Modern Pharmaceutics
`Third Edition, Revised and Expanded
`
`DRL EXHIBIT 1026 PAGE 1
`
`DRL EXHIBIT 1026 PAGE 1
`
`

`

`Library of Congress Cataloging-in-Publication Data
`
`Modern pharmaceutics / edited by Gilbert S. Banker, Christopher T.
`Rhodes.—3rd ed., rev. and expanded.
`p.
`cm.—(Drugsand the pharmaceutical sciences ; v. 72)
`Includes bibliographical references and index.
`ISBN 0-8247-9371-4 (alk. paper)
`1. Drugs—Dosage forms.
`2. Biopharmaceutics.
`3. Pharmacokinetics.
`4. Pharmaceutical industry——Quality control.
`I. Banker, Gilbert S.
`II. Rhodes, Christopher T.
`III. Series.
`RS200.M63
`1995
`615’.1—dc20
`
`95-33238
`CIP
`
`The publisher offers discounts on this book when ordered in bulk quantities. For more information, write
`to Special Sales/Professional Marketing at the address below.
`This book is printed on acid-free paper.
`Copyright © 1996 by Marcel Dekker, Inc. All Rights Reserved.
`Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic
`or mechanical, including photocopying, microfilming, and recording, or by any information storage and
`retrieval system, without permission in writing from the publisher.
`Marcel Dekker, Inc.
`270 Madison Avenue, New York, New York 10016
`
`Current printing (last digit)
`10
`9 8.7 6
`5
`
`“PRINTED IN THE UNITED STATES OF AMERICA
`
`DRL EXHIBIT 1026 PAGE 2
`
`DRL EXHIBIT 1026 PAGE 2
`
`

`

`
`
`Principles of Drug Absorption
`
`Michael Mayersohn
`College of Pharmacy, University of Arizona, Tucson, Arizona
`
`I.
`
`INTRODUCTION
`
`Drugs are most often’ introduced into the body by the oral route of administration. In fact, the
`vast majority of drug dosage forms are designed for oral ingestion, primarily for ease of admin-
`istration. It should be recognized, however, that this route may result in inefficient and erratic
`drug therapy. Whenever a drug is ingested orally (or by any nonvascular route), one would like
`to have rapid and complete absorption into the bloodstream for the following reasons:
`
`1.
`
`2.
`
`If we assume that there is somerelationship between drug concentration in the body
`and the magnitude of the therapeutic response (which is often the case), the greater the
`concentration achieved, the greater the response.
`In addition to desiring therapeutic concentrations, one would like to obtain these con-
`centrations rapidly. The more rapidly the drug is absorbed, the sooner the pharmaco-
`logical response is achieved.
`In general, one finds that the more rapid and complete the absorption, the more uniform
`and reproducible the pharmacological response becomes.
`4. The morerapidly the drug is absorbed, the less chance there is of drug degradation or
`interactions with other materials present in the gastrointestinal tract.
`
`3.
`
`In a broad sense, one can divide the primary factors that influence oral drug absorption and,
`thereby, govern the efficacy of drug therapy into the following categories: (a) physicochemical
`variables, (b) physiological variables, and (c) dosage form variables. For the most part, these
`variables will determine the clinical response to any drug administered by an extravascular
`route. Although often the total response to a drug given orally is a complex function of the
`aforementioned variables interacting together, the present discussion is limited to primarily the
`first two categories involving physicochemical and physiological factors. Dosage form variables
`influencing the response to a drug and the effect of route of administration are discussed in
`Chapters 4 and 5.
`
`21
`
`DRL EXHIBIT 1026 PAGE 3
`
`DRL EXHIBIT 1026 PAGE 3
`
`

`

`22
`
`Mayersohn
`
`Almost all drugs in current use and those under developmentare relatively simple organic
`molecules obtained from either natural sources or by synthetic methods. This statement is
`especially true of those drugs administered orally, the route emphasized in this chapter. How-
`ever, I would be derelict in not noting the virtual revolution in development of new therapeutic
`entities; those based on the incredible advances being made in the application of molecular
`biology and biotechnology. These new drugs, especially peptides and proteins, are not the
`small organic molecules stressed in this chapter. Indeed, those compounds have unique phys-
`icochemical properties that are quite different from those of small organic molecules, and they
`offer remarkable challenges for drug delivery. As a result, new and more complex physical
`delivery systems are being designed in conjunction with an examination of other, less tradi-
`tional, routes of administration (e.g., nasal, pulmonary, transdermal, or other). Because of issues
`of instability in the gastrointestinal tract and poor intrinsic membrane permeability,
`it now
`appears unlikely that these new biotechnology-derived drugs will employ the oral route for
`administration to any appreciable extent. Numerous strategies, however, are being explored,
`and there is evidence that some measure of gastrointestinal absorption can be achieved for
`some peptides [1].
`
`ll. ANATOMICAL AND PHYSIOLOGICAL CONSIDERATIONS OF THE
`GASTROINTESTINAL TRACT
`
`The gastrointestinal tract (GIT) is a highly specialized region of the body, the primary functions
`of which involve the processes of secretion, digestion, and absorption. Since all nutrients
`needed by the body, with the exception of oxygen, must first be ingested orally, processed by
`the GIT, and then made available for absorption into the bloodstream, the GIT represents an
`important barrier and interface with the environment. The primary defense mechanisms em-
`ployed by the gut to rid it of noxious or irritating materials are vomiting and diarrhea. In fact,
`emesis is often a first approach to the treatment of oral poisoning. Diarrheal conditions, initiated
`by either a pathological state or a physiological mechanism, will result in the flushing away
`of toxins or bacteria or will represent the responseto a stressful condition. Indeed, the GIT is
`often the first site of the body’s responseto stress, a fact readily appreciated by students taking
`a final examination. The nearly instinctive gut response to stress may be particularly pertinent
`to patients needing oral drug therapy. Since stress is a fact of our daily lives, and since any
`illness requiring drug therapy may, in some degree, be considered stressful, the implications
`of the body’s response to stress and the resulting influence on drug absorption from the gut
`may be extremely important.
`Figure 1 illustrates the gross functional regions of the GIT. The liver, gallbladder, and
`pancreas, although notpart of the gut, have been included, since these organs secrete materials
`vital to the digestive and certain absorptive functions of the gut. The lengths of various regions
`of the GIT are presented in Table 1. The small intestine, comprising the duodenum, jejunum,
`and ileum, represents greater than 60% of the length of the GIT, whichis consistent with its
`primary digestive and absorptive functions. In addition to daily food and fluid intake (about
`1—2 liters), the GIT and associated organs secrete about 8 liters of fluid per day. Of this total,
`between 100 and 200 ml of stool water is lost per day, indicating efficient absorption of water
`throughout the tract.
`
`A. Stomach
`
`After oral ingestion, materials are presented to the stomach, the primary functions of which
`are storage, mixing, and reducing all componentsto a slurry with the aid of gastric secretions;
`
`DRL EXHIBIT 1026 PAGE 4
`
`DRL EXHIBIT 1026 PAGE 4
`
`

`

`Principles of Drug Absorption
`
`23
`
`
`
`SALIVARY GLANDS ————_
`
`{Sj “¢@—--— PHARYNX
`
`—— ESOPHAGUS
`
`LIVER
`GALLBLADDER
`transverse colon
`ascending colon
`LARGE
`INTESTINE descending colon ———-
`cecum
`rectum
`
`
`
`;
`
`:
`
`’
`
`ai
`fos
`E
`
`STOMACH
`ir
`SFHt pyloric valve
`x
`PANCREAS
`Juodenuin
`SMALL
`jejunum
`INTESTINE
`ileum
`ikeocecal valve
`
`
`
`1 Diagrammatic sketch of the gastrointestinal tract (and subdivisions of the small and large intes-
`Fig.
`tines) along with associated organs. (Modified from Ref. 2.)
`
`and then emptying these contents in a controlled manner into the upper small intestine (duo-
`denum). All of these functions are accomplished by complex neural, muscular, and hormonal
`processes. Anatomically, the stomach has classically been divided into three parts: fundus,
`body, and antrum (or pyloric part), as illustrated in Fig. 2. Although there are no sharp dis-
`tinctions among these regions, the proximal stomach, made up of the fundus and body, serves
`as a reservoir for ingested material, and the distal region (antrum) is the major site of mixing
`motions and acts as a pump to accomplish gastric emptying. The fundus and body regions of
`the stomach haverelatively little tone in their muscular wall, as a result these regions can
`distend outward to accommodate a meal of up to 1 liter.
`A commonanatomical feature of the entire GIT is its four concentric layers. Beginning with
`the luminal surface, these are the mucosa, submucosa, muscularis mucosa, and serosa. The
`three outer layers are similar throughout most of the tract; however, the mucosa has distinctive
`structural and functional characteristics. The mucosal surface of the stomach is lined by an
`epithelial layer of columnarcells, the surface mucouscells. Along this surface are many tubular
`invaginations, referred to as gastric pits, at the bottom of which are found specialized secretory
`cells. These secretory cells form part of an extensive network of gastric glands that produce
`and secrete about 2 liters of gastric fluid daily. The epithelial cells of the gastric mucosa
`represent one of the most rapidly proliferating epithelial tissues, being shed by the normal
`stomach at the rate of about a half million cells per minute. As a result, the surface epithelial
`
`Table 1 Approximate Lengths of Various Regions of the Human Gastrointestinal Tract
`
`Region
`
`Duodenum
`Jejunum
`Tleum
`Large Intestine
`
`Length (m)
`
`0.3
`2.4
`3.6
`0.9-1.5
`
`DRL EXHIBIT 1026 PAGE 5
`
`DRL EXHIBIT 1026 PAGE 5
`
`

`

`24
`
`Mayersohn
`
`
`
`cardia
`
` esophagus
`
`pyloric antrum
`
`Fig. 2. Diagrammatic sketch of the stomach and anatomical regions. (Modified from Ref. 3.)
`
`layer is renewed every 1—3 days. Covering the epithelial cell surface is a layer of mucus 1.0—
`1.5 mm thick. This material, made up primarily of mucopolysaccharides, provides a protective
`lubricating coat for the cell lining.
`The next region, the muscularis mucosa, consists of an inner circular and an outer longi-
`tudinal layer of smooth muscle. This area is responsible for the muscular contractions of the
`stomach wall, which are needed to accommodate a mealby stretching, and for the mixing and
`propulsive movements of gastric contents. An area known as the lamina propria lies below the
`muscularis mucosa and contains a variety of tissue types, including connective and smooth
`muscles, nerve fibers, and the blood and lymphvessels. It is the blood flow to this region and
`to the muscularis mucosa that delivers nutrients to the gastric mucosa. The major vessels
`providing a vascular supply to the GIT are the celiac and the inferior and superior mesenteric
`arteries. Venous return from the GIT is through the splenic and the inferior and superior mes-
`enteric veins. The outermost region of the stomach wall provides structural support for the
`organ.
`
`B. Small Intestine
`
`The small intestine has the shape of a convoluted tube and represents the major length of the
`GIT. The small intestine, comprising the duodenum, jejunum, and ileum, has a unique surface
`structure, making it ideally suited for its primary role of digestion and absorption. The most
`important structural aspect of the small intestine is the means by which it greatly increasesits
`effective luminal surface area. The initial increase in surface area, compared with the area of
`a smooth cylinder, is due to the projection within the lumen of folds of mucosa, referred to as
`the folds of Kerckring. Lining the entire epithelial surface are fingerlike projections, the villi,
`extending into the lumen. These villi range in length from 0.5 to 1.5 mm,and it has been
`estimated that there are about 10—40 villi per square millimeter of mucosal surface. Projecting
`from the villi surface are fine structures, the microvilli (average length, 1 mm), which represent
`the final large increase in the surface area of the small intestine. There are approximately 600
`Microvilli protruding from each absorptive cell lining the villi. Relative to the surface of a
`smooth cylinder, the folds, villi, and microvilli increase the effective surface area by factors
`
`DRL EXHIBIT 1026 PAGE 6
`
`DRL EXHIBIT 1026 PAGE 6
`
`

`

`Principles of Drug Absorption
`
`25
`
`
`
`{A}
`
`(B)
`
`{C)
`
`(A) Photomicrograph of the human duodenal surface illustrating the projection ofvilli into the
`Fig. 3.
`lumen (magnification *75). The goblet cells appear as white dots on the villus surface. (B) Photomicro-
`graph of a single human duodenal villus illustrating surface coverage by microvilli and the presence of
`goblet cells (white areas; magnification 2400). (C) Photomicrographillustrating the microvilli of the
`small intestine of the dog (magnification 33,000). (From Ref. 4.)
`
`of 3, 30, and 600, respectively. These structural features are clearly indicated in the photomi-
`crographs shown in Fig. 3. A diagrammatic sketch of the villus is shown in Fig. 4.
`The mucosa of the small intestine can be divided into three distinct layers. The muscularis
`mucosa, the deepest layer, consists of a thin sheet of smooth muscle three to ten cells thick
`and separates the mucosa from the submucosa. The lamina propria, the section between the
`muscularis mucosa and the intestinal epithelia, represents the subepithelial connective tissue
`space and, together with the surface epithelium, formsthe villi structure. The lamina propria
`
`SMALL INTESTINAL LUMEN
`
`COLUMNAR (ABSORBING)
`
`LAMINA PROPRIA
`
`CENTRAL LACTEAL
`
`
`
`
`
`
`
`
`
`
`CAPILLARIES
`
`INTESTINAL
`GLAND (CRYPT)
`
`GOBLET CELLS
`
`
`Be
`3. MUSCULARIS MUCOSA
`
`
`
`
`LYMPH VESSEL
`
`Fig. 4 Diagrammatic sketch of the small intestine illustrating the projection of the villi into the lumen
`(left) and anatomic features of a single villus (right). (Modified from Ref. 5.)
`
`DRL EXHIBIT 1026 PAGE 7
`
`DRL EXHIBIT 1026 PAGE 7
`
`

`

`26
`
`Mayersohn
`
`contains a variety of cell types, including blood and lymph vessels and nerve fibers. Molecules
`to be absorbed must penetrate this region to gain access to the bloodstream.
`The third mucosal layer is that lining the entire length of the small intestine and represents
`a continuous sheet of epithelial cells. These epithelial cells are columnar, and the luminalcell
`membrane, upon which the microvilli reside, is called the apical cell membrane. Opposite this
`membrane is the basal plasma membrane, which is separated from the lamina propria by a
`basement membrane. A sketch of this cell is shown in Fig. 5. The primary function of the villi
`is absorption.
`The microvilli region has also been referred to as the striated border. It is in this region that
`the process of absorption is initiated. In close contact with the microvilli is a coating of fine
`filaments composed of weakly acidic, sulfated mucopolysaccharides. It has been suggested that
`this region may serve as a relatively impermeable barrier to substances within the gut, such as
`bacteria and other foreign materials. In addition to increasing the effective luminal surface area,
`the microvilli region appears to be an area of important biochemical activity.
`The surface epithelial cells of the small intestine are renewed rapidly and regularly. It takes
`about 2 days for the cells of the duodenum to be renewed completely. As a result of its rapid
`renewalrate, the intestinal epithelium is susceptible to various factors that may influence pro-
`liferation. Exposure of the intestine to ionizing radiation and cytotoxic drugs (such as folic
`acid antagonists and colchicine) reduce the cell renewalrate.
`
`Cc. Large intestine
`
`The large intestine, often referred,to as the colon, has two primary functions: the absorption
`of water and electrolytes, and the storage and elimination of fecal material. The large intestine,
`which has a greater diameter than the small intestine (ca., 6 cm), is connected to the latter at
`the ileocecal junction. The wall of the ileum at this point has a thickened muscular coat, called
`the ileocecal sphincter, which forms the ileocecal valve, the principal function of which is to
`prevent backflow of fecal material from the colon into the small intestine. From a functional
`point of view the large intestine may be divided into two parts. The proximal half, concerned
`
`INTESTINAL LUMEN
`
`a
`
`RIBOSOMES
`GOLGI MATERIAL
`
`a SMOOTH
`RETICULUM
`
`TIGHT JUNCTION——_“) MICROTUBULES+g .
`MICROVILLI
`MITOCHONDRIA= &
`RETICULUM
`LYSOSOMES
`A)va GRANULAR
`
`
`
`NUCLEUS
`
`.
`)BASAL MEMBRANE
`“~_ LAMINA
`PROPRIA
`
`a“
`
`INTERCELLULAR ——>
`SPACE
`BASEMENT
`
`
`
`*
`
`MEMBRANE
`
`Fig. 5 Diagrammatic sketch of the intestinal absorptive cell. (Modified from Ref. 6.)
`
`DRL EXHIBIT 1026 PAGE 8
`
`DRL EXHIBIT 1026 PAGE 8
`
`

`

`Principles of Drug Absorption
`
`27
`
`primarily with absorption, includes the cecum, ascending colon, and portions of the transverse
`colon. The distal half, concerned with storage and mass movement of fecal matter, includes
`part of the transverse and descending colon, the rectum, and anal regions, terminating at the
`internal anal sphincter (see Fig. 1).
`In humans,
`the large intestine usually receives about 500 ml of fluidlike food material
`(chyme) per day. As this material movesdistally through the large intestine, water is absorbed,
`producing a viscous and, finally, a solid mass of matter. Of the 500 ml normally reaching the
`large intestine, approximately 80 ml are eliminated from the gut as fecal material, indicating
`efficient water absorption.
`Structurally, the large intestine is similar to the small intestine, although the luminal surface
`epithelium of the former lacks villi. The muscularis mucosa, as in the small intestine, consists
`of inner circular and outer longitudinal
`layers. Figure 6 illustrates a photomicrograph and
`diagrammatic sketches of this region.
`
`
`
`OPENING OF
`INTESTINAL GLAND
`
`ABSORPTIVE CELL
`
`~ LAMINA PROPRIA
`* SMOOTH MUSCLE
`
`
`
`
`ARTERIOLE ET, Gon OTS Lt
`
`i
`
`LYMPHATIC
`ee
`:
`ie} NODULE
`
`(B)
`
`(C)
`
`(A) Scanning electron micrograph of the luminal surface of the large intestine (transverse colon;
`Fig. 6
`magnification 60). (From Ref. 7.) (B) Schematic diagram showing a longitudinal cross section of the
`large intestine. (C) Enlargement of cross section shown in (B). (B and C modified from. Ref. 8.)
`
`DRL EXHIBIT 1026 PAGE 9
`
`DRL EXHIBIT 1026 PAGE 9
`
`

`

`28
`
`Mayersohn
`
`D. Pathways of Drug Absorption
`Once a drug molecule is in solution, it has the potential to be absorbed. Whether or notit is
`in a form available for absorption depends on the physicochemical characteristics of the drug
`(i.e., its inherent absorbability) and the characteristics of its immediate environment(e.g., pH,
`the presence of interacting materials, and the local properties of the absorbing membrane). If
`there are no interfering substances present to impede absorption, the drug molecule must come
`in contact with the absorbing membrane. To accomplish this, the drug molecule must diffuse
`from the gastrointestinal fluids to the membrane surface. The most appropriate definition of
`drug absorption is the penetration of the drug across the intestinal “‘membrane’’ and its ap-
`pearance, unchanged in the blood draining the GIT. There are two important points to this
`definition: (a) It is often assumed that drug disappearance from the GI fluids represents ab-
`sorption. This is true only if disappearance from the gut
`represents appearance in the
`bloodstream. This is often not the situation; for example, if the drug degrades in GI fluids or
`if it is metabolized within the intestinal cells. (b) The term intestinal membraneis rather mis-
`leading, since this membraneis not a unicellular structure, but really a number of unicellular
`membranes parallel to one another. In fact, relative to the molecular size of most drug mole-
`cules, the compound must diffuse a considerable distance. Thus, for a drug molecule to reach
`the blood, it must penetrate the mucous layer and brush border covering the GI lumen, the
`apical cell surface, the fluids within this cell, the basal membrane, the basement membrane,
`the tissue region of the lamina propria, the external capillary membrane, the cytoplasma of the
`capillary cell, and finally, the inner capillary membrane. Therefore, when the expression intes-
`tinal membraneis used, we are discussing a barrier to absorption consisting of several distinct
`unicellular membranesand fluid regions bounded by these membranes. Throughout this chapter,
`intestinal membrane will be used in that sense.
`For a drug molecule to be absorbed from the GIT and gain access to the body(i.e., the
`systemic circulation) it must effectively penetrate all
`the regions of the intestine just cited.
`There are primarily three factors governing this absorption process once a drug is in solution:
`the physicochemical characteristics of the molecule, the properties and components of the GI
`fluids, and the nature of the absorbing membrane. Although penetration of the intestinal
`membraneis obviously the first part of absorption, we discuss the factors controlling penetra-
`tion extensively in the following section. At this point, assume that the drug molecule has
`penetrated most of the barriers in the intestine and has reached the lamina propria region. Once
`in this region, the drug mayeither diffuse through the blood capillary membrane and be carried
`away in the bloodstream, or penetrate the central lacteal and reach the lymph. These functional
`units of the villi are illustrated in Fig. 4. Most drugs, if not all, reach the systemic circulation
`by the bloodstream of the capillary network in the villi. The primary reason for this route
`being dominant over lymphatic penetration is that the villi are highly and rapidly perfused by
`the bloodstream. Blood flow rate to the GIT in humans is approximately 500—1000 times
`greater than lymph flow. Thus, although the lymphatic system is a potential route for drug
`absorption from the intestine, under normal circumstances, it will account for only a small
`fraction of the total amount absorbed. The major exception to this rule will be drugs (and
`environmental
`toxicants, such as insecticides) that have extremely large oil/water partition
`coefficients (on the order of 10,000). By increasing lymph flow or, alternatively, by reducing
`blood flow, drug absorption by the lymphatic system may become more important. The cap-
`illary and lymphatic vessels are rather permeable to most
`low-molecular-weight and lipid-
`soluble compounds. The capillary membrane represents a more substantial barrier than the
`central lacteal to the penetration of very large molecules or combinations of molecules, as a
`result of frequent separations of cells along the lacteal surface. This route of movement is
`
`DRL EXHIBIT 1026 PAGE 10
`
`DRL EXHIBIT 1026 PAGE 10
`
`

`

`Principles of Drug Absorption
`
`29
`
`important for the absorption of triglycerides,
`large (about 0.5 xm in the diameter).
`
`in the form of chylomicrons, which are rather
`
`Ili.
`
`PHYSICOCHEMICAL FACTORS GOVERNING DRUG ABSORPTION
`
`A. Ojil/Water Partition Coefficient and Chemical Structure
`
`As a result of extensive experimentation, it has been found that the primary physicochemical
`properties of a drug influencing its passive absorption into and across biological membranes
`are its oil/water partition coefficient (Ko,w), extent of ionization in biological fluids, determined
`by its pK, value and pH of the fluid in which it
`is dissolved, and its molecular weight or
`volume. That these variables govern drug absorption is a direct reflection of the nature of
`biological membranes. The cell surface of biological membranes (including those lining the
`entire GIT) is lipid; as a result, one may view penetration into the intestine as a competition
`for drug molecules between the aqueous environment on one hand, and the lipidlike materials
`of the membrane, on the other. To a large extent, then, the principles of solution chemistry and
`the molecular attractive forces to which the drug molecules are exposed will govern movement
`from an aqueous phase to the lipidlike phase of the membrane.
`At the turn of this century, Overton examined the osmotic behavior of frog sartorius muscle
`soaked in a buffer solution containing various dissolved organic compounds. He reasoned that,
`if the solute entered the tissue, the weight of the muscle would remain essentially unchanged;
`whereas, loss of weight would indicate an osmotic withdrawal of fluid and, hence, imperme-
`ability to the solute. He noted that, in general, the tissue was most readily penetrated by lipid-
`soluble compounds and poorly penetrated by lipid-insoluble substances. Overton was one of
`the first investigators to illustrate that compounds penetrate cells in the same relative order as
`their oil/water partition coefficients, suggesting the lipid nature of cell membranes. With animal
`or plant cells, other workers provided data in support of Overton’s observations. The only
`exception to this general rule was the observation that very small molecules penetrate cell
`membranes faster than would be expected based on their Ko,w values. To explain the rapid
`penetration of these small molecules (e.g., urea, methanol, formamide), it was suggested that
`cell membranes, although lipid, were not continuous, but were interrupted by small water-filled
`channels or ‘‘pores’’; such membranesare best described as being lipid-sieve membranes. As
`a result, one could imagine lipid-soluble molecules readily penetrating the lipid regions of the
`membrane while small water-soluble molecules pass through the aqueous pores. Fordtran et
`al. [9] estimated the effective pore radius to be 7—8.5 and 3~—3.8 A in human jejunum and
`ileum, respectively. There may be a continuous distribution of pore sizes; a smaller fraction of
`larger ones and a greater fraction of smaller pores.
`Our knowledge of biological membrane ultrastructure has increased considerably over the
`years as a result of rapid advances in instrumentation. Although there is still controversy over
`the most correct biological membrane model, the concept of membranestructure presented by
`Davson and Danielli, as a lipid bilayer is perhaps the one best accepted [10,11]. The most
`current version of that basic modelis illustrated in Fig. 7 and is referred to as the fluid mosaic
`model of membrane structure. That model is consistent with what we have learned about the
`existence of specific ion channels and receptors within and along surface membranes.
`Table 2 summarizes someliterature data supporting the general dependence of the rate of
`intestinal absorption on Ko,y, as measured in the rat [13,14]. As with other examples that are
`available, as Ko,w increases, the rate of absorption increases. One very extensive study [15—
`
`DRL EXHIBIT 1026 PAGE 11
`
`DRL EXHIBIT 1026 PAGE 11
`
`

`

`30
`
`Mayersohn
`
`CARBOHYDRATES BOUND
`
`TO LIPIDS AND TO PROTENNS
`
`SOS
`
`Ff
`[ese
`Steeer
`PASE
`
`TRANSMEMBRANE
`PROTEIN
`
`PERIPHERAL
`PROTEIN
`
`Fig. 7 Diagrammatic representation of the fluid mosaic model of the cell membrane. The basic structure
`of the membrane is that of a lipid bilayer in which the lipid portion (long tails) points inward and the
`polar portion (round head) points outward. The membrane is penetrated by transmembrane (orintegral)
`proteins. Attached to the surface of the membrane are peripheral proteins (inner surface) and carbohy-
`drates that bind to lipid and protein molecules (outer surface). (Modified from Ref. 12.)
`
`17] has examined in depth the physicochemical factors governing nonelectrolyte permeability
`for several hundred compounds. This study employed an in vitro rabbit gallbladder preparation,
`the mucosal surface of which is lined by epithelial cells. The method used to assess solute
`permeability is based on measurement of differences in electrical potential (streaming poten-
`tials) across the membrane. The more permeable the compound, the smaller the osmotic pres-
`
`
`
`Table 2 Influence of Oil/Water Partition Coefficient (Ko»w) on Absorption from the Rat Intestine
`
`Percentage
`Compound
`Kony
`absorbed
`
`
`Olive oil/water
`Valeramide
`Lactamide
`Malonamide
`
`Chloroform/water
`Hexethal
`Secobarbital
`Pentobarbital
`Cyclobarbital
`Butethal
`Allybarbituric Acid
`Phenobarbital
`Aprobarbital
`Barbital
`
`0.023
`0.00058
`0.00008
`
`85
`67
`27
`
`44
`> 100
`40
`50.7
`30
`28.0
`24
`13.9
`24
`11.7
`23
`10.5
`20
`4.8
`17
`4.9
`
`0.7 12
`
`DRL EXHIBIT 1026 PAGE 12
`
`DRL EXHIBIT 1026 PAGE 12
`
`

`

`Principles of Drug Absorption
`
`31
`
`sure it exerts, and the smaller the osmotic fluid flow it produces in the opposite direction; this
`results in a small potential difference. If the compound is impermeable, it produces a large
`osmotic pressure and osmotic fluid flow, resulting in a large potential difference. Experimen-
`tally, one exposes the mucosal membrane surface to a buffer solution containing a reference
`compound to which the membrane is completely impermeable and measures the resulting
`potential difference. This is followed by exposing the same membraneto a solution ofatest
`compound and again measuring the resulting potential difference. The ratio of the potential
`difference of the test compoundto that of the reference compoundis referred to as the reflection
`coefficient (o). The reflection coefficient is a measure of the permeability of the test compound
`relative to a reference solute with the particular membrane being used. The less permeable the
`test compound,the closer the reflection coefficient approaches 1; the more permeable thetest
`compound,the closer the coefficient approaches zero.
`By using this method, Wright and Diamond were able to reach a number of important
`conclusions concerning patterns of nonelectrolyte permeability. In general, membrane perme-
`ability of a solute increases with Ko,w, supporting previous findings mentionedearlier. The two
`classes of exceptions to this pattern are (a) highly branched compounds, which penetrate the
`membrane more slowly than would be expected based on their Ko,y; and (b) smaller polar
`molecules, which penetrate the membrane more readily than would be expected based ontheir
`Kon. The latter observation has been noted by other workers and, as mentionedearlier, it has
`resulted in the developmentof the lipid-sieve membrane concept, whereby one envisions aque-
`ous ‘pores in the membrane surface. The authors postulate that these small, polar, relatively
`lipid-insoluble compounds penetrate the membrane by following a route lined by the polar
`groupings of membraneconstituents(i.e., localized polar regions). This concept is an attractive
`structural explanation of what have been referred to as pores. The accessibility of this route
`would be limited primarily by the molecular size of the compound asa result of steric hin-
`drance. In fact, it is the first one or two members of a homologous series of compoundsthat
`are readily permeable, but beyond these members, it is primarily Ko,w that dictates permeability.
`Table 3 illustrates this effect for several members of various homologous series. Recall that
`
`Influence of Chain Length on Membrane
`Table 3
`Permeability Within Several Homologous Series"
`
`Compound
`
`Urea
`Methyl urea
`Ethyl urea
`Propyl urea
`Butyl urea
`
`Malononitrile
`Succinonitrile
`Glutaronitrile
`
`Reflection coefficient, o
`
`0.29
`0.54
`0.92
`0.93.
`0.70
`
`*
`|
`|
`-
`J
`
`f
`0.09
`0.30 —
`0.21
`4
`
`f
`0.28
`Methylformamide
`—
`0.51
`Methylacetamide
`|
`0.22
`Methylproprionamide
`“The reflection coefficient o is defined in the text. The di-
`rection of the arrows indicates an increase in permeability
`from the least permeable memberofthe series.
`
`DRL EXHIBIT 1026 PAGE 13
`
`DRL EXHIBIT 1026 PAGE 13
`
`

`

`32
`
`Mayersohn
`
`the smaller the o, the more permeable the compound. In each instance, permeability decreases
`after the first member, reaches a minimum, and then increases again.
`The other anomalous behavior was the smaller-than-expected permeability of highly
`branched compounds. This deviation has been explained on the basis that membranelipids are
`subject to a more highly constrained orientation (probably a parallel configuration of hydro-
`carbon chains of fatty acids) than are those in a bulk lipid solvent. As a result, branched
`compounds must disrupt this local lipid structure of the membrane and will encounter greater
`steric hindrance than will a straight-chain molecule. This effect with branched compoundsis
`not adequately reflected in simple aqueous—lipid partitioning studies(i.c., in the Ko,w value).
`With the exception of rather small polar molecules, most compounds, including drugs, ap-
`pear to penetrate biological membranesbya lipid route. As a result, the membrane permeability
`of most compounds is