Aliment Pharmacol Ther 2000; 14: 138371401.
`
`Review article: the control of gastric acid and Helicobacter pylori
`
`eradication
`
`G. SACHS*, ]. M. SHIN*, K. MUNSON*, O. VAGIN*, N. LAMBRECHT*, D. R. SCOTT*,
`D. L. WEEKS" & K. MELCHERST
`*UCLA Digestive Research Center, Departments of Physiology and Medicine, University of California, Los Angeles, CA, USA;
`and TByk Gulden, Konstanz, Germany
`
`Accepted for publication 5 June 2000
`
`
`
`SUMMARY
`
`ureI, E, F, G and H are explored in order to explain the
`unique location of
`this pathogen. The dominant
`This review focuses on the gastric acid pump as a
`requirement for acid resistance is the presence of a
`therapeutic target for the control of acid secretion in
`proton gated urea transporter, UreI, which increases
`peptic ulcer and gastro-oesophageal reflux disease. The
`access of gastric juice urea to the intrabacterial urease
`mechanism of the proton pump inhibitors is discussed as
`300-fold. This enables rapid and continuous buffering of
`well as their clinical use. The biology of Helicobacter pylori
`the bacterial periplasm to sz 6.0, allowing acid
`as a gastric denizen is then discussed, with special regard
`resistance and growth at acidic pH in the presence of
`to its mechanisms of acid resistance. Here the properties
`1 mM urea. A hypothesis for the basis of combination
`of the products of the urease gene clusters, ureA, B and
`therapy for eradication is also presented.
`
`INTRODUCTION
`
`In the last 25 years of the 20th century, a series of
`revolutions occurred in our understanding of
`the
`biology of gastric acid secretion and of acid related
`disease, particularly with respect to the role of Helicob—
`acter pylori. Whereas ‘no acid, no ulcer’ remains as true
`as it were in 1910, a new aphorism is required to state
`our current therapeutic strategy. At the beginning of
`the 21st century, the treatment and outcome of acid
`related disease have changed radically. Peculiarly, these
`changes, although remarkable, have passed largely
`unheralded by the general public. On the other hand,
`we have become much more aware of the long-term
`consequences of gastro-oesophageal
`reflux disease
`
`Correspondence to: Dr G. Sachs, VA Greater Las Angeles Heath Care
`System, West LA, Building 113 Room 324, Los Angeles, CA 90073,
`USA.
`
`E—mail: gsachs@acla.eda
`
`© 2000 Blackwell Science Ltd
`
`(GERD) and of H. pylori infection, consequences that
`will again change our method of treatment.
`The introduction of histamine-2 receptor antagonists
`allowed, for the first time, successful, relatively routine,
`medical management of peptic ulcer disease.1 However,
`it
`then became clear that duodenal ulcer treatment
`
`required maintenance therapy with these drugs, raising
`the question as to the reasons for recurrence.2 Addi-
`tionally, the effectiveness of these antagonists against
`not only histamine—, but also gastrin—stimulated secre—
`tion, placed the release of histamine as the pivotal event
`in the regulation of acid release from the parietal cell.1
`This, in turn, led to the recognition of the enterochrom-
`affin-like cell rather than the mast cell, as the master
`
`neuro-endocrine cell regulating parietal cell function.3
`After their introduction, it was realized that the clinical
`
`effectiveness of these receptor antagonist drugs was
`somewhat limited, even at high doses. Whereas acute
`treatment and maintenance of duodenal ulcer disease
`
`was satisfactory, certainly compared to before their
`
`1383
`
`ATTORNEY CONFIDENTIAL
`
`AZV00293734
`
`Page 1
`
`Dr. Reddy's Exh. 1047
`
`Page 1 Dr. Reddy's Exh. 1047
`
`

`

`1384
`
`G. SACHS et (11.
`
`introduction, the treatment of gastro-oesophageal ero-
`sion was less so. However, at the same time as the first
`
`of these receptor antagonists was introduced, the acid
`pump of the stomach (the gastric H, K ATPase) was
`described and its potential as a better therapeutic target
`was recognizedf‘f—6
`The drugs that were introduced as an improvement in
`therapy, particularly of GERD, were targeted at the acid
`pump. Now, truly reliable treatment of all types of acid
`related diseases became available in the form of proton
`pump inhibitors.7 However, the problem of recurrence
`of peptic ulcers was still present in spite of superior
`inhibition of acid secretion during initial therapy.
`At the same time that the first proton pump inhibitor
`(omeprazole) was patented (1979), a pathologist
`in
`Western Australia, Robyn Warren, was intrigued by the
`observation that bacteria were present at the sites of
`ulcer disease. He managed to convince a young
`resident, Barry Marshall,
`that this might provide an
`explanation for ulcer
`recurrence and the door
`to
`Campylobacter now Helicobactcr pylori was opened.8
`Today it is recognized that infection by this organism
`is
`the major culprit
`in peptic ulcer disease and
`eradication is now commonly performed by a combi-
`nation of a proton pump inhibitor and amoxycillin and
`clarithromycin.9 This review will discuss the last two of
`these advances, the proton pump inhibitors and their
`mechanisms and profiles and the peculiarities of an
`organism that has chosen the stomach as its favoured
`habitat.
`
`THERAPEUTIC ASPECTS OF THE GASTRIC
`
`H, K ATPASE, THE GASTRIC ACID PUMP
`
`Ion transport cycle of the acid pump
`
`The gastric H, K ATPase is an ion-motive ATPase,
`belonging to a family of
`these, called the P type
`ATPases. These ATPases transport ions as a function
`of a cycle of phosphorylation and dephosphorylation of
`the transport protein.10 A subfamily of these P type
`ATPases are the P2 ATPases, which transport small
`cations such as Na+, K+, Ca2+, Mg2+ and H+. In this P2
`group are the closely related Na, K and H, K ATPases
`distinguished by the presence of two, rather than one,
`sub-units, a catalytic,
`ion transporting sub-unit (the
`alpha sub-unit), and a stabilizing sub-unit (the beta sub-
`unit).11 The Na, K and H, K ATPase are about 65%
`homologous. However, digoxin is absolutely selective
`
`the Na, K ATPase whereas the proton pump
`for
`inhibitors, owing to their chemical properties, are
`absolutely selective for the gastric H, K ATPase. These
`ATPases have six membrane inserted segments which
`are regarded as their core structure containing the ion
`transport pathway and either
`two such segments
`preceding the core structure to give eight membrane
`segments in the transition metal pumps (P1 ATPases) or
`four segments following the core structure to provide 10
`segments in the P2 ATPases. The beta sub-unit has a
`single trans-membrane segment.
`The gastric pump is present in high concentrations in
`the parietal cell and at much lower concentrations in
`the collecting duct of the kidney. Whereas maximal
`acidity in the parietal cell canaliculus is < 1.0, that in
`the kidney is never
`less
`than 4.0. Whatever
`the
`pathway of stimulation of the parietal cell, the common
`event is recruitment of the pump, present in cytoplasmic
`tubules into the membrane of the secretory canaliculus,
`to form microvilli lining this space, and activation of a
`KCl pathway to allow K+ to access the external surface
`of the ATPase and secretion of Cl‘.
`
`The coupling between the cycle of phosphorylation
`and dephosphorylation and ion transport
`is best
`described as due to a sequential set of conformational
`changes. In the absence of MgATP the pump is in a
`conformation able to bind MgATP and the outward ion
`
`the E1 conformation. With the binding of
`(H30+),
`MgATP and the ion, the protein is phosphorylated in
`a sequence common to all the P type ATPases and the
`ion moves into the membrane domain of the catalytic
`sub-unit
`(El-P). Within the membrane domain,
`the
`transport ion is enclosed so that it loses communication
`with the cytoplasmic side and has not yet opened to the
`outside face,
`the occluded conformation. The outside
`
`face now opens forming the EZ-P conformation. At this
`stage the outward ion is extruded and the inward ion
`binds. With this,
`the pump dephosphorylates
`(E2
`conformation),
`followed by the inward ion, become
`occluded. From this state, the ion-binding site opens to
`the interior and the inward ion is released as MgATP
`rebinds. Figure 1 illustrates these steps for the gastric
`H, K ATPase.
`
`The pump is thought to pump hydronium ions (H30+)
`rather than protons, since at high pH values it is able to
`
`pump Na+. It is able to release H30+ at a concentration
`of 160 mM, therefore generating an external pH of 0.8.
`This corresponds to a 4 million-fold gradient with
`respect to the inside of the parietal cell. It is able to
`
`© 2000 Blackwell Science Ltd, Aliment Pharmacol Ther 14, 138371401
`
`ATTORNEY CONFIDENTIAL
`
`AZV00293735
`
`Page 2
`
`Dr. Reddy‘s Exh. 1047
`
`Page 2 Dr. Reddy's Exh. 1047
`
`

`

`
`
`REVIEW: GASTRIC ACID AND H. PYLORI ERADICATION
`
`1385
`
`
`cytoplasmic
`Hg?
`
`5‘
`
` & -1-
`
`orientation from facing inward to facing outward. The
`cytoplasmic domain in the ATP and inward ion binding
`conformation (E1) is a relatively open structure with at
`least
`two large separate lobes, corresponding to the
`loops between M2 and M3 and M4 and M5. The latter
`contains the phosphorylation site and MgATP binding
`region. In the ion site facing outward (Ez-P) conforma-
`tion, the cytoplasmic domain is more compact and the
`separation between the lobes less evident.
`The stalk contains the extensions of the membrane
`
`segments into the cytoplasmic region and forms a link
`for
`the passage of
`ions into the membrane. The
`membrane domain of the catalytic sub-unit of the H, K
`ATPase contains 10 inserted segments. The arrange-
`ment of these segments and the relative mobility of
`these is vital for the transport of the ions. The beta sub-
`unit has a single transmembrane segment with most of
`the protein on the outside surface of the pump, closely
`associated with the beginning of M8 in the catalytic
`sub-unit. It is N-glycosylated at seven sites. Its function
`appears to be stabilization of the membrane region of
`the alpha sub—unit, perhaps to lock TMS in place so that
`
`K+ can bind for inward transport. The external domain
`contains the loops connecting the membrane segments
`of the catalytic sub-unit and most of the beta sub-unit,
`with six or seven N-linked glycosylation sites. The
`largest loop between the membrane segments of the
`alpha sub-unit is between M7 and M8.
`Different biochemical and mutagenesis experiments
`have provided evidence that there is a close association
`between M4 and M6, M5 and M7 and M6 and M9.
`
`Recent data on 2D and three dimensional crystals of the
`sarcoplasmic reticulum (sr) Ca2+ ATPase have allowed
`placement of the membrane segments relative to each
`other and soon we should be able to place the side-
`chains of the membrane-inserted segments in their
`appropriate positions
`to reveal
`the ion pathways
`through the membrane. Prior to crystallization a series
`of site directed mutations have been made, which
`
`implicate M4, M5 and M6 as being intimately involved
`
`in the pathway transporting H30+ outward and K+
`inward.
`
`Activation of the H, K ATPase
`
`In the absence of stimulation, the vast majority of the
`pump molecules present
`in the parietal cell are in
`cytoplasmic membrane structures called tubulovesicles.
`Upon stimulation,
`these tubulovesicles transform into
`
`
`
`Mg.E1-P 1330*
`I
`Mg.E1—P[ 335+]
`I
`Mg-Ez-P
`
`extracytoplasmic
`
`350
`
`Figure 1. The reaction cycle of the gastric H+, KJr ATPase
`showing the E1 form of the enzyme that is phosphorylated by
`ATP as the hydronium ion, H30+, is bound. This form converts to
`EZ—P as the hydronium ion is released after passing through the
`trapped or occluded form. With binding of K+, the cycle progresses
`by dephosphorylation, occlusion and release of KJr to the
`cytoplasm from the E1 form of the enzyme.
`
`reabsorb K+, generating an inward gradient of about
`tenfold. As such, this is the most powerful pump known
`in mammalian systems and generates an acidic space
`within the secretory canaliculus of the parietal cell at
`least a thousand-fold greater than anywhere else in the
`
`body. In the absence of K+ on its external surface,
`the pump locks into the Ez-P conformation and does not
`turnover. Essential for activation of acid secretion is the
`
`presentation of K+ to the outside surface of the pump.
`This occurs as the pump moves into the microvilli of the
`
`secretory canaliculus due to activation of a K+Cl_
`pathway.
`
`Structure of the ATPase
`
`The three-dimensional structure of these pumps is
`becoming clearer so that these conformational changes
`can be described in molecular terms. There are four
`
`the catalytic sub-unit: a large
`distinct domains of
`cytoplasmic domain, a stalk domain which connects
`the cytoplasmic to the membrane domain, and an
`extracytoplasmic domain. The cytoplasmic domain
`undergoes large conformational changes as the ion is
`transported outwards, changing from a loose structure
`to a more compact one as the ion moves to the external
`surface (E1 to E2 transition). The ion passes through the
`stalk and into the membrane domain, which also
`
`changes conformation as the ion binding site changes
`
`© 2000 Blackwell Science Ltd, Aliment Pharmacol Ther 14, 138371401
`
`ATTORNEY CONFIDENTIAL
`
`AZV00293736
`
`Page 3
`
`Dr. Reddy‘s Exh. 1047
`
`Page 3 Dr. Reddy's Exh. 1047
`
`

`

`1386
`
`G. SACHS et al.
`
`the microvilli of an infolding of the parietal cell, termed
`the secretory canaliculus. Along with this transforma-
`tion, the canalicular membrane acquires the property of
`a KCl conductance, thus supplying K+ to the outside
`surface of the enzyme. This enables the enzyme to cycle,
`
`leaving the Cl‘ to
`exporting H+ and reabsorbing K+,
`accompany the secreted H+. Various proteins, such as
`ezrin and Rab, generally involved in membrane recyc-
`ling, have been identified in this activation mechanism.
`In humans, there is continuous recycling of the pump
`between the tubulovesicles and the secretory canalicu-
`lus, to give a basal acid secretion. Upon stimulation by
`food, there is a rapid recruitment of the pump into the
`canalicular membrane.
`
`confirm the presence of 10 membrane segments in the
`catalytic sub-unit of the P2 type ATPases and show
`the likely presence of a vestibule at the luminal face of
`the enzymes. They both show a large cytoplasmic
`sector, a stalk and a membrane sector with a small
`external
`sector. The advent of
`three dimensional
`
`crystals of the sr Ca2+ ATPase at 3.7 A resolution
`(Toyoshima, C) will do much to advance the field. The
`illustrates
`the
`model
`shown in Figure 2
`different
`associations by other methods that have been described
`in the Na, K, Ca and H, K ATPases, where M1, M2 and
`
`M3, M4 are juxtaposed and M4, M5, M6 and M8 are
`close to the transport pathway of the pump and M9,
`M10 and the beta sub-unit are peripheral
`to the
`transport sector.
`
`Towards a three-dimensional map of the H, K ATPase
`
`Recently, three dimensional reconstructions have been
`published of the H+ ATPase of Neurospora and the Ca2+
`ATPase of
`the sarcoplasmic reticulum.12‘ 13 These
`
`Putative transport mechanism
`
`The rate of ion transport by a P2 type ATPase is about
`104 times slower than through a channel and 102 times
`
`
`
`ADP
`/
` Mg+Pi
`MgATP
`
`.\
`/'
`MgEfl‘ 44» mag?
`
`CYTOPLASMIC
`DOMAIN
`
`
`
`
`
`
`
`stalk
`
`
`MEMBRANE
`330MA] N
`
`Figure 2. The probable arrangement of the membrane segments of the alpha and beta sub—units of the H+, KJr ATPase with interactions
`of M8 and beta residue regions numbered and M6 to M9 and M5 to M7 numbered at the end of M8 in the figure. The transport region of
`the membrane domain is considered to involve particularly M5 and M6 along with M4 and perhaps MS, as illustrated schematically. The
`cytoplasmic domain is visualized with a groove—admitting ATP and ion. The groove closes as the enzyme goes from E1 to E2. The majority
`of the beta sub—unit is on the outside surface and in the case of hog enzyme has 6 N—linked glycosylation sites. A vestibule is present
`bounded by M4, M5, M5—M6 loop M6 and M7, M8.
`
`© 2000 Blackwell Science Ltd, Aliment Pharmacol Ther 14, 138371401
`
`ATTORNEY CONFIDENTIAL
`
`AZV00293737
`
`Page 4
`
`Dr. Reddy‘s Exh. 1047
`
`Page 4 Dr. Reddy's Exh. 1047
`
`

`

`REVIEW: GASTRIC ACID AND H. PYLORI ERADICATION
`
`1387
`
`slower than through a carrier. This is due to the larger
`conformational changes necessary in a pump, com-
`pared to these other types of
`transporters. Several
`approaches have shown that these ATPases undergo
`conformational changes as a function of ion binding.
`For example,
`the fluorescence of FITC labelled H, K
`
`ATPase is rapidly quenched with the addition of K+ as is
`the Na, K ATPase.14‘ 15 The rate of exchange of tritium
`or deuterium is markedly altered as the E1 conformation
`changes to E2.16‘ 17 Electron diffraction images of the
`pumps in the E1 or E2 state show large changes in shape
`of the cytoplasmic domain.18‘ 19
`The location of ion transport is through the membrane
`domain. In the case of the Na, K, Ca and H, K ATPases,
`
`the ion transport pathway becomes occluded during the
`transport cycle, that is to say the ions are trapped within
`the membrane domain as the pump transits from E1 to
`E2 and vice versa, as if a gate on either side of the ion is
`closed during ion movement across the membrane
`domain. Site directed mutagenesis has identified several
`residues within the membrane domain as relevant to
`
`In particular, hydrophilic
`occlusion and transport.
`amino acids such as aspartic acid, glutamic acid, serine
`and threonine in M4, M5, M6 and perhaps M8 have
`been identified by this approach as related to ion flux
`through the membrane. M5 and M6 have also been
`identified as probably mobile within the membrane
`domain, as a function of
`loss of K binding after
`trypsinization.20‘ 21
`The crystal structure of other ATPases of the same
`family suggests the presence of a luminal vestibule in
`the pump. By analogy to the structure of the K channel
`of bacteria that has been resolved, the vestibule contains
`
`binding sites for K” that remove the water of hydration
`from the ion as it moves through an inverted filter
`funnel and then moves through the transport site as the
`naked ion. In the K channel, this was recognized even
`before crystallization since hydrophobic amines blocked
`K movement. In the case of the H, K ATPase a series of
`
`hydrophobic amines, such as imidazopyridines, quinaz-
`
`olines or aza-indoles are all K+ competitive inhibitors
`binding at
`the luminal
`surface.22_241 Site directed
`mutations have identified residues
`in M3/M4 and
`
`M5/M6 as involved in binding an imidazo-pyridine,
`SCH2 8080. These may then be considered to form part
`of the vestibule admitting K+.25‘ 26
`If so, then the region containing M5 and M6 and the
`connecting loop between these membrane segments is a
`
`luminal vestibule of the pump. An example of this class
`of drug is the proton pump inhibitors.
`
`THE PROTON PUMP INHIBITORS
`
`Mechanism of the proton pump inhibitors
`
`All the proton pump inhibitors have a similar core
`structure, a substituted pyridyl methylsulfinyl benzim-
`idazole. The pyridine moiety gives them the property of
`being a protonatable weak base with pKa in the range of
`4075.0. In the un-protonated form they are membrane
`permeable prodrugs, so that they will concentrate in
`spaces more acidic than pH 4.0. In people, the only
`space where the pH can go lower than 4.0 is the
`canaliculus of the active gastric parietal cell. They are
`
`prodrugs since these compounds are subject to a H+
`catalysed conversion where they undergo a rearrange-
`ment to sulfenic acids and thence to sulfenamides.27
`
`The latter are the product in acidic solutions but the
`former may be responsible for inhibition of the ATPase
`under acid secreting conditions. As a class, these drugs
`are therefore acid-unstable, requiring protection against
`gastric acidity for oral
`formulation and requiring
`reconstitution for i.v. administration. The four com-
`
`in terms of acid stability.
`pounds available differ
`Pantoprazole is the most stable, followed by omeprazole
`and lansoprazole and then rabeprazole. The first three
`have a pKa of z4.0 whereas the pKa of rabeprazole is
`close to 5.0. The higher pKa of rabeprazole accounts for
`its relative acid instability but
`this drug will also
`concentrate 10-fold relative to the other compounds,
`hence it will concentrate 10-fold in an acidic space of
`pH 4.0, compared to no accumulation for the other
`drugs. At
`the pH of a fully activated parietal cell
`canaliculus, accumulation is theoretically 1000-fold
`for omeprazole,
`lansoprazole and pantoprazole and
`10 000-fold for rabeprazole.
`The acid space accumulation is a key feature of these
`drugs, giving them a large therapeutic index, since they
`are highly concentrated at their target, the site of acid
`secretion. Following this, their conversion to a reactive
`sulfenic acid or sulfenamide (which form disulphides
`with one or more of the accessible cysteines in the
`pump) provides an ability to covalently inhibit
`the
`gastric ATPase, providing acid inhibition of
`long
`duration even after the drug has disappeared from the
`blood. Both the sulfenic acid and the sulfenamide are
`
`target
`
`for hydrophobic drugs able to penetrate the
`
`permanent
`
`cations,
`
`and
`
`are
`
`therefore
`
`relatively
`
`© 2000 Blackwell Science Ltd, Aliment Pharmacol Ther 14, 138371401
`
`ATTORNEY CONFIDENTIAL
`
`AZV00293738
`
`Page 5
`
`Dr. Reddy‘s Exh. 1047
`
`Page 5 Dr. Reddy's Exh. 1047
`
`

`

`1388
`
`G. SACHS et a1.
`
`CI—I-
`CH_ ° CH_
`ON
`
`so
`omeprazole
`H
`N/<N
`
`CH
`
`protonation
`
`sulfenic acid sulfenamide
`
`R.
`
`+
`
`H
`
`
`
`3095::CH_
`
`(
`
`)
`
`Iansoprazole
`N/<
`
`D
`H
`
`©8
`
`“: g:
`
`
`
`\
`
`"AIME—SATPase
`
`Figure 3. The chemistry of proton pump inhibitors. The four currently available drugs are shown on the left. The protonation step
`results in selective accumulation in the secretory canaliculus of the parietal cell. In acid, there is an acid catalysed conversion to the
`sulfenic acid and thence to the sulfenamide. Either of these can inhibit the H, K ATPase although it appears more likely that the sulfenic
`acid is the primary inhibitor.
`
`membrane impermeant, remaining trapped at their site
`of formation or being secreted into the gastric juice.
`Their mechanism of action is illustrated in two figures,
`mechanism
`one
`detailing
`the
`putative
`chemical
`(Figure 3), the other superimposing their mechanism
`onto the illustration of the pump to define their target
`pictorially (Figure 4).
`
`Clinical effects of proton pump inhibitors
`
`Kinetics of inhibition. The drugs have become the
`mainstay of therapy for GERD and are also used in the
`treatment of duodenal and gastric ulcer disease, espe-
`cially since treatment of many of these patients involves
`eradication of H. pylori.
`Since proton pump inhibitors are covalent and require
`acid secretion for maximal effect, their use requires a
`degree of understanding of these characteristics for
`rational treatment of patients. The drugs are given in
`association with food, so as to stimulate the parietal cell
`to make acid. They are also protected to varying extents,
`
`based on their formulation, from gastric acidity. They all
`have a relatively short plasma half-life. They therefore
`appear in the blood about 30 min after oral dosage and
`are present in the blood for about 90 min. Therefore, the
`window of opportunity for inhibition of active gastric
`pumps is about 90 min. After 90 min,
`if there are
`pumps that were inactive but now can be activated, they
`will not be inhibited since the prodrug has disappeared
`from the blood. It can be estimated, from the rate of
`
`onset of inhibition by proton pump inhibitors, that about
`75% of the pumps are morning meal activated.
`On the other hand, those pumps that are inhibited are
`inhibited covalently and thus inhibition of these pumps
`lasts well beyond the plasma dwell
`time of
`the
`compounds. Two possibilities exist for these inhibited
`pumps: either
`the inhibition is not
`reversed and
`eventually the pumps are consumed by the normal
`protein degradative pathways, or there is reversal of
`inhibition by glutathione as a reagent able to break the
`disulphide bond and reverse the inhibition. In contrast,
`there is continuing de novo synthesis of pumps; in the
`
`© 2000 Blackwell Science Ltd, Aliment Pharmacol Ther 14, 138371401
`
`ATTORNEY CONFIDENTIAL
`
`AZV00293739
`
`Page 6
`
`Dr. Reddy‘s Exh. 1047
`
`Page 6 Dr. Reddy's Exh. 1047
`
`

`

`REVIEW: GASTRIC ACID AND H. PYLORI ERADICATION
`
`1389
`
`€73.4in exam
`
`Farie‘tai {flail
`
`
`
`Figure 4. A pictorial representation of the mechanism of inhibition of the gastric proton pump by proton pump inhibitors. Illustrated
`is the drug in the cytoplasm of the parietal cell diffusing through the membrane of the canaliculus and accumulating due to protonation.
`The sulfenic acid then forms a disulphide with cys 813 in the loop between M5 and M6, to form a covalently inhibited complex.
`Other cysteines can also react but may not be directly involved in inhibition.
`
`rat, the half-life of the pump protein is z 50 h.28 This
`means that about 25% of pumps are replaced each day.
`If then, there is no reversal of the disulphide bond, it is
`predicted that steady state inhibition would, on once-a-
`day morning dosage, happen on the 3rd day of
`treatment and would be about 66% of stimulated acid
`
`output. For example, on the first dose, 25% of pumps
`would remain and by the next morning 50% of the
`normal complement of pumps would be present due to
`
`replacement by de novo synthesis. With this dose, 75% of
`these would be inhibited,
`leaving only 12.5% of the
`normal complement of pumps present immediately after
`dose. By the next morning, 37.5% of the normal
`secretory capacity would be present and after a further
`dose of the drug, again 75% active, 9% of capacity
`would remain immediately after the dose. On the third
`morning, after synthesis of again, 25% of pump protein,
`only 34% of secretory capacity would remain. With
`further
`iteration of
`this calculation, no significant
`increase in acid inhibition would occur. In order to
`
`improve the rate of onset of inhibition, either a very
`large dose can be given so that the drug can remain at
`high levels in blood for longer than 90 min, or divided
`
`doses are given along with meals either once or twice a
`day. On twice-a-day dosage, inhibition of acid secretory
`capacity improves to about 80% of maximally stimu-
`lated output. Although there are claims of a more rapid
`onset of inhibition with some of the drugs available,
`there are as yet no clinical data establishing that these
`claims provide superior results.
`Just after these drugs were introduced, it was feared
`that
`their use would result
`in achlorhydria and
`consequent bacterial overgrowth in the stomach. The
`idea was that these were irreversible inhibitors of the
`
`gastric ATPase and that this translated into irreversible
`inhibition of gastric acid secretion. This is clearly not
`the case, as illustrated by the arithmetical argument
`above, and achlorhydria is not found with treatment
`using any oral proton pump inhibitor.
`
`Therapeutic target for pH elevation. A number of clinical
`trials have established the superiority of proton pump
`inhibitors, compared to HZ-receptor antagonists, for the
`treatment of any acid related disease.
`However, the margin of superiority varies according to
`the success of the receptor antagonist
`for a given
`
`© 2000 Blackwell Science Ltd, Aliment Pharmacol Ther 14, 138371401
`
`ATTORNEY CONFIDENTIAL
`
`AZV00293740
`
`Page 7
`
`Dr. Reddy‘s Exh. 1047
`
`Page 7 Dr. Reddy's Exh. 1047
`
`

`

`1390
`
`G. SACHS et a1.
`
`disease. In the case of healing of GERD, moderately
`severe to severe, proton pump inhibitors are effective to
`about the 90th percentile within 4 weeks, compared to
`the 65th percentile for receptor antagonists. They are
`also significantly better in symptom relief and preven-
`tion of heartburn.
`
`superiority in treatment of
`there is
`Additionally,
`duodenal and gastric ulcers, whether NSAID related or
`H. pylori related. In earlier trials,
`infection was not
`considered and overall, proton pump inhibitors healed
`peptic ulcers twice as fast as Hz-receptor antagonists.
`In a series of meta-analyses of more than 300 clinical
`trials where the degree of pH elevation was plotted against
`the rate of healing, the optimal pH for healing of GERD
`and duodenal ulcers was estimated, based on predicted
`daily median pH achieved by HZ-receptor antagonism
`and by omeprazole as a representative proton pump
`inhibitor. It appears that achieving a median pH of 3 .0 for
`1 8 h per day is sufficient for ‘optimal’ healing of duodenal
`ulcer disease.29 This optimum is deduced from the pH
`expected from current proton pump inhibitor treatment.
`In the case of GERD, a median pH of 4.0 is required for
`optimization.30 Re-plotting the values obtained from
`these meta-analyses allows a direct comparison of proton
`pump inhibitors with HZ-RAs as therapeutic contenders
`for healing of duodenal ulcer and GERD, as shown in
`Figure 5. A discussion of the use of these agents in the
`
`eradication of H. pylori will be discussed in the second
`segment of this review.
`
`PROTON PUMP INHIBITORS AS DRUGS
`
`There are four proton pump inhibitors currently on the
`market: omeprazole,
`lansoprazole, pantoprazole and
`rabeprazole,
`the most
`recent of
`the proton pump
`inhibitors. In this section we shall discuss their use
`
`mainly for GERD and purely NSAID-induced gastric
`ulcers. Their employment in the eradication of H. pylori
`will be dealt with in a separate section of this review.
`
`GERD
`
`Gastro-oesophageal reflux disease is a rapidly growing
`ailment world-wide. Although the acid (or peptic)
`damage to the oesophageal epithelium is due to an
`incompetence of
`the lower oesophageal
`sphincter,
`effective therapy directed at normalizing its complex
`relaxation pattern has not surfaced. Hence acid control
`is the effective method of choice for treatment of simple
`heartburn to grade IV erosions.
`A guiding postulate of treatment for healing of GERD
`based on meta-analysis is, as shown in Figure 5,
`to
`elevate median intragastric pH to at least 4.0 for 18 h
`or more per day. This is predicted to optimize healing
`
`LEVEL and DURATION of INTRAGASTRIC pH ELEVATION
`for
`OPTIMAL TREATMENT
`of:
`
`=
`
`H
`@ am Hi mmaem?
`fl op PUMPHNHHBHT©R
`
`Hrs per DAY
`with
`PH >3:4 0" 5
`24hr
`
`IL 115
`SEE-g;
`DUODENAL
`eradgga‘iion
`+
`
`l
`U LCER
`A'biotica ..
`1 6h r "
`'
`
`
`
`8hr
`
`
`
`
`
`
`pH>3
`
`pH>4
`INTRAGASTRIC pH
`
`pH>5
`
`Figure 5. Meta—analysis of the pH elevation and duration of elevation required for optimization of healing of duodenal ulcer and GERD
`(modified from29' 30). C, untreated; R b.d., ranitidine; P., omeprazole o.d.
`
`© 2000 Blackwell Science Ltd, Aliment Pharmacol Ther 14, 138371401
`
`ATTORNEY CONFIDENTIAL
`
`AZV00293741
`
`Page 8
`
`Dr. Reddy‘s Exh. 1047
`
`Page 8 Dr. Reddy's Exh. 1047
`
`

`

`REVHNN:GASTRHIACH)ANI)H.PYLORIERADICATION
`
`1391
`
`and has, perhaps, a rational basis. The disease is caused
`by the incompetence of the lower oesophageal sphinc-
`ter, allowing acid and other gastric contents to reflux
`into the oesophagus. Although the major damaging
`agent
`is acid, as evidenced by the efficacy of acid
`secretory inhibition, other factors such as pepsin and
`perhaps bile may also play a role in determining the
`severity of the disease. In the absence of effective and
`safe medical anti-reflux therapy, effective acid suppres-
`sion is at this time the only medical option.
`With the increasing incidence of oesophageal cancer
`related to the effect of acid on the oesophageal
`epithelium,
`resulting in the metaplasia of Barrett’s
`disease, a pre-cancerous state, treatment of acid reflux
`is becoming more heroic, not only attempting to heal
`the lesions induced by reflux, but also aiming for rigid
`control of symptoms. Hence it
`is not only control of
`acidity to a mean diurnal pH of >4.0 but prevention of
`any excursions to a pH < 2.0 that may be desirable
`targets. Here, even large doses of Hz-RAs (q.d.s.) do not
`achieve the pH of 4.0 required for optimal healing,
`whereas once—a—day dosage of proton pump inhibitors is
`usually effective. Medical treatment of GERD therefore
`requires proton pump inhibitors as first
`line therapy
`since other agents are less effective for acid suppression.
`The human oesophageal epithelium is a multilayered,
`stratified squamous epithelium. Afferent nerves are
`present
`reaching into the superficial
`layers of the
`epithelium. Rather than being protected by a continu-
`ous tight junction, this epithelium has largely regional
`cell-to-cell contact via desmosomes. The multilayered
`nature of the