, Lys, Asn) and Ac-Tyr-Pro-X-Z-Val-NH 2 (X=GIy, Ile and Z = Asp, Asn), in Caco-2 cell monolayers did not show any discrimination based on the difference in charge [73,74]. It was concluded that the molecular radius and not the charge predominantly limits the permeation of these peptides through the cell monolayers.

Besides the importance of size and charge, the third physicochemical property that determines the pathway by which a molecule traverses the intestinal mucosa is hydrophilicity. Since the paracellular pathway is an aqueous route, a sufficient hydrophilicity is a prerequisite. The more lipophilic a peptide is, the more likely it will interact with the

8 G.M. Pauletti et al. I Journal of Controlled Release 41 (1996) 3 - 1 7

cell membrane, specifically with the phospholipid bilayer. Interactions of this nature would shift the permeation toward a transcellular route [64].

3.2. Transcellular

'Transcellular permeation requires a distinct interaction of the solute with the membrane. Since the latter represents a very diverse chemical environment, the ability of a peptide to traverse this barrier by passive diffusion depends on its interaction not only with the lipid bilayer but also with many different integral and peripheral membrane proteins [75]. Most of the studies performed to date have focused on the individual contributions of physicochemical properties, including size, lipophilicity (hydrophobicity and hydrogen bonding potential) and conformation, on the transcellular permeation of peptides. However, the controlling features for this route are still not well understood.

Traditionally, lipophilicity has been viewed as the most important molecular characteristic in determin- ing passive diffusion through biological membranes, mainly because a membrane is simplistically considered a 'lipophilic' barrier. Nevertheless, early in vivo data suggested that intestinal absorption may decline when lipophilicity becomes too high [76]. These results imply an 'optimal' rather than a high lipophilicity for improved transmucosal permeability of a molecule. Burton and co-workers [75] extensively studied the effect of lipophilicity on transport of model peptides in Caco-2 cell monolayers. It was demonstrated that lipid solubility as expressed by the octanol/water partition coefficient does not influence the membrane permeability of peptides to the same extent as shown for small organic molecules. A better predictor for peptide permeability was the respective hydrogen bonding potential of these compounds, which was experimentally assessed by the difference in partition coefficients determined in the octanol/water and the isooctane/water system [75,77,78]. The same conclusion was drawn when the transport of these model peptides was studied in an in situ perfused rat ileum model [79]. This implies that hydrogen bonding potential may represent a valuable parameter to characterize the lipophilicity of peptides with respect to their transmembrane permeability.

Another approach to explore the ideal physico-

chemical properties for the transcellular pathway of peptides is estimation of the polar surface area of the molecule by means of molecular modeling [7]. From

0 2 this study, a polar surface area of --<50 oA, as compared to a total surface area of --<350 A-, was found to provide a tetra- to hexapeptide with 'optimal' characteristics for transcellular permeability.

The concept of optimized physicochemical properties has been shown to be very useful in the rational design of orally available peptidic renin inhibitors [80]. By chemical modification, the oral bioavailability could be substantially increased (up to >50%) for the 'optimal' derivative, whereas the parent compound exhibited a bioavailability of <2%.

4. Structural features that influence the carrier- mediated transport of peptides

It is well known that the intestinal absorption of natural di- and tripeptides is mainly an active, carrier-mediated process [81-86]. In the past 10 years, interest has increasingly focused on this carrier-mediated pathway, since there is evidence that the oligopeptide transporter(s) contributes to the intestinal transport of various peptidomimetics (e.g. amino-/3-1actam antibiotics, angiotensin-converting enzyme inhibitors, renin inhibitors) [87-92]. To take advantage of the oligopeptide transporter(s) for drug delivery, one needs a greater understanding of what structural features of a peptide optimize its binding to the transporter(s), its cellular uptake and its transepithelial transport. However, this information is particularly difficult to obtain without knowing the specific number of transporters involved.

In our overview we summarize structural elements that were found as favorable or unfavorable for affinity to the intestinal oligopeptide transporter(s). Unfortunately, these data are mostly based on the results obtained from competition studies rather than correlation between peptide structure and transport. Since molecules can bind to the peptide transporter(s) without necessarily undergoing transport [93- 95], it is obvious that the proposed optimal structural features for binding do not represent sufficient characteristics for transport via this carrier-mediated route. Nevertheless, binding to the transporter(s) represents the initial step for a peptide traversing the intestinal cell by this pathway. Assuming that more


information on the effect of these structural elements not only for binding but also for transcellular transport is forthcoming, it may be possible in the near future to design peptide drugs that are efficiently absorbed via the oligopeptide transporter(s) pathway.

Modification of the C-terminal carboxyl group generally leads to reduction or abolition of the affinity to the transporter(s). Examples include studies with L-Val-L-Val or o-Val-L-Val dipeptides, in which the free carboxyl group was esterified to a methyl ester, or with L-VaI-L-Val, in which the free carboxyl was reduced to the alcohol [96,97]. Further- more, affinity was significantly reduced, when the carboxyl group was amidated [98,991. This implies that, for optimal interaction with the transporter(s), a free carboxyl group is required. Similarly, cyclization of the dipeptide Gly-Pro and Ala-Ala led to abolition of affinity to the intestinal oligopeptide transporter(s) [96]. However, when the dipeptide contained a net negative charge in the side chain (e.g. Glu-Glu, Asp-Asp), cyclization did not alter its ability to interact with the oligopeptide transporter(s). Thus, the inability of conformationally constrained (e.g. cyclized) dipeptides to interact with the oligopeptide transporter(s) can be overcome through judicious selection of amino acids in the dipeptides.

In contrast to the carboxyl group, which appears to be an essential structural element for the affinity to the transporter(s), the N-terminal a-amino group is not required for recognition by the oligopeptide transporter(s) [100]. These findings are consistent with the fact that some /3-1actam antibiotics [91,101,102] and ACE inhibitors [92,103], which are transported via the oligopeptide transporter(s), lack an N-terminal a-amino group.

The effect of the net charge on the affinity of a dipeptide for the intestinal oligopeptide transporter(s) is still controversial. Wootton and Hazelwood [104] found that negative charges are more detrimental to the affinity for the carrier system than are positive charges. In comparison, Wood et al. [105] reported that dipeptides bearing two positive charges exhibited the lowest affinity, while neutral dipeptides showed the highest affinity.

The stereoselectivity of the intestinal oligopeptide transporter(s) has been characterized using various dipeptides that contain L- or D-amino acids as well as dipeptides consisting of both enantiomers (i.e. 'mixed' dipeptides) [87,96,106,107]. It appears that

only one L-amino acid must be present in dipeptides in order to maintain the necessary structure for interaction with the carrier.

The effect of molecular weight and size on the affinity to the intestinal oligopeptide transporter(s) has been studied for a wide range of peptidomimetics [108-110]. However, no systematic study has been conducted thus far with peptides.

5. Structural features that influence the efflux- mediated transport of peptides

One of the most interesting features of the efflux systems exhibiting multidrug resistance is their non- specificity toward a wide variety of chemotherapeutic agents including peptides. Various studies on the structure-activity relationship of drugs, differing in their ability to be taken up by the multidrug transporter and being transported to the luminal side of the membrane, demonstrate that the most important structural feature needed for substrate activity with this efflux carrier is hydrophobicity [111,112]. Moreover, the drugs that are substrates for the efflux systems are shown to have octanol/water partition coefficients of approximately one or greater [59]. These observations are supported by the efflux of hydrophobic peptides such as gramicidin D, valinomycin and N-acetyl-leucyl-leucyl-norleucinal [59,113]. However, sparingly water-soluble, highly hydrophobic drugs are not substrates for these efflux systems, which suggests that some water solubility is a necessary property for interaction with a multidrug transporter [114]. Recently, Burton and co-workers demonstrated that the permeability of various N- methylated peptides across Caco-2 cell monolayers was limited by apical polarized efflux systems [115]. In this case, no specific structural feature could be identified that favors interaction with the efflux systems.

6. Strategies to increase the metabolic stability of peptides

Several approaches were proposed to circumvent protease action, thus leading to a higher and more reproducible bioavailability for peptide drugs. The focus of the discussion below will be on the chemi-

10 G.M. Pauletti et al. / Journal (~t' Controlled Release 41 (1996) 3 - 1 7

cal modification of peptides and the effect of peptidase inhibitors.

7. Chemical modifications

7.1. N- and C-terminal modifications

In recent years, medicinal chemists have made tremendous progress in resolving the metabolic instability problem through structural manipulation of peptides. The chemical modifications tend to optimize not only potency and selectivity but, most importantly, stability of peptides toward proteolysis. One of the most common modifications is substitution of L-amino acid with D-amino acid at either or both termini in order to prevent degradation. For example, when D-Ala was substituted in place of Gly 2 in methionyl-enkephalin (Tyr-Gly-Gly-Phe- Met), it showed a significantly longer half-life than the parent peptide. This is attributed to the inability of aminopeptidases to cleave the Tyr-D-Ala amide bond readily [ 116]. Other modifications, which have been used to prevent degradation by aminopeptidases, include N"-acetyl, -formyl, and other N ~- alkylations [117]. The use of a D-amino acid in place of an L-amino acid at the C-terminus can also inhibit carboxypeptidase-mediated degradation. In cases where D-amino acid replacement is not possible due to loss of biological activity, other strategies have been used (e.g. conversion of the C-terminal amino acid residue into an ester or an amide). In the case of metkephamid (Tyr-D-Ala-Gly-Phe-MeMet-NH2), which contains both N- and C-terminal modifications, the half-life in vivo was significantly longer [118]. Some other modifications of the C-terminal end are converting the carboxy end to amino alcohol, methyl ketone, nitriles, tetrazoles and decarboxy amino acids. The amino alcohol method was successful with enkephalin peptide, where the C-terminal methionine was converted to a methioniol residue [119,1201.

One modification, which inhibits both aminopeptidases as well as carboxypeptidases, is converting a linear peptide to a cyclic analog [121]. This could be achieved by forming an amide bond between the N-and C-termini, between a side chain and the N-or C-terminus, or between two side chains. In the case of a cyclic analog between Asp 5 and Lys l° of a -

melanotropin (c~-MSH), the half-life was 100 times greater than that of the parent linear peptide [122]. Recently, our laboratory reported the effect of N- to C-terminus cyclization on the metabolism and permeability properties of a series of model peptides (Trp-Ala-Gly-Gly-X-Ala, e.g. X=Asp) in Caco-2 cell monolayers [67]. The linear Asp-containing hexapeptide, for example, was rapidly metabolized (apparent half-life: 11 min), mainly by the action of aminopeptidases, when applied to the apical side of the cell monolayer. However, for the cyclic analog, no decrease in concentration was observed within a 180 min incubation period. This indicated a significantly increased metabolic stability of the peptide due to the formation of a N- to C-terminal linked cyclic analog. However, peptide cyclization can result in major structural and conformational changes that may affect the biological activity of the peptide.

7.2. Prodrug approaches

The use of prodrugs is another important approach for the chemical modification of peptide drugs. A prodrug is defined as a pharmacologically inactive chemical derivative of a parent drug molecule that requires spontaneous or enzymatic transformation within the body in order to release the active drug. Prodrugs are usually designed to overcome pharmaceutical and/or pharmacokinetic problems [ 123,124]. The existing literature on prodrugs of peptides is limited and still evolving. The prodrug approach has been shown to work quite nicely in the case of TRH (pyro-Glu-L-His-L-Pro-NH2). Various prodrugs (N- alkoxycarbonyl derivatives) of TRH have been shown to be resistant to cleavage by enzymes [ 125,126]. Prodrugs (4-imidazolidinone derivative) of Leu-enkephalin and Met-enkephalin have also been shown to have greater metabolic stability. These derivatives are converted to the parent enkephalin via non-enzymatic hydrolysis [127].

Our laboratory has prepared recently cyclic esterase-sensitive prodrugs of a model peptide, Trp- Ala-Gly-Gly-Asp-Ala, using two different pro-moi- eties [128,129]. These prodrug systems utilize either a acyloxyalkoxycarbamate pro-moiety (Fig. 3A) or a 3-(2'-hydroxy-4',6'-dimethylphenyl)-2,2-dimethyl propionic acid pro-moiety (Fig. 3B). In human blood, the conversion of both prodrugs to the peptide was significantly more rapid than in buffer, suggesting

G.M. Pauletti et al. I Journal of Controlled Release 41 (1996) 3 - 1 7 11

S' ½ H

Ester~e ~[Slow

RCHO + C O 2 " ~ / I Chemical

~ Fast

+H3N-[ PEPTIDE |COO-

B I PEPTIDE

Esterase I Slow

Chemical Fast

+H3N- I PEPTIDE|COO" +

O

Fig. 3. Prodrug approaches to prepare cyclic peptides with enhanced membrane permeability and enzymatic stability. (A) acyloxyalkoxycarbamate pro-moiety, (B) 3-(2'-hydroxy-4',6'-dimethylphenyl)-2,2-dimethyl propionic acid pro-moiety.

esterase-mediated hydrolysis. In addition, when per- meabilities were determined in Caco-2 cell monolayers, the prodrugs were five times more permeable than the metabolically stabilized, linear hexapeptide, Ac-Trp-AIa-Gly-Gly-Asp-Ala-NH 2 [130,131]. The linear, unprotected hexapeptide, however, was rapidly metabolized (apparent half-life: 11 min) on the apical side of the monolayer and no transport was observed.

7.3. Peptidomimetics

Recently, rapid screening of small molecule li- braries and rational design approaches have produced peptidomimetics as a new generation of promising drugs. A peptidomimetic is defined as a molecule

that mimics the biological activity of a peptide, but is no longer completely peptidic in nature, such as pseudo-peptides, semi-peptides and peptoids [132]. The progress in the design of peptidomimetics includes strategies ranging from simple replacement of a single atom in the amide bond to more complex mimetics showing specific secondary structure elements of peptides. One of the major goals of peptide backbone modification is to gain stability against rapid proteolysis. Practically every surrogate of the peptide bond, except the ester bond as in depsipep- tide, has been found to increase resistance toward enzymatic degradation. This suggests that there is a recognition of the polar amide bond by most of the proteolytic enzymes. Any type of backbone modification, except of the ester bond, will interrupt this recognition step, thereby enhancing the stability of

12 G.M. Pauletti et al. / Journal of Controlled Release 41 (1996) 3 - 1 7

peptidomimetic drugs against enzyme degradation [133].

Several types of amide bond surrogates have been thoroughly reviewed by Spatola and colleagues [134]. For example, the [Gly3~F(CH2 S) Phe4] - leucine enkephalin analog showed much greater stability against enzymatic degradation [135]. The alkylidene amide surrogate qr(C=C) is quite an interesting substitution because it does keep the high double bond character of an amide bond. Several enkephalin analogs with the amide bond surrogate have been shown to have a high metabolic stability against proteolysis [136].

As shown by the above examples, there are many modification strategies for peptide drugs now available to eliminate proteolysis problems. Depending on the site and mechanism by which a given peptide drug is metabolized, appropriate chemical modifications can be used to design metabolically stable analogs. It should be noted that these chemical modifications also change the overall lipophilicity of the molecule, which might also facilitate their passage through biological membranes.

Z4. Peptidase inhibitors

Co-administration of peptidase inhibitors has been shown to substantially improve the oral absorption of several peptides. Jennewein et al. [137] reported an l l.5-fold increase in the absorption of pentagastrin from the duodenum of the rat when the pancreatic duct was ligated to exclude the pancreatic digestive enzymes from the small intestine. The nonapeptide renin inhibitor Pro-His-Pro-Phe-His-Leu-Phe-Val-Phe was found to cross the adult rabbit jejunum 90% intact in the presence of phosphoramidon, a metallo- protease inhibitor [138]. In the absence of the inhibitor, however, more than 55% of the nonapeptide was degraded in 5 min and none was left after 30 min. In the case of arginine vasopressin, co- administration of aprotinin, a natural inhibitor of trypsin, caused a substantial increase in the an- tidiuretic response of nearly 50% [139].

All of these examples suggest that for these particular peptides the enzymatic barrier is more of an impediment to absorption than the physical barrier. In contrast, TRH is absorbed in man to less than 3% despite its metabolic stability in the gut

[140]. This is partly because absorption of this peptide is limited to the upper regions of the intestine, at least in the rat and dog [140]. Conse- quently, co-administration of peptidase inhibitors does not always increase the fraction absorbed from the luminal side of the mucosa into the systemic circulation. Physicochemical properties of the peptides, as well as the route of absorption (i.e. paracellular or/and transcellular), play a major role in limiting the overall permeability.

The selection of peptidase inhibitors depends on the type of target protease and its subcellular distribution. Primarily membrane-bound protease activity can easily be suppressed by simple co-administration of an effective inhibitor. Where the principal proteolytic activity is cytosolic, however, it is im- perative that peptidase inhibitors themselves be absorbed into the cell. This may be associated with significant problems, since such compounds are typically too large and too hydrophilic to partition into and diffuse across biological membranes. The extent of inhibition of an enzymatic reaction is dependent on both the nature and concentration of the inhibitor [141]. Recently, our laboratory described the metabolism of DSIP (Trp-Ala-Gly-Gly- Asp-Ala-Ser-Gly-Glu) in human intestinal (Caco-2) cell monolayers [142]. To achieve a substantial increase in the metabolic stability of DSIP on Caco-2 cell monolayers, a 'cocktail' of three different peptidase inhibitors was necessary, although aminopeptidases appeared to mediate the major pathway of metabolism in the absence of peptidase inhibitors. However, in the presence of bestatin alone, a potent arninopeptidase inhibitor, the metabolic pattern seemed to be shifted to degradation by the action of dipeptidyl peptidase IV and peptidyl dipeptidase A. This result indicates that stabilization of a peptide to prevent intestinal metabolism may require the use of several peptidase inhibitors acting in concert.

Unquestionably, peptidase inhibitors have been recognized as offering great potential to reduce intestinal metabolism of peptides and, therefore, increase their oral bioavailability. However, a con- cern when using peptidase inhibitors to promote peptide absorption is their possible effect on the mucosal cell (e.g. change on the tight junctions) [139] and physiological processes following absorption into the systemic circulation [143,144].


8. Conclusions

The mult ipl ic i ty o f barr ier m e c h a n i s m s in the

gas t rointes t inal tract represen ts a cons ide rab le chal-

lenge for successfu l pharmaceu t ica l de l ivery o f

pep t ides via enteral routes. Howeve r , intest inal ab-

sorpt ion o f b io logica l ly act ive pep t ides may be

poss ib le through an under s t and ing o f the m a n y

d i f ferent m e c h a n i s m s that regulate the mucosa l

barrier. Hence , pep t ides can be modi f i ed accord ing ly

in order to achieve e n h a n c e d chemica l and enzymat i c

stabil i ty and im proved permeabi l i ty proper t ies .

Never the less , every b io logica l ly act ive pept ide has to

be t reated individual ly to improve its pe rmea t ion

th rough the intest inal mucosa .

References

[1] M.J. Humphrey and RS. Ringrose, Peptides and related drugs: a review of their absorption, metabolism and excre- tion, Drug Metab. Rev. 17 (1986) 283-310.

[2] V.H.L. Lee and A. Yamamoto, Penetration and enzymatic barriers to peptide and protein absorption, Adv. Drug Deliv- ery Rev. 4 (1990) 171-207.

[3] V. Bocci, Catabolism of therapeutic proteins and peptides with implications for drug delivery, Adv. Drug Delivery Rev. 4 (1990) 149-169.

[4] N.B. Modi, Pharmacokinetics and pharmacodynamics of recombinant proteins and peptides, J. Control. Release 29 (1994) 269-281.

[5] X.H. Zhou, Overcoming enzymatic and absorption barriers to non-parenterally administered protein and peptide drugs, J. Control. Release 29 (1994) 239-252.

[6] G.L. Amidon and H.J. Lee, Absorption of peptide and peptidomimetic drugs, Annu. Rev. Pharmacol. Toxicol. 34 (1994) 321-341.

[7] D. Barlow and T. Satoh, The design of peptide analogues for improved absorption, J. Control. Release 29 (1994) 283- 291.

[8] C.Y. Bowers, A.V. Schally, F. Enzman, J. BNer and K. Folkers, Porcine thyrotropin-releasing hormone is (pyro)Glu- His-Pro(NH2), Endocrinology 86 (1970) 1143-1153.

[9] H.S. Fiissl, J. Domin and S.R. Bloom, Oral absorption of the somatostatin analogue SMS 201-995: theoretical and practi- cal implications, Clin. Sci. 72 (1987) 255-257.

[10] S. Lundin and H. Vilhardt, Absorption of l-deamino-8-D- arginine vasopressin from different regions of the gastrointestinal tract in rabbits, Acta Endocrinol. 112 (1986) 457- 460.

[11] H. Vilhardt and S. Lundin, In vitro intestinal transport of vasopressin and its analogues, Acta Physiol. Scand. 126 (1986) 601-607.

[12] J. Hastewell, S. Lynch, I. Williamson, R. Fox and M. Mackay, Absorption of human calcitonin across the rat colon in vivo, Clin. Sci. 82 (1992) 589-594.

[13] W.A. Banks, A.J. Kastin and D.H. Coy, Delta sleep-inducing peptide (DSIP)-like material is absorbed by the gastrointestinal tract of the neonatal rat, Life Sci. 33 (1983) 1587-1597.

[14] V.H.L. Lee, R.D. Traver and M.E. Taub, Enzymatic barriers to peptide and protein drug delivery, in: V.H.L. Lee (Ed.), Peptide and Protein Drug Delivery, Marcel Dekker, New York, 1991, pp. 303-357.

[15] D.B. Silk, J.A. Nicholson and Y.S. Kim, Relationships between mucosal hydrolysis and transport of two phenylalanine dipeptides, Gut 17 (1976) 870-876.

[16] S.A. Adibi and E.L. Morse, The number of glycine residues which limits intact absorption of glycine oligopeptides in human jejunum, J. Clin. Invest. 60 (1977) 1008-1016.

[17] S.A. Adibi, Intestinal transport of dipeptides in man: relative importance of hydrolysis and intact absorption, J. Clin. Invest. 50 (1971) 2266-2275.

[18] V.H.L. Lee, Enzymatic barriers to peptide and protein absorption, Crit. Rev. Ther. Drug Cartier Syst. 5 (1988) 69-97.

[19] R.F. Crampton, M.T. Lis and D.M. Matthews, Sites of maximal absorption and hydrolysis of two dipeptides by rat small intestine in vivo, Clin. Sci. 44 (1973) 583-594.

[20] D.M. Matthews and J.W. Payne, Transmembrane transport of small peptides, Curt. Top. Membr. Transp. 14 (1980) 331- 425.

[21] R.J. Kania, N.A. Santiago and G.M. Gray, Intestinal cell surface peptidase: potential role in protein digestion, Gas- troenterology 62 (1972) 768A.

[22] Y.S. Kim, W. Birtwhistle and Y.W. Kim, Peptide hydrolases in the brush-border and soluble fractions of small intestinal mucosa of rat and man, J. Clin. Invest. 51 (1972) 1419- 1430.

[23] A.J. Barrett and J.K. McDonald, Mammalian Proteases: a Glossary and Bibliography, Vol. 1, Endopeptidases, Academ- ic Press, New York, 1980.

[24] J.K. McDonald and A.J. Barrett, Mammalian Proteases: a Glossary and Bibliography, Vol. 2, Exopeptidases, Academic Press, New York, 1986.

[25] J.S. Bond and RJ. Beynon, Proteolysis and physiological regulation, Mol. Aspects Med. 9 (1987) 173-287.

[26] J.RF. Bai, Comparison of distribution of brush-border exo- and endopeptidases in rat and rabbit intestine, J. Pharm. Pharmacol. 46 (1994) 928-930.

[27] G.J. Leitch, Regional variations in the composition of purified brush-borders isolated from infant and adult rabbit small intestine, Arch. Int. Physiol. Biochem. 79 (1971) 279-286.

[28] T. Lindberg, Intestinal dipeptidases: dipeptidase activity in the mucosa of the gastrointestinal tract of the adult human, Acta Physiol. Scand. 66 (1966) 437-443.

[29] S. Miura, S.I. Song, A. Morita, R.H. Erickson and Y.S. Kim, Distribution and biosynthesis of aminopeptidase N and dipeptidyl aminopeptidase IV in rat small intestine, Biochim. Biophys. Acta 761 (1983) 66-75.

14 G.M. Pauletti et al. / Journal o f Controlled Release 41 (1996) 3 - 1 7

[30] E.E. Sterchi, The distribution of brush-border peptidases along the small intestine of the adult human, Pediatr. Res. 15 (1981) 884-885.

[31] J.EF. Bai, The regional differences in the mucosal-cell lysosomal proteases within the rat small intestine, Int. J. Pharm. 107 (1994) 133-140.

[32] G.F. Vaeth and S.J. Henning, Postnatal development of peptidase enzymes in rat small intestine, J. Pediatr. Gastroen- terol. Nutr. 1 (1982) 111-117.

[33] J.A. Nicholson, D.M. McCarthy and Y.S. Kim, The re- sponses of rat intestinal brush-border and cytosol peptide hydrolase activities to variation in dietary protein content, J. Clin. Invest. 54 (1974) 890-898.

[34] Y. Suzuki, R.H. Erickson, B.C. Yoon and Y.S. Kim, Tran- scriptional regulation of rat intestinal angiotensin-converting enzyme and dipeptidyl peptidase IV by a high proline diet, Nutr. Res. 15 (1995) 571-579.

[35] S. Miura, A. Morita, R.H. Erickson and Y.S. Kim, Content and turnover of rat intestinal microvillus membrane aminopeptidase. Effect of methylprednisolone, Gastroenterology 85 (1983) 1340-1349.

[36] J. Baker, l.J. Hidalgo and R.T. Borchardt, Intestinal epithelial and vascular endothelial barriers to peptide and protein delivery, in: V.H.L. Lee (Ed.), Peptide and Protein Drug Delivery, Marcel Dekker, New York, 1991, pp. 359-390.

[37] J.L. Madara and J.S. Trier, The functional morphology of the mucosa of the small intestine, in: L.R. Johnson (Ed.), Physiology of the Gastrointestinal Tract, Raven Press, New York, 1994, pp. 1577-1622.

[38] M. Cereijido, O. Ruiz, L. Gonzfilez-Mariscal, R.G. Con- treras, M.S. Balda and M.R. Garcfa-Villegas, The paracellular pathway, in: K.L. Audus and T. Raub (Eds.), Biological Barriers to Protein Delivery, Plenum Press, New York, 1993, pp. 3-21.

[39] M.G. Farquhar and G.E. Palade, Junctional complexes in various epithelia, J. Cell Biol. 17 (1963) 375-412.

[40] M. Cereijido, Introduction: evolution of ideas on the tight junction, in: M. Cereijido (Ed.), Tight Junctions, CRC Press, Boca Raton, FL, 1991, pp, 1-13.

[41] M. Cereijido, I. Meza and A. Martfnez-Palomo, Occluding junctions in cultured epithelial monolayers, Am. J. Physiol. 240 (1981) C96-C102.

[42] J.L. Madara and K. Dharmsathaphorn, Occluding junction structure-function relationships in a cultured epithelial monolayer, J. Cell Biol. 101 (1985) 2124-2133.

[43] P.S. Burton, R.B. Hill and R.A, Conradi, Metabolism and transport of peptides across the intestinal mucosa, Proc. Int. Symp. Control. Rel. Bioact. Mater. 14 (1987) 6-7.

[44] N.F.H. Ho, J.S. Day, C.L. Barsuhn, P.S. Burton and T.J. Ranb, Biophysical model approaches to mechanistic transepithelial studies of peptides, J. Control. Release 11 (1990) 3-24.

[45] M. Tomita, M, Shiga, M. Hayashi and S. Awazu, Enhance- ment of colonic drug absorption by the paracellular permeation route, Pharm. Res. 5 (1988) 341-346.

[46] K. Kubes, Nitric oxide modulates epithelial permeability in the feline small intestine, Am, J. Physiol. 266 (1992) GI 138-G1142.

[47] A.D. Marmorstein, K.H. Martell, D.R. Ratcliffe and E.B. Cramer, Epithelial permeability factor: a serum protein that condenses actin and opens tight junctions, Am. J. Physiol. 262 (1992) C1403-C1410.

[48] E.E. Schneeberger and R.D. Lynch, Structure, function, and regulation of cellular tight junctions, Am. J. Physiol. 262 (1992) L647-L661.

[49] J. Hochman and P. Artursson, Mechanisms of absorption enhancement and tight junction regulation, J. Control, Re- lease 29 (1994) 253-267.

[50] K. Atisook and J.L. Madara, An oligopeptide permeates intestinal tight junctions at glucose-elicited dilatations: implications for oligopeptide absorption, Gastroenterology 100 (1991) 719-724.

[51] D.M. Matthews, R.H. Gandy, E. Taylor and D. Burston, Influx of two peptides, glycylsarcosine and L-glutamyl-L- glutamic acid, into hamster jejunum in vitro, Clin. Sci. 56 (1979) 15-23.

[52] J. Lowther, S.M. Hammond, K. Russel and P.D. Fairclough, Uptake of cephalosporins by human intestinal brush-border membrane vesicles, J. Antimicrob. Chemother. 25 (1990) 183-184.

[53] K. Siminoski, P. Gonnella, J. Bemanke, L. Owen, M.R. Neutra and R.A. Murphy, Uptake and transepithelial transport of nerve growth factor in suckling rat ileum, J. Cell Biol. 103 (1986) 1979-1990.

[54] J.P. Kr~ihenbiihl and M.A. Campiche, Early stages of intestinal absorption of specific antibodies in the newborn. An ultrastructural, cytochemical, and immunological study in the pig, rat, and rabbit, J. Cell Biol. 42 (1969) 345-365.

[55] V.H.L. Lee, S. Dodda-Kashi, G.M. Grass and W. Rubas, Oral route of peptide and protein delivery, in: V.H.L. Lee (Ed.), Peptide and Protein Drug Delivery, Marcel Dekker, New York, 1991, pp. 691-738.

[56] G.J. Strous and J. Dekker, Mucin-type glycoproteins, Crit. Rev. Biochem. Mol. Biol. 27 (1992) 57-92.

[57] J.F. Forstner and G.G. Forstner, Gastrointestinal mucus, in: L.R. Johnson (Ed.), Physiology of the Gastrointestinal Tract, Raven Press, New York, 1994, pp. 1255-1283.

[58] W,H.M. Peters, C.E.W. Boon, H.M.J. Roelofs, T. Wobbes, F.M. Nagengast and P.G. Kremers, Expression of drug- metabolizing enzymes and P-170 glycoprotein in colorectal carcinoma and normal mucosa, Gastroenterology 103 (1992) 448 -455.

[59] M.M. Gottesman and I. Pastan, Biochemistry of multidrug resistance mediated by the multidrug transporter, Annu. Rev. Biochem. 62 (1993) 385-427.

[60] G.K. Collington, J. Hunter, C.N. Allen, N.L. Simmons and B.H. Hirst, Polarized efftux of 2'-7'-bis(2-carboxyethyl)- 5(6)-carboxyfluorescein from cultured epithelium cell monolayers, Biochem. Pharmacol. 44 (1992) 417-424.

[61] M.M. Nerurkar, P.S. Burton and R.T. Borchardt, The use of surfactants to enhance the permeability of peptides through Caco-2 cells by inhibition of an apically polarized efflux system, Pharm. Res. 13 (1996) 528-534.

[62] J. Drewe, G. Fricker, J. Vonderscher and C. Beglinger, Enteral absorption of octreotide: absorption enhancement by polyoxyethylene-24-cholesterol ether, Br. J. Pharmacol. 108 (1993) 298 303.


[63] U. Jaehde, R. Masereeuw, A.G. De Boer, G. Fricker, J.F. Nagelkerke, J. Vonderscher and D.D. Breimer, Quantification and visualization of the transport of octreotide, a somatostatin analogue, across monolayers of cerebrovascular endothelial cells, Pharm. Res. 11 (1994) 442-448.

[64] S. Lundin and P. Artursson, Absorption of a vasopressin analogue, 1-deamino-8-D-arginine-vasopressin (dDAVP), in a human intestinal epithelial cell line, Caco-2, Int. J. Pharm. 64 (1990) 181-186.

[65] D.T. Thwaites, B.H. Hirst and N.L. Simmons, Passive transepithelial absorption of thyrotropin-releasing hormone (THR) via a paracellular route in cultured intestinal and renal epithelial cell lines, Pharm. Res. 10 (1993) 674-681.

[66] A. Adson, T.J. Raub, P.S. Burton, C.L. Barsuhn, A.R. Hilgers, K.L. Audus and N.F.H. Ho, Quantitative approaches to delineate paracellular diffusion in cultured epithelial cell monolayers, J. Pharm. Sci. 83 (1994) 1529-1536.

[67] F.W. Okumu, G.M. Pauletti, D.G. Vander Velde, T.J. Siahaan and R.T. Borchardt, The effect of conformational flexibility on the permeability of a hexapeptide across monolayers of a cultured human intestinal epithelial cell (Caco-2 cells), Pharm. Res. 12 (1995) S-302.

[68] J.R. Pappenheimer and K.Z. Reiss, Contribution of solvent drag through intercellular junctions to absorption of nutrients by the small intestine of the rat, J. Membrane Biol. 100 (1987) 123-136.

[69] J.L. Madara, D. Barenberg and S.L. Carlson, Effects of cytochalasin D on occluding junctions of intestinal absorptive cells: further evidence that the cytoskeleton may influence paracellular permeability and junctional charge selectivity, J. Cell Biol. 102 (1986) 2125-2136.

[70] P. Claude and D.A. Goddenough, Fracture faces of zonulae occludentes from 'tight' and 'leaky' epithelia, J. Cell Biol. 58 (1973) 390-400.

[71] W. Rubas, M. Cromwell, T. Gadek, D. Narindray and R. Mrsny, Structural elements which govern the resistance of intestinal tissues to compound transport, Mat. Res. Soc. Symp. Proc. 331 (1994) 179-185.

[72] W. Rubas, J. Villagran, M. Cromwell, A. McLeod, J. Wassenberg and R. Mrsny, Correlation of solute flux across Caco-2 monolayers and colonic tissue in vitro, S.T.P. Pharm. Sci. 5 (1995) 93-97.

[73] F.W. Okumu, G.M. Pauletti, D.G. Vander Velde, T.J. Siahaan and R.T. Borchardt, The effect of charge and conformation on the permeability of a hexapeptide across monolayers of a cultured human intestinal epithelial cell (Caco-2 cells), Pharm. Res. 12 (1995) S-302.

[74] G.T. Knipp, D.G. Vander Velde, T.J. Siahaan and R.T. Borchardt, The effect of solution conformation and charge on the paracellular permeability of model pentapeptides across Caco-2 cell monolayers, Pharm. Res. 12 (1995) S-303.

[75] P.S. Burton Conradi, R.A., Hilgers, A.R., (B) Mechanisms of peptide and protein absorption. (2) Transcellular mechanism of peptide and protein absorption: passive aspects, Adv. Drug Delivery Rev. 7 (1991) 365-386.

[76] C. Hansch and T. Fujita, p, ~ ~- Analysis. A method for the correlation of biological activity and chemical structure, J. Am. Chem. Soc. 86 (1964) 1616-1626.

[77] R.A. Conradi, A.R. Hilgers, N.F.H. Ho and ES. Burton, The influence of peptide structure on transport across Caco-2 cells. II. Peptide bond modification which results in improved permeability, Pharm. Res. 9 (1992) 435-439.

[78] R.A. Conradi, A.R. Hilgers, N.F.H. Ho and ES. Burton, The influence of peptide structure on transport across Caco-2 cells, Pharm. Res. 8 (1991) 1453-1460.

[79] D.C. Kim, ES. Burton and R.T. Borchardt, A correlation between the permeability characteristics of a series of peptides using an in vitro cell culture model (Caco-2) and those using an in situ perfused rat ileum model of the intestinal mucosa, Pharm. Res. 10 (1993) 1710-1714.

[80] H.D. Kleinert, S.H. Rosenberg, W.R. Baker, H.H. Stein, V. Klinghofer, J. Barlow, K. Spina, J. Polakowski, E Kovar, J. Cohen and J. Denissen, Discovery of a peptide-based renin inhibitor with oral bioavailability and efficacy, Science 257 (1992) 1940-1943.

[81] D.M. Matthews, Intestinal absorption of peptides, Physiol. Rev. 55 (1975) 537-608.

[82] D.M. Matthews and S.A. Adibi, Peptide absorption, Gas- troenterology 71 (1976) 151-161.

[83] S.A. Adibi and Y.S. Kim, Peptide absorption and hydrolysis, in: L.R. Johnson (Ed.), Physiology of the Gastrointestinal Tract, Raven Press, New York, 1981, pp. 1073-1096.

[84] V. Ganapathy and F.H. Leibach, Peptide transport in intestinal and renal brush-border membrane vesicles, Life Sci. 30 (1982) 2137-2146.

[85] V. Ganapathy and F.H. Leibach, Is intestinal peptide transport energized by a proton gradient?, Am. J. Physiol. 249 (1985) GI53-G160.

[86] F. Navab and A.M. Asatoor, Studies of intestinal absorption of amino acids and a dipeptide in a case of Hartnup disease, Gut 11 (1970) 373-379.

[87] I.J. Hidalgo, F.M. Ryan, G.J. Marks and P.L. Smith, pH- dependent transepithelial transport of cephalexin in rabbit intestinal mucosa, Int. J. Pharm. 98 (1993) 83-92.

[88] K.I. Inui, T. Okano, H. Maegawa, M. Kato, M. Takano and R. Hori, H + coupled transport of p.o. cephalosporins via dipeptide carriers in rabbit intestinal brush-border membranes: difference of transport characteristics between cefix- ime and cephradine, J. Pharmacol. Exp. Ther. 247 (1988) 235-241.

[89] K. Iseki, M. Sugawara, H. Saitoh, K. Miyazaki and T. Arita, Comparison of transport characteristics of amino fl-lactam antibiotics and dipeptides across rat intestinal brush border membrane, J. Pharm. Pharmacol. 41 (1989) 628-632.

[90] A.H. Dantzig and L. Bergin, Carrier-mediated uptake of cephalexin in human intestinal cells, Biochem. Biophys. Res. Commun. 155 (1988) 1082-1087.

[91] W. Kramer, C. Dechent, F. Girbig, U. Gutjahr and H. Neubauer, Intestinal uptake of dipeptides and fl-lactam antibiotics: I. The intestinal uptake system for dipeptides and fl-lactam antibiotics is not part of a brush-border membrane peptidase, Biochim. Biophys. Acta 1030 (1990) 41-49.

[92] M. Hu and G.L. Amidon, Passive and carrier-mediated intestinal absorption components of captopril, J. Pharm. Sci. 77 (1988) 1007-1011.

16 G.M. Pauletti et al. / Journal of Controlled Release 41 (1996) 3 -17

[93] K. Miyazaki, K, Iseki and T. Arita, Intestinal absorption of several /3-1actam antibiotics. IV. Binding to the various components in the intestinal mucosa of rat and role in absorption process, J. Pharmacobiodyn. 5 (1982) 593-602.

[94] F.M. Ryan and EL. Smith, Effect of altering fluid transport on the absorption of benzylpenicillin and mannitol across rat intestinal mucosa in vitro, Pharm. Res. 6 (1989) S-88.

[95] W. Kramer, F. Girbig, U. Gutjahr, S. Kowalewski, F. Adam and W. Schiebler, Intestinal absorption of /3-1actam antibiotics and oligopeptides. Functional and stereospecific reconstitution of the oligopeptide transport system from rabbit small intestine, Eur. J. Biochem. 204 (1992) 923-930.

[96] I.J. Hidalgo, E Bhatnagar, C.P. Lee, J. Miller, G. Cucullino and EL. Smith, Structural requirements for interaction with the oligopeptide transporter in Caco-2 cells, Pharm. Res. 12 (1995) 317-319.

[97] I.J. Hidalgo, E Bhatnagar, C.P. Lee and P.L. Smith, Structur- al specificity for uptake via the di-/tripeptide transporter in Caco-2 cells, Pharm. Res. 10 (1993) S-181.

[98] J.M. Addison, D. Burston, J.A. Dalrymple, D.M. Matthews, J.W. Payne, M.H. Sleisenger and S. Wilkinson, A common mechanism for transport of di- and tripeptides by hamster jejunum in vitro, Clin. Sci. Mol. Med. 49 (1975) 313-322.

[99] J.M. Addison, D.M. Matthews and D. Burston, Evidence for active transport of the dipeptide carnosine (/3-alanyl-L-his- tidine) by hamster jejunum in vitro, Clin. Sci. Mol. Med. 46 (1974) 707-714.

[100] J.EF. Bai, E Subramanian, H.I. Mosberg and G.L. Amidon, Structural requirements for the intestinal mucosal cell peptide transporter: the need for N-terminal a-amino group, Phann. Res. 8 (1991) 593-599.

[101] C.H. Gochoco, F.M. Ryan, J. Miller, EL. Smith and I.J. Hidalgo, Uptake and transepithelial transport of the orally absorbed cephalosporin cephalexin, in the human intestinal cell line, Caco-2, Int. J, Pharm. 104 (1994) 187-202.

[102] E. Nakashima, A. Tsuji, H. Mizuno and T. Yamaha, Kinetics and mechanism of in vitro uptake of amino-/3- lactam antibiotics by rat small intestine and relation to the intact peptide transport system, Biochem. Pharmacol. 33 (1984) 3345-3352.

[103] D.I. Friedman and G.L. Amidon, Intestinal absorption mechanism of two prodrug ACE inhibitors in rat: enalapril maleate and fosinopril sodium, Pharm. Res. 6 (1989) 1043- 1047.

[104] R. Wootton and R. Hazelwood, Relative affinity of a series of charged dipeptides for the peptide carrier of rabbit intestinal brush-border membranes, Biochem. Soc. Trans. 17 (1989) 691-692.

[105] C. Wood, J. Miller, G. Wilson and I.J. Hidalgo, Effect of charge on affinity for the intestinal di-/tripeptide transporter, Pharm. Res. 9 (1992) S-254.

[106] N. Muranushi, T. Yoshikawa, M. Yoshida, T. Oguma, K. Hirano and H. Yamada, Transport characteristics of cef- tibuten, a new oral cephem, in rat intestinal brush-border membrane vesicles: relationship to oligopeptide and amino /3-1actam transport, Pharm. Res. 6 (1989) 308-312.

[107] J.R. Bronk, N. Lister and EA. Helliwell, Stereoselectivity of dipeptide transport in rat small intestine, in vitro, J. Physiol. 467 (1993) 189E

[108] E.E Eddy, G. Wilson and l.J. Hidalgo, Uptake of glycyl-L- proline in Caco-2 cells involves multiple transport mechanisms, Pharm. Res. 10 (1993) S-207.

[109] E.E Eddy, G. Wilson and 1.J. Hidalgo, Comparison of affinity of dipeptides and antibiotics for the di-/tripeptide transporter in Caco-2 cells, Pharm. Res. 10 (1993) S-208.

[110] H.W. Kleemann, H. Heitsch, R. Henning, W. Kramer, W. Kocher, W. Lerch, W. Rinz, W.U. Nickel, D. Ruppert, H. Urbach, R. Utz, A. Wagner, R. Weck and F. Wiegand, Renin inhibitory pentols showing improved enteral bioavailability, J. Med. Chem. 35 (1992) 559-567.

[111] I. Nogae, K. Kohno, J. Kikuchi, M. Kuwano, S. Akiyama, A. Kiue, K. Suzuki, Y. Yoshida, M.M. Kornwell, I. Pastan and M,M. Gottesman, Analysis of structural features of dihydropyridine analogues needed to reverse multidrug resistance and to inhibit photoaffinity labeling of P- glycoprotein, Biochem. Pharmacol. 38 (1989) 519-527.

[112] J.M. Zamora, H.L. Pearce and W.T. Beck, Physical-chemical properties shared by compounds that modulate multidrug resistance in human leukemic cells, Mol. Pharmacol. 33 (1988) 454-462.

[113] R.C. Sharma, S. Inoue, J. Roitelman, R.T. Schimke and R.D. Simoni, Peptide transport by multidrug resistance pump, J. Biol. Chem. 267 (1991) 5731-5734.

[114] A.Y. Chen, C. Yu, M. Potmesil, M.E. Wall, M.C. Wani and L.F. Liu, Camptothecin overcomes MDRl-mediated resistance in human KB carcinoma cells, Cancer Res. 51 (1991) 6039-6044.

[115] ES. Burton, R.A. Conradi and N.F.H. Ho, Evidence for a polarized efflux system for peptides in the apical region of Caco-2 cells, Biochem. Biophys. Res. Commun. 190 (1993) 760-766.

[116] ED. Gesellchen, C.J. Parli and R.C.A. Frederickson, A novel method for analysis of the in vivo stability of opioid peptides, in: D.H. Rich and E. Gross (Eds.), Peptides: Synthesis, Structure, Function, Pierce, Rockford, 1981, pp. 637-640.

[117] J.S. Morley, C.F. Hayward, R.J. Carter and S. Shuster, MPF analogue with high stability to proteolysis, Neuropeptides 2 (1981) 109-114.

[118] ED. Gesellchen and R.F, Santerre, Synthesis of peptides and proteins by chemical and biotechnological means, in: V.H.L. Lee (Ed.), Peptide and Protein Drug Delivery, Marcel Dekker, New York, 1991, pp. 57-136.

[119] L.F. Tseng, H.H. Lob and C.H. Li, [D-THR 2, THZ~] - enkephalinamide: a potent analgesic by subcutaneous and oral administration, Life Sci. 23 (1978) 2053-2056.

[120] D. RGmer, H.H. Biischer, R.C. Hill, J. Pless, W. Bauer, F. Cardinaux, A. Closse, D. Hauser and R. Huguenin, A synthetic enkephalin analogue with prolonged parenteral and oral analgesic activity, Nature 268 (1977) 547-549.

[121] S. Terada, T. Kato and N. Izumiya, Synthesis and hydrolysis by pepsin and trypsin of a cyclic hexapeptide containing lysine and phenylalanine, Eur. J. Biochem. 52 (1975) 273-282.

[122] F. AI-Obeidi, M.E. Hadley, B.M. Pettitt and V.J. Hruby, Design of a new class of superpotent cyclic a-melano- tropins based on quenched dynamic simulations, J. Am. Chem. Soc. 111 (1989) 3413-3416.


[123] R. Oliyai and V.J. Stella, Prodrugs of peptides and proteins for improved formulation and delivery, Annu. Rev. Phar- macol. Toxicol. 32 (1993) 521-544.

[124] H. Bundgaard, Prodrugs as a means to improve the delivery of peptide drugs, Adv. Drug Delivery Rev. 8 (1992) 1-38.

[125] H. Bundgaard and J. Moss, Prodrugs of peptides. 6. Bioreversible derivatives of thyrotropin-releasing hormone (TRH) with increased lipophilicity and resistance to cleavage by the TRH-specific serum enzyme, Pharm. Res. 7 (1990) 885-892.

[126] G.J. Rasmussen and H. Bundgaard, Prodrugs of peptides. 15. 4-imida-zolidinone prodrug derivatives of enkephalins to prevent aminopeptidase-catalyzed metabolism in plasma and absorptive mucosae, Int. J. Pharm. 76 (1991 ) 113-122.

[127] H. Bundgaard and G.J. Rasmussen, Prodrugs of peptides 11. Chemical and enzymatic hydrolysis kinetics of N- acyloxymethyl derivatives of a peptide like bond, Pharm. Res. 8 (1991) 1238-1242.

[128] S. Gangwar, G.M. Panletti, T.J. Siahaan, V.J. Stella and R.T. Borchardt, Synthesis of a novel, enzyme-sensitive cyclic prodrug of a model hexapeptide having enhanced membrane permeability and enzymatic stability. I. Acyloxy- alkoxycarbamate pro-moiety, 210th Annual Meeting of the American Chemical Society, Chicago, 20-24 August 1995.

[129] B. Wang, S. Gangwar, G.M. Pauletti, T.J. Siahaan and R.T. Borchardt, Synthesis of a novel, enzyme-sensitive cyclic prodrug of a model hexapeptide having enhanced membrane permeability and enzymatic stability. II. 3-(2'-hydroxy-4',6'-dimethylphenyl)-2,2-dimethyl propionic acid pro-moiety, 210th Annual Meeting of the American Chemi- cal Society, Chicago, 20-24 August 1995.

[130] G.M. Pauletti, S. Gangwar, B. Wang, T.J. Siahaan, D.G. Vander Velde and R.T. Borchardt, Novel prodrug approaches to prepare cyclic peptides with enhanced membrane permeability and enzymatic stability. II. 3-(2'-Hy- droxy-4',6'-dimethylphenyl)-2,2-dimethyl propionic acid pro-moiety, Pharm. Res. 12 (1995) S-208.

[ 131 [ S. Gangwar, G.M. Pauletti, T.J. Siahaan, D.G. Vander Velde, V.J. Stella and R.T. Borchardt, Novel prodrug approaches to prepare cyclic peptides with enhanced membrane permeability and enzymatic stability. I. Acyloxy-alkoxycarbamate pro-moiety, Pharm. Res. 12 (1995) S-208.

[132] B.A. Morgan and J.A. Gainor, Approaches to the discovery of non-peptide ligands for peptide receptors and peptidases, Annu. Rep. Med. Chem. 24 (1989) 243-252.

[133] A.F. Spatola, H. Saneii, J.V. Edwards, A.L. Bettag, M.K. Anwar, P. Rowell, B. Browne, R. Lahti and P. Von Voigtlander, Structure-activity relationships of enkephalins containing serially replaced thiomethylene amide bond surrogates, Life Sci. 38 (1986) 1243-1249.

[134] A.F. Spatola, Peptide backbone modifications: a structure- activity analysis of peptides containing amid bond surrogates, in: B. Weinstein (Ed.) Chemistry and Biochemistry of Amino Acids, Peptides and Proteins, Marcel Dekker, New York, 1983 p. 276-357.

[135] J.J. Shaw, L. Miller, M.J. Turubull, J.J. Gormley and J.S. Morley, Selective antagonists at the opiate delta-receptor, Life Sci. 31 (1982) 1259-1262.

[136] M.M. Hann, P.G. Sammes, P.D. Kennewell and J.B. Taylor, On double bond isosteres of the peptide bond; an enkephalin analogue, J. Chem. Soc., Chem. Commun. 5 (1980) 234-235.

[137] H.M. Jennewein, F. Waldeck and W. Konz, The absorption of tetragastrin from different sites in rats and dogs, Arz- neim. Forsch./Drug Res. 24 (1974) 1225-1228.

[138] K. Takaori, J. Burton and M. Donowitz, The transport of an intact oligopeptide across adult mammalian jejunum, Bio- chem. Biophys. Res. Commun. 137 (1986) 682-687.

[139] M. Saffran, C. Bedra, G.S. Kumar and D.C. Neckers, Vasopressin: a model for the study of effects of additives on the oral and rectal administration of peptide drugs, J. Pharm. Sci. 77 (1988) 33-38.

[140] S. Yokohama, K. Yamashita, H. Toguchi, J. Takeuchi and N. Kitamori, Absorption of thyrotropin-releasing hormone (TRH) after oral administration of TRH tartrate monohy- drate in the rat, dog and human, J. Pharmacobiodyn. 7 (1984) 101-111.

[141] E. Hayakawa, A. Yamamoto, Y. Shoji and V.H.L. Lee, Effect of sodium glycocholate and polyoxyethylene-9-1auryl ether on the hydrolysis of varying concentrations of insulin in the nasal homogenates of the albino rabbit, Life Sci. 45 (1989) 167-174.

[142] P.F. Augustijns and R.T. Borchardt, Transport and metabolism of Delta-Sleep-inducing Peptide (DSIP) in cultured human intestinal epithelial (Caco-2) cell monolayers, Drug Metab. Dispos. 23 (1995) 1372-1378.

[143] M.A. Ondetti and D.W. Cushman, Angiotensin-converting enzyme inhibitors: biochemical properties and biological actions, Crit. Rev. Biochem. 16 (1985) 381-411.

[144] M. Verstraete, Clinical application of inhibitors of fib- rinolysis, Drugs 29 (1985) 236-261.

Structural requirements for intestinal absorption of peptide drugs

Documents