`
`(Printed in Great Britain)
`
`REVIEW ARTICLE
`Regulation of intracellular pH in eukaryotic cells
`
`Inger Helene MADSHUS
`Department of Biochemistry, Institute for Cancer Research, The Norwegian Radium Hospital, Montebello,
`0310 Oslo 3, Norway
`
`1
`
`INTRODUCTION
`The fact that cytoplasmic pH is strictly regulated has
`only been appreciated during the last 10 years. Eukaryotic
`cells clamp cytoplasmic pH at 7.0-7.4 by ion transport
`mechanisms and a high buffering capacity of the cytosol.
`The values of internal pH observed (pH 7.0-7.4) are
`higher than expected if the protons had been passively
`distributed across the cell membrane according to
`electrochemical gradients. Thus, when the membrane
`potential is -59 mV, pHi should be 1 unit less than
`pHo. The equilibrium relation between membrane poten-
`tial, Vm, and internal and external H+ concentrations is
`given by the Nernst expression:
`Vm = 1000 (R x TIF) ln [H+]O/[H+],)
`where R and F are the gas constant and Faraday
`constant respectively, and Vm is membrane voltage in
`mV. At 22 °C, Vm = 59 (pH, - pHO). At extracellular
`pH 7.4 the calculations would predict the cytosolic pH to
`be 6.4. This pH value is cytotoxic and far below the one
`actually observed. In barnacle muscle pH, was measured
`to be 0.5 unit higher than expected from the electro-
`chemical gradient (Roos & Boron, 1982), and similar
`findings have been made with a number of cell types.
`This fact clearly shows that there are mechanisms actively
`removing acid equivalents from the cytosol.
`The buffering capacity of cells has been determined to
`be between 10 and 50 mm per pH unit, depending on cell
`type investigated and on the conditions of the measure-
`ments (whether or not the buffering capacity has been
`measured in the presence of bicarbonate).
`Because of the great importance of the internal pH for
`many cellular processes and because the field concerning
`pH-regulating mechanisms is rapidly expanding, I here
`present an updated overview of the reasons why intra-
`cellular pH must be strictly controlled, the methods
`used to study regulation of intracellular pH and the
`mechanisms involved in the regulation of intracellular
`pH.
`
`REASONS WHY INTRACELLULAR pH MUST BE
`STRICTLY CONTROLLED
`Intracellular pH is important for the activity of a
`number of enzymes with pH optima within the physio-
`logical pH range as well as for the efficiency of contractile
`elements and the conductivity of ion channels. Also, pH
`oscillations seem to be important in controlling the cell
`cycle and the proliferative capacity of cells.
`Effect of pH on the activity of metabolic enzymes and
`synthesis of macromolecules
`The activity of a large number of intracellular enzymes
`taking part in the cellular metabolism is pH-sensitive. An
`
`Vol. 250
`
`important example is phosphofructokinase, the rate-
`limiting enzyme of glycolysis. The activity of this enzyme
`strongly increases with increasing pH over a small pH
`interval within the physiological range (Fidelman et al.,
`1982; Ui, 1966). Insulin stimulates the key enzyme of the
`glycolysis by increasing the pH, by activating the
`electroneutral Na+/H+-exchanger in the plasma mem-
`brane (Moore, 1981). In agreement with the results
`described above, Seglen (1972) reported that in perfused
`rat liver cells both glycolysis and respiratory activity
`were inhibited by low pH.
`Also, protein synthesis is affected by pH. Thus, in a
`cell-free translation system Winkler (1982) found a sharp
`increase in the rate of protein synthesis starting at pH 6.9
`with an optimum at pH 7.4.
`The -synthesis of DNA and RNA increase with
`increasing intracellular pH within the physiological
`range. The pH optimum ofDNA polymerases is generally
`quite high. The activity increases with increasing pH
`from 7.0 to 8.0, which encompasses the usual physio-
`logical pH, range (Gerson, 1982). This can probably be
`related to the rise in the free energy of hydrolysis of
`ATP and other nucleoside triphosphates observed with
`increasing pH (Alberty, 1968).
`Within the physiological range, it appears to be a
`general rule that with increasing pH, the metabolic
`activities of cells increase.
`
`Effect of pH on contractile elements
`The contractile activity of purified preparations of
`actin and myosin has been shown to be dramatically
`influenced by comparatively small changes in pH
`(Condeelis & Taylor, 1977) with low pH reducing the
`Also, microtubule assembly and dis-
`contractility.
`assembly is affected by pH with an increased disassembly
`at alkaline pH (Regula et al., 1981).
`Acidification dramatically reduces the contractility of
`muscles. Apparently, intracellular acidosis may account
`for between 40 % and 50 % of the immediate negative
`inotropic effect of ischaemia in the heart muscle (Jacobus
`et al., 1982).
`
`Effect of pH on ion conductivities
`Some ion channels have a pH-dependent conductivity.
`In particular, potassium channels in excitable cells are
`often pH-dependent (see Moody, 1984). Intracellular
`acidification blocks the K+ conductance and depolarizes
`the membrane, thereby facilitating the occurrence of
`action potentials. pH-sensitive K+ channels have been
`shown to be of importance for the generation of Ca2+-
`dependent action potentials in crayfish slow muscle fibres
`(Moody, 1982). Also, in the case of vertebrate muscle
`fibres (Blatz, 1980) and in the squid giant axon (Wanke
`et al., 1979) the K+ conductance was shown to decrease
`with increasing intracellular acidification.
`
`MYLAN INST. EXHIBIT 1049 PAGE 1
`
`
`
`2
`
`Changes in ionic conductance in the pancreatic fl-cell
`plasma membrane most likely represent fundamental
`steps in stimulus-secretion coupling. The glucose-in-
`duced electrical activity in fl-cells has been shown to be
`modulated by pH1 through the effect on K+ channels
`(Tarvin et al., 1981; Rosario & Rojas, 1986). When the
`K+-channel conductance decreases, the membrane is
`depolarized and voltage-gated Ca2+ channels and Na+
`channels are activated (Pace et al., 1982). The resulting
`increase in Ca2 1 is a stimulus for exocytosis of insulin-
`containing vesicles. pH, changes thereby indirectly
`regulate the release of insulin.
`With respect to Ca2+ channels, Umbach (1982) found
`that in Paramecium the Ca21 currents were decreased by
`decreasing pH,. The titration effect indicated a single
`titratable group with an apparent dissociation constant
`of 6.2. In Aplysia neurons, however, no effect on Ca2"
`currents was observed when pH1 increased by 0.35 unit
`(Zucker, 1981). However, the resting pHi in Aplysia
`neurons is 7.17, as opposed to 6.8 in Paramecium. The
`titration curve obtained by Umbach predicts that no
`effect on the Ca2" currents is obtained when pHi is raised
`to values above 7.17. In contrast with K+ channels,
`comparatively little is known about the pH-dependency
`of Ca21 channels. A number of workers have measured
`the effects of an imposed pHi change on Ca2 1. However,
`the effect on Ca2 i of changing pHi is not uniform among
`cells and cannot be predicted with confidence (see
`Moody, 1984).
`In Xenopus embryos the conductance of gap junctions
`was found to be blocked by decreasing pH (Turin &
`Warner, 1977, 1980). The conductance changed steeply
`over a small pH1 interval. In experiments on isolated
`coupled cells in which the pH1 and the junctional
`conductance were directly meaured, the relationship was
`shown to be a simple sigmoid curve (Spray et al., 1982).
`The relationship was well fitted by a Hill plot with an
`apparent pKH of 7.3.
`that pHi changes might play a
`This suggested
`physiological role in modifying electrical communi-
`cations between cells (Spray et al., 1982).
`Altogether, it appears that changes in pHi modify the
`electrical properties of excitable cells. Thus, small pH1
`changes dramatically influence the responsiveness of
`such cells (Moody, 1984). Also, intercellular communi-
`cation via gap junctions, which is important in develop-
`ment and in organized functioning of tissues, is highly
`sensitive to pH changes (Spray et al., 1982).
`Control of the cell cycle
`Oscillations in intracellular pH have been postulated
`to be of importance in the control of the cell cycle and
`cell division in several cell types. Low intracellular pH is
`common both to prokaryotic and eukaryotic resting
`cells. This is one reason that these cells have low
`metabolic activities. A rapid increase in intracellular pH
`may be important to bring cells from Go/Gl and into S
`phase.
`Many observations exist with respect to pH oscillations
`during the cell cycle. In the ciliated protozoan Tetra-
`hymena pyriformis two pH increases of about 0.3-0.4 pH
`unit above a baseline of 7.2 were observed within 30 min
`of the onset of S phase (Gillies & Deamer, 1979). A single
`pH1 shift of similar magnitude has been reported in the
`slime mould Physarum polycephalum (Gerson & Burton,
`1977). In this case the pH, increase was associated with
`
`I. H. Madshus
`
`cell division and mitosis. These authors suggest that
`pH, increases play a significant role in induction of cell
`proliferation and cell division.
`Steinhardt & Morisawa (1982) also presented evidence
`for a pH-sensitive step prior to mitosis in Physarum
`polycephalum. In the cell cycle the pH, was increased to
`pH 7.4-7.5 within 1 h prior to mitosis. If pHi was
`artificially lowered, a delay of mitosis was induced.
`Furthermore, starvation of the cells lowered the pH, and
`interrupted the cell cycle, thereby inhibiting mitosis.
`According to Gerson (1982) a biphasic increase in
`pH, was observed when lymphocytes were stimulated
`mitogenically with concanavalin A. The first peak was
`seen after 6-8 h and the second peak 48 h after the
`stimulation. The first peak correlated with early events,
`such as increase in the synthesis of phospholipids and
`protein, while the second peak correlated with the
`synthesis of DNA.
`The results described above support the hypothesis
`that intracellular pH is an important modulator of cell
`proliferation and cell division. Experiments with early
`Xenopus embryos indicated, however, that pH, is not
`necessarily a universal regulator of mitosis. It was
`reported that even if small oscillations in pHi occurred
`during the mitotic cycle in early Xenopus embryos, no
`delay in mitosis occurred when the pHi was artificially
`reduced (Lee & Steinhardt, 1981). Furthermore, clamping
`at high pHi values did not interfere with normal
`chromosome cycling or cell division in the sea urchin
`embryo (Grainger et al., 1979). This suggests that the
`pH, shifts observed in other systems (Tetrahymena and
`Physarum) reflect changes associated with cell growth
`but not mitosis or cytokinesis per se.
`Pouyssegur et al. (1985) reported that in response to
`growth factors, quiescent fibroblast mutants lacking
`Na+/H+ exchange activity failed to elevate their cyto-
`plasmic pH and to reinitiate DNA synthesis at neutral
`and acidic pHo. These authors claim that a pHi threshold
`of 7.2 exists, below which growth factors cannot induce
`the Go to S phase transition.
`In contrast with this, Mills et al. (1985) suggested that
`pH, increase is not essential for proliferation. This con-
`clusion was based upon observations with interleukin 2-
`induced lymphocyte proliferation. Interleukin 2 stimu-
`lates electroneutral Na+/H+ exchange, thereby increasing
`pHi. However, if this exchanger was inhibited with
`amiloride analogues in a HCO3--free medium, pro-
`liferation occurred even if pHi did not increase.
`A principal difference in the experimental design
`between the experiments of Pouyssegur et al. (1985) and
`Mills et al. (1985) could explain why these authors
`reached different conclusions. Lymphocytes having inter-
`leukin 2 receptors are in G1 phase; this means that the
`cells are partly stimulated and thereby already in the cell
`cycle, while quiescent fibroblasts are in Go and not in the
`cell cycle. Possibly, a pHi increase is necessary for the
`transition of cells from Go to G1 in order to bring
`quiescent cells back into the cell cycle.
`It was recently reported by Ober & Pardee (1987) that
`a series of tumourigenic Chinese hamster embryo
`fibroblast cell lines maintain an internal pH that is
`0.12 + 0.04 pH unit above that of the non-tumorigenic
`parental cell line. This suggests a critical role for pHi in
`the regulation of DNA synthesis and suggests that
`aberrations in pH1 can contribute to the acquisition of
`altered growth properties.
`
`1988
`
`MYLAN INST. EXHIBIT 1049 PAGE 2
`
`
`
`Regulation of intracellular pH in eukaryotic cells
`
`NH4CI added
`
`NH4CI removed
`
`NH4CI added
`l
`
`NH4CI removed
`
`Na+
`
`140 mm-Na+
`
`140 mM-Na+
`plus amiloride
`No Na+
`
`5
`
`10
`
`15
`
`7.4
`
`7.2
`
`I:
`
`7.0
`
`6.81-
`
`6.4
`
`0
`
`Time (min)
`Fig. 1. Na+/H+ antiport studied by the ammonium prepulse
`technique
`When the prepulse is applied, an abrupt acidification of
`the cytosol is produced. This acidification is followed by
`a rapid pH, normalization in a sodium-containing buffer
`due to Na+/H+ antiport. The normalization follows an
`exponential time-course, being complete after 5 min in a
`bicarbonate-free medium.
`
`Altogether, most data presented support the hypo-
`thesis that a strictly controlled pHi acts as a second
`messenger in growth control.
`
`METHODS TO STUDY INTRACELLULAR pH
`REGULATION
`A number of different techniques are now available to
`measure intracellular pH. For a detailed description of
`these methods the reader is referred to Nuccitelli &
`Deamer (1982). Here the different methods are only
`mentioned briefly.
`A widely used method for estimating pHi is to measure
`the equilibrium distribution of a radioactively labelled
`weak acid or base across the plasma membrane. However,
`this method has limitations such as sensitivity to cellular
`volume changes and intracellular compartmentalization
`of the tracer (Roos & Boron, 1981). Also, this method
`shows a poor temporal resolution.
`N.m.r. was first used to measure intracellular pH in
`red blood cells by Moon & Richards (1973). The
`inorganic phosphate signal is most commonly used,
`because it is readily observable in the majority of 31P
`spectra and because its frequency is particularly sensitive
`to pH changes in the region around neutrality (Gadian
`et al., 1982). A disadvantage is that n.m.r. equipment is
`expensive and complicated to use. The requirement for a
`high cell density also creates problems, but these can be
`solved by various methods of superfusion (Nuccitelli,
`1982).
`
`Vol. 250
`
`3
`
`pH-sensitive microelectrodes offer one way of ob-
`taining continuous pH measurements. The recessed-tip
`type of pH microelectrode, first described in 1974 by
`Thomas, can be made with tip diameters of less than
`1 ,um, and can thus be used on a wide variety of cells.
`However, one is at risk of destroying the cell when the
`electrode is introduced through the cell membrane.
`When regulatory mechanisms are studied, it is con-
`venient to use fluorescent probes with pH-dependent
`fluorescence. Most often carboxylfluorescein derivatives
`are used. For these compounds the excitation, and
`thereby the fluorescence intensity,
`increases almost
`linearly from pH 6.5 to pH 8.0. Fluorescent probes can
`be introduced to the cytosol in the form of lipid-soluble
`esters, which are cleaved by cytoplasmic esterases, and
`the fluorophore is thereby trapped in the cytosol (Thomas
`et al., 1979; Rink et al., 1982; Moolenaar et al., 1983).
`Such techniques are non-invasive, and spectrally moni-
`tored fluorescence can be recorded continuously with
`excellent temporal resolution.
`To study pH regulation the cytosol may be acidified or
`alkalinized and then the regulation of the pH back to
`normal value is studied under different conditions by
`a continuous recording of pH. In this way different
`mechanisms for pH regulation have been characterized,
`and, in the following, four such mechanisms will be
`described in detail.
`
`MECHANISMS INVOLVED IN THE
`REGULATION OF INTRACELLULAR pH
`Na+/H+ antiport
`The first demonstrations of Na+/H+ antiport in
`eukaryotic plasma membranes were made in vesicles
`from the brush borders of rabbit kidney and small
`intestine (Murer et al., 1976). Electroneutral Na+/H+
`antiport was described as an important pH-regulating
`mechanism in sheep heart Purkinje cells after cytoplasmic
`acidification (Deitmer & Ellis, 1980). Recently, much
`information has accumulated regarding the Na+/H+
`exchanger in mammalian cells, and there seems to be
`general agreement that all animal cells possess an
`electroneutral Na+/H+ antiporter
`review,
`(for
`see
`Krulwich, 1983). The antiporter responds to a fall in
`extracellular pH by quickly extruding protons in ex-
`change with extracellular Na+ (Fig. 1). The energy for the
`extrusion is provided by the large inward-directed Na+
`gradient.
`The cytosol can be acidified by the ammonium prepulse
`technique described by Boron & De Weer (1976).
`Addition of NH4C1 to the culture medium first produces
`a rapid rise in pH, due to entry of NH3, which combines
`with protons in the cytosol. The pH, subsequently slowly
`decays due to entry of NH4'. When NH4C1 is withdrawn
`from the medium, NH3 leaves the cells while H+ ions are
`left behind. Therefore, the pH becomes lower than the
`original pH, value before the pulse was applied (Fig. 1).
`The Na+/H+ antiporter is inhibited by the potassium-
`saving diuretic amiloride and by analogues of this drug.
`Amiloride inhibits competitively binding to the extra-
`cellular site which accommodates Na+ ions (Haggerty
`et al., 1985).
`During extrusion of protons after an acid load, a large
`amount of Na+ ions enter the cell due to Na+/H+
`exchange. The Na+ ions are then extruded into the
`
`MYLAN INST. EXHIBIT 1049 PAGE 3
`
`
`
`4
`
`medium by the Na+/K+-ATPase. If the ATPase is
`the Na+/H+ antiport
`inhibited with ouabain,
`will
`eventually be inhibited, because the sodium gradient,
`which drives the process, is dissipated. Depending on the
`direction of the Na+ gradient, HI ions will either enter or
`leave the cell by Na+/H+ exchange. At low extracellular
`Na+ the sodium gradient is reversed, and cytoplasmic
`acidification will follow when the Na+/H+ exchanger is
`stimulated (Moolenaar et al., 1983; Paris & Pouyssegur,
`1983).
`Normally, the rate of Na+/H+ exchange is small, just
`balancing the passive H+ influx and the intracellular
`production of acidic metabolites. The Na+/H+ exchanger
`is regulated; otherwise the actual Na+ gradient would
`mediate an H+ effilux, resulting in a resting pHi more
`alkaline than the actual pHi (Moolenaar, 1986).
`There is evidence that the intracellular pH is the
`important parameter regulating the activity of the
`exchanger. Aronson and coworkers proposed that
`cytoplasmic H+ acts as an allosteric activator of the
`Na+/H+ exchanger (Aronson et al., 1982). It has been
`suggested that the antiport molecule has on its cyto-
`plasmic face two proton-binding sites that are separate
`and functionally independent. One site is a regulatory or
`modifier site. When this site is occupied, a conformational
`change is triggered, activating the exchanger. A distinct
`H+-transport site mediates the net extrusion of H+ once
`the exchanger is activated (Aronson et al., 1982; Aronson,
`1985).
`The amiloride-sensitive Na+/H+ exchanger seems to
`be involved in hormonal stimulation of cell growth
`(Smith & Rozengurt, 1978; Moolenaar et al., 1982;
`Schulinder & Rozengurt, 1982). There is evidence that
`mitogen-induced cytoplasmic alkalinization is due to
`Na+/H+ exchange (Rothenberg et al., 1982, 1983a,b;
`1982, 1983;
`Cassel et al., 1983; Moolenaar et al.,
`L'Allemain et al., 1984). Experiments indicate that serum
`in human fibroblasts (Moolenaar et al., 1983), thrombin
`in hamster fibroblasts (Paris & Pouyssegur, 1983) and
`phorbol esters in T-lymphocytes (Grinstein et al., 1985)
`may activate the Na+/H+ antiporter by increasing the
`affinity of the pH, sensor, the allosteric H+-binding
`site.
`The understanding of the sequence ofevents mediating
`the growth-factor-induced rise in pHi is still incomplete.
`Protein kinase C is a strong candidate for a transducer of
`growth-factor-mediated activation of the Na+/H+ anti-
`porter. The fact that a variety of extracellular signals
`stimulate inositol phospholipid breakdown and therefore
`the formation of endogenous diacylglycerol supports this
`1984; Berridge,
`1984). Also,
`possibility (Nishizuka,
`phorbol 12-myristate 13-acetate and other phorbol esters
`were suggested to stimulate Na+/H+ exchange through
`an effect on protein kinase C (Grinstein et al., 1985).
`However, the activation of the antiporter by growth
`factors may also be mediated by other pathways, such as
`steps activated by Ca2+-calmodulin or GTP-binding
`proteins. Also, transmethylation reactions have been
`proposed to modulate the activity of the Na+/H+-
`exchanger (Dudeja et al., 1986).
`It can be concluded that the Na+/H+ antiporter is
`apparently present in all animal cells and that it plays an
`important role in regulation of pHi back to normal value
`after an acid load. Also, the antiporter seems to mediate
`an alkalinization of the cytosol when hormones and
`growth factors combine with their cellular receptors.
`
`I. H. Madshus
`
`1600
`
`m~~~~
`
`ci
`
`U3
`
`-c
`
`V11
`
`M
`
`Cu
`
`1200~
`
`_
`
`800 t
`
`4001
`
`UI
`
`7
`001
`I
`
`o S.
`
`6.5
`
`7.5
`
`0
`
`7.0
`pHj
`Fig. 2. Effect of intracellular pH on the uptake of 36Cl- by anion
`antiport in Vero cells
`For experimental details, see Olsnes et al. (1987b).
`
`Anion antiport
`Most cells require bicarbonate for continuous growth.
`One reason for this may be that Cl-/HCO3- antiport is
`important in the regulation of pH,. A bicarbonate-linked
`mechanism was suggested by Russell & Boron (1976) to
`be the most important pH regulatory mechanism in
`invertebrate cells.
`The earliest experiments studying pH, regulation were
`performed on large invertebrate cells such as snail
`neurons and squid axons. The registrations of intra-
`cellular pH were continuous by pH-sensitive micro-
`electrodes applied intracellularly. Thomas (1982) re-
`ported that in snail neurons intracellular chloride and
`extracellular HCO3- were necessary to normalize pHi
`after an acid load. He also found the system to be
`electroneutral and dependent on extracellular Na+.
`Furthermore, the pH regulation was completely inhibited
`by the anion-exchange inhibitor SITS. Becker & Duhm
`(1978) proposed that pH1 might be regulated by an
`exchange of internal Cl- for an external ion pair
`consisting of one Na+ ion linked to one CO3 2- ion.
`Cl-/HCO3- exchange has been described to be Na+-
`dependent in hamster fibroblasts (L'Allemain et al.,
`1985), and in squid axon (Boron, 1985), while Na+-
`independent Cl-/HCO3- exchange has been described in
`sheep heart Purkinje fibres (Vaughan-Jones, 1979), and
`in LLC-PK1 cells (Chaillet et al., 1986). Evidence was
`recently presented that both Na+-dependent and Na+-
`independent Cl-/HCO3- antiport take place in Vero cells
`(T0nnessen et al., 1987; Olsnes et al., 1987a; Madshus
`& Olsnes, 1987).
`It was suggested by Russell & Boron (1976) that the
`squid axon might possess an ATP-dependent pump
`which responded to acid challenges by taking up
`HC03- from the medium in exchange with intracellular
`chloride. However, convincing evidence for the existence
`of an energy-dependent mechanism involved in anion
`
`1988
`
`MYLAN INST. EXHIBIT 1049 PAGE 4
`
`
`
`Regulation of intracellular pH in eukaryotic cells
`
`5
`
`et al. (1986) that the Na+-independent anion antiporter
`shows a steep rise in Vmax with increasing pH, above
`pH 7.0. In Fig. 2 is shown how 36Cl- uptake by anion
`antiport increases strongly over a narrow pH interval.
`An electroneutral Cl-/HCO3- exchanger could also
`participate in acid extrusion, but only if pH, falls
`sufficiently to raise [HC03-]O/[HC03-] above [Cl-]0/
`[Cl-]1. Also, if the chloride gradient was reversed by
`removing extracellular chloride the sodium-independent
`anion antiporter was shown to extrude acid equivalents
`(T0nnessen et al., 1987; Madshus & Olsnes, 1987).
`Compared with the great amount of information
`accumulating regarding the regulation of Na+/H+ anti-
`port, there has been little information regarding the
`regulation of the anion antiport. Possibly, the activity of
`the sodium-linked and the sodium-independent Cl-/
`HCO3 antiporters is subject to regulation by hormones
`and growth factors, in a similar way as the Na+/H+
`antiporter.
`It can be concluded that anion antiport mechanisms
`constitute important pH regulatory mechanisms and
`that by anion antiport pHi can be normalized both
`after acid and alkali loads. After an acid load, pHi is
`normalized by the uptake of a charged complex of Na+
`and HCO3- in exchange with intracellular Cl-, while
`after an alkali load pH, is normalized by extrusion of
`intracellular HCO3- in exchange with extracellular Cl-.
`It is still unclear whether different molecular entities are
`involved in the two kinds of anion antiport.
`
`Na+/HCO3- symport
`A number of authors have recently presented evidence
`for electrogenic symport of Na+ and HCO3- in a variety
`of cells involved in transepithelial transport of acid
`(Boron & Boulpaep, 1983; Alpern, 1985; Jentsch et al.,
`1985, 1986b; Yoshitomi et al., 1985; Biagi & Sohtell,
`1986). This symport has been described to couple tightly
`the transport of one Na+ ion to two or three HCO3- ions.
`It has also been suggested that the negatively charged
`complex is the ion pair NaCO3- (Jentsch et al., 1986a).
`The transport is inhibited by the anion exchange
`inhibitors SITS and DIDS. The first example of such
`transport was described to take place in the basolateral
`membrane of the proximal tubule of the salamander
`Ambystoma tigrinum (Boron & Boulpaep, 1983). The
`transporter was postulated to mediate net efflux of
`HCO3- and Na+, giving rise to transcellular transport of
`HCO3-. A similar process was described in monkey
`kidney epithelial cells (Jentsch et al., 1985), and in
`proximal tubular cells ofthe rat (Alpern, 1985; Yoshitomi
`et al., 1985) and the rabbit (Biagi & Sohtell, 1986; Sasaki
`et al., 1985).
`In all these cases it was found that a sudden peritubular
`reduction of Na+- and/or HCO3- concentration caused a
`sudden transient depolarization as well as a reduction in
`pHi. The cytoplasmic acidification indicates that bi-
`carbonate efflux takes place in response to peritubular
`reduction of bicarbonate and sodium ions. The fact that
`bicarbonate efflux takes place in response to an altered
`sodium gradient indicates strongly that a complex of
`Na+ and HCO3- is transported, and the resulting
`depolarization indicates that this transport is electro-
`genic.
`The 'leakage' of HCO3- across the basolateral side of
`the polarized cell membrane mediates intracellular
`
`Fig. 3. Schematic drawing showing pH-regulatory mechanisms
`described in the text
`On the left side mechanisms correcting acidification of the
`cytosol are depicted, while on the right side pH-regulatory
`mechanisms producing intracellular acidification or cor-
`recting intracellular alkalinization are depicted.
`
`antiport has not been presented. The energy available
`from the large inward-directed Na+ gradient is more than
`sufficient to drive the uptake of NaCO3- and the effilux of
`Cl-.
`Cl- and HCO3- ions are relatively
`close
`to
`electrochemical equilibrium.
`It can be concluded that cytoplasmic pH can be
`effectively normalized after an acid load by anion
`antiport. HCOJ-/Na+ as a negatively charged complex is
`exchanged with intracellular chloride, and the process is
`probably driven by the sodium gradient and in some
`cases by an additional inward-directed HC03- gradient.
`The ion gradients are such that the Na+-independent
`Cl-/HCO3- exchanger will under physiological con-
`ditions bring chloride ions into the cell and bicarbonate
`out of the cell. Thus, the [CI-]/[CI-] is generally greater
`than [HC03-10/[HC03-11 and a passive and electrically
`neutral Cl-/HCO3- exchanger would therefore normally
`mediate Cl- uptake and HCO3- extrusion. This process
`could be of importance in keeping the Cl- activity above
`the electrochemical equilibrium value in certain cell types
`(Vaughan-Jones, 1979). Because of the HCO3- efflux this
`antiport would with time result in an acidification, which
`would have to be compensated for by another pH
`regulatory mechanism.
`The Na+-independent Cl-/HCO3- exchanger may be
`of importance in situations where the pH1 rises above
`neutrality. pH1 regulation on the alkaline side of the
`normal pH1 is of interest because both the coupled
`NaCO3-/Cl- exchange and the Na+/H+ exchange regu-
`late pH1 only on the acid side. It was shown by Olsnes
`
`Vol. 250
`
`MYLAN INST. EXHIBIT 1049 PAGE 5
`
`
`
`6
`
`acidification and must be compensated for by H+
`extrusion across
`the luminal membrane. Such HI
`extrusion has been shown to occur by the Na+/H+
`antiport and by H+-translocating ATPases (see below).
`The H+ extruded from the cell across the luminal
`membrane constitutes transcellular acid secretion. Thus,
`acid secretion is merely a byproduct of pHi regulation;
`the efflux of HCO3- across the basolateral membrane
`provides a sustained intracellular acid load to which the
`Na+/H+ exchanger and the electrogenic H+-translocating
`ATPase respond by extruding protons across the luminal
`membrane (see Boron, 1983).
`The possibility exists that some of the transport
`phenomena described as Na+/HCO3 symport actually
`partly consists of a Cl-/NaCO3- antiport. It seems that
`the Cl- independence for the transport has not been fully
`documented in all cases. Cl-/NaCO3 antiport has been
`to be an important mechanism in pHi
`described
`normalization after an acid load. The possibility exists
`that this antiporter could function in both directions
`depending on the electrochemical ion gradients.
`In the case of the Cl-/HCO3- antiport and Cl-/Cl-
`antiport, a certain slippage has been shown to take place
`at alkaline pH values in the sense that the coupling ratio
`between the anions is different from 1 (Olsnes & Sandvig,
`1986; Olsnes et al., 1987b). The same could be true for
`the Cl-/NaCO;- antiporter; at certain conditions more
`or less slippage and thereby electrogenic transport could
`occur.
`Na+/HCO3- symport mechanisms have been described
`almost exclusively in cells from renal tubules. These cells
`are specialized in the sense that they have the important
`function of transcellular transport of acid. Such cells
`have high concentrations of the enzyme carbonic
`still not clear whether Na+/HCO3
`anhydrase. It is
`symport constitutes a general pH, regulatory mechanism
`or if this symport is only operating in specialized acid-
`secreting cells.
`
`H+-translocating ATPases
`Proton pumps driven by hydrolysis of ATP have been
`described in the plasma membrane of certain tight
`epithelia, such as turtle urinary bladder (Steinmetz,
`1974; Gluck et al., 1982). Proton-translocating ATPases
`have also been described in a number of intracellular
`organelles (Anderson et al., 1982; Hutton, 1982; Dean
`et al., 1984; Forgac et al., 1983; Galloway et al., 1983;
`Glickman et al., 1983). Intracellular vesicles with proton
`pumps fuse with and form from the plasma membrane,
`and it has been suggested that H+ pumps may participate
`in cellular pH regulation (Adelsberg & Al-Awqati,
`1986).
`Proton pumps are electrogenic, but are either coupled
`in parallel with anion channels or antiparallel to K+
`channels in order to counteract the buildup of a
`membrane potential (for review, see Mellman et al.,
`1986).
`Even if the proton pumps are not located at the plasma
`membrane they can actively remove H+ ions from the
`cytosol.
`The pH of intracellular vesicles has been shown to be
`between 4.5 and 6, with the lowest pH values found in
`lysosomes (see Mellman et al., 1986).
`When Hep 2 cells were treated with a hypo-osmotic
`buffer, an immediate cytoplasmic acidification of approx.
`
`I. H. Madshus
`
`0.2 pH units was observed simultaneously with an
`alkalinization of the vesicular compartment (Madshus
`et al., 1987). This suggests that swelling of intracellular
`vesicles has taken place with resulting leakage of
`intravesicular protons to the cytosol and indicates that
`the vesicular compartment normally is of great impor-
`tance in removing H+ ions from the cytosol.
`Secretion of acid in renal proximal tubular cells is
`suggested to be partly mediated by proton-translocating
`ATPases (Steinmetz & Andersen,
`1982). The acid
`secretion is a two-step process. Acid first enters the cell
`across the basolateral membrane by HC03- efflux (see
`Na+/HCO3 symport) and acid then exits the cell across
`the luminal membrane, by luminal H+ pumps and by
`Na+/H+ antiporters in the luminal membrane.
`Cannon et al. (1985) reported that the number of H+-
`ATPases in the apical plasma membrane of turtle-
`bladder cells was regulated by exocytic insertion of
`intracellular vesicles whose membrane contained the HI
`pump. CO2, which is a major stimulus for urinary
`acidification, was shown to cause rapid fusion of these
`vesicles with the luminal membrane, thereby inserting
`these pumps and increasing the rate of net transepithelial
`H+ secretion. Exocytosis was also shown to take place
`when intracellular pH was decreased by incubation of
`the cells wi