modern diuretics and the kidney - journal of clinical ... · changes in physical forces within the...

J Clin Pathol 1981;34:1267-1275

Modern diuretics and the kidneyAF LANT

From the Department of Therapeutics, Westminster Medical School, London SWIP 2AP

It is particularly appropriate to discuss the pharma-cology and mechanisms of action of diuretic drugsin a symposium on diseases of the kidney, sincediuretics were the first group of drugs to be devel-oped for the control and manipulation of selectedaspects of renal function. Despite a relatively shorthistory of less than three decades since the discoveryof the first orally acting diuretic, chlorothiazide,modern diuretic therapy continues to enjoy an ever-expanding clinical demand: worldwide expenditurehas been estimated at around US $800 millionper annum during the last five years. Thoughoriginally introduced for the treatment of oedema-tous conditions it is interesting that application ofdiuretics to the treatment of hypertension has out-stripped their use in oedema. This has happenednotwithstanding the parallel discoveries of othermajor antihypertensive agents such as the beta-adrenoceptor blocking drugs and powerful vaso-dilators.The advent of novel, orally effective diuretics

following on the prototype benzothiadiazine, chloro-thiazide, has had a major influence in stimulatingprogess in the basic sciences relating to nephrologyor electrolyte transport in body tissues in general.Advances in fundamental knowledge have in turngiven the impetus to further discoveries of new and,in some cases, unique substances possessing diureticactivity. By and large these compounds have attained aremarkable degree of sophistication while, at thesame time, remaining relatively safe during longterm treatment of patients. The widespread use and,at times, the regrettable abuse of any group ofdrugs whose primary effect is directed towardinterference with the renal handling of electrolytesmust inevitably generate secondary disturbances inbody homeostasis which have particular relevanceto the chemical pathologist. To understand these,demands an understanding of how diuretics workin the context of modern views of the functionalorganisation of the kidney.

Renal regulation of sodium excretionl 2

Four main mechanisms are believed to be involvedin the control of sodium excretion by the kidney.

THE DISTRIBUTION OF GLOMERULARFILTRATION BETWEEN OUTER CORTICALAND JUXTAMEDULLARY NEPHRONSThe functional unit of each kidney comprisesapproximately one million nephrons which lackhomogeneity in their structure. About 80% ofthem are in the outer cortex, have short loops ofHenle, and have relatively low reabsorptive capacityfor sodium. The remaining 20% are juxtamedullary,possessing long loops of Henle, and are largelyresponsible for creating the hyperosmotic inter-stitium in the medulla which meditates the processof urine concentration. Redistribution of blood flowfrom outer cortical to juxtamedullary nephronscan contribute to abnormal sodium retention,whilst predominance of the effect of outer corticalnephrons may lead to saluresis. A reduction in bloodflow to the outer part of the cortex has been foundto occur in some sodium-retaining states. Conversely,a drug which could shift blood flow from juxta-medullary to outer cortical nephrons would reducesodium reabsorption and result in an effectivediuretic response.

HAEMODYNAMIC AND PHYSICAL FACTORS3Changes in physical forces within the peritubularcapillaries have been shown to be important de-terminants of renal sodium reabsorption, especiallyin the proximal tubule. Thus, for example, renalarteriolar dilatation may, by increasing the hydro-static pressure in the vasa recta, decrease net tubularreabsorption of sodium with resulting natriuresis,while the reverse occurs with any increase of plasmaoncotic pressure. It seems probable that changesin the oncotic or hydrostatic pressures within theperitubular blood vessels achieve their effects onnet tubular reabsorption of sodium by influencingthe resistance of the intercellular channels or "shunt"paths through which sodium ions pass to reach theperitubular capillaries.

HORMONAL FACTORSA number of hormonal mechanisms operate singlyor together in encouraging sodium retention by thekidney.

1267

copyright. on January 12, 2020 by guest. P

rotected byhttp://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.34.11.1267 on 1 Novem

ber 1981. Dow

nloaded from

http://jcp.bmj.com/

1268

Renin-angiotensin-aldosterone4 5The role of this humoral system in renal sodiumregulation is considerable. Yet, because of the slug-gish characteristics of the system, it is unlikelythat these hormones, especially aldosterone, areinvolved in fine "moment to moment" modulation.Aldosterone influences epithelial transport of sodiumthrough activation of DNA-dependent RNA syn-thesis, and the response to it follows a significantlag period which corresponds to the time needed forthe induction of new protein synthesis; this mayamount to an hour or more in experimental systems.Yet it is clear that changes in sodium excretionin vivo may be induced abruptly by various ma-noeuvres. Moreover, whenexcessive amounts ofaldos-terone or other mineralocorticoids are given experi-mentally sodium retention occurs, but is transient.The "escape" from the action of aldosterone cannotbe accounted for by changes in glomerular filtrationrate or renal blood flow, and it is one of the piecesof evidence supporting the presence of anotherhumoral substance stimulating natriuresis-calledThird Factor or natriuretic hormone.

Natriuretic hormone6A growing body of evidence supports the existenceof a natriuretic humoral agent which promotesthe renal elimination of sodium, by inhibiting sodiumreabsorption in the proximal tubule. It may alsowork in harmony with the renin-angiotensin-aldosterone system to provide fine control ofsodium excretion by an action exerted mainly onthe collecting ducts. The chemical characterisationof the natriuretic hormone remains incompleteand its role in the sodium-retaining states is alsouncertain.

Prostaglandins and kinins7 8Many experimental studies have demonstratedthe ability of renal tissues to generate prostaglandinsand kinins. Whereas renin, angiotensin and aldos-terone, like prostacyclin, are transported primarilyin the vascular compartment, kallikrein-kinin andprostaglandins of the E series are associated withthe renal interstitium and tubular lumen. One ormore prostaglandins, primarily PGE2, is probablyresponsible for mediating the increase in medullaryblood flow that occurs in response to various stimuliincluding surgical trauma, salt loading, and theaction of "loop" diuretics. The generation ofkinins in the distal tubules and collecting ductsresults in the release of prostaglandins which inturn inhibit the local effects of antidiuretic hormone(ADH) and thereby contribute to the tubularexcretion of solute-free water.

FEEDBACK CONTROL SYSTEMS INVOLVINGTHE MACULA DENSA9A fourth intrarenal regulating system which hasbeen proposed is that of a servo-mechanism operat-ing between the macula densa cells of the ascendinglimb of Henle's loop and the glomerulus of thesame nephron. This system functions by the releaseof renin and angiotensin It locally in response tothe sodium concentration in the tubular fluidimpinging on the macula densa, the feedbackloop being completed by appropriate alterations inboth GFR and proximal tubular reabsorption. Thedetailed mechanisms involved in such an auto-regulatory system which couples distal salt deliveryto the filtration rate in individual nephrons remainincompletely understood.

Organisation of tubular function'° 11

Diuretic drugs have actions on ion-transportingtissues as diverse as amphibian skin, intestinalepithelium, red and white blood cells and cornea.The primary target for the action of these drugshowever is the kidney, where they promote theexcretion of water and certain electrolytes such assodium and chloride by interfering with tubularreasorptive mechanisms. Because these reabsorptivemechanisms vary according to the degree of sophisti-cation of different portions of the epithelium liningthe tubule, a brief survey of the organisation oftubular functions is relevant to understandingdiuretic action (Fig. 1).

GLOMERULUS AND PROXIMAL TUBULEIn normal man each day the renal glomeruli produceapproximately 180 litres of filtrate, and urine isfinally produced by the progressive reabsorptionof 99% of this ultrafiltrate at various stages alongthe nephron. About two-thirds of the glomerularfiltrate is reabsorbed iso-osmotically in the proximaltubule as a result of the active reabsorption ofsodium chloride and sodium bicarbonate from thetubular lumen into the peritubular fluid. Themechanisms involved in transcellular ion movementare complex and involve a variety of energy-dependent ion pumps as well as transfer paths orchannels in between the loose-fitting cells of theproximal tubule. The resistance to these intercellularshunts of ions is influenced considerably by changesin the oncotic and hydrostatic pressures within theperitubular capillaries.

ASCENDING LIMB OF HENLE'S LOOP(MEDULLARY DILUTING SEGMENT)1213There are two morphologically distinct kinds ofnephron, namely the outer cortical nephrons,

Lant



j.com/

J Clin P


ber 1981. Dow

nloaded from

http://jcp.bmj.com/

Modern diuretics and the kidney

which are the more plentiful and have short loopsof Henle, and the juxtamedullary nephrons whichhave long loops plunging down into the inner partsof the medulla. The suggestion that loops of Henlemight act as a countercurrent multiplier systemwas first proposed in 1942 by Kuhn and Ryffel,14but only received widespread attention in 1951after Wirz, et a115 showed a striking osmotic gradientin the renal interstitium increasing from cortexto papilla. In the ascending limb of the loop ofHenle two anatomically distinct portions can bedistinguished, both of which are concerned withurinary dilution, namely a medullary portionlined by cuboidal cells and a cortical portion linedby flattened cells. The tubular epithelium of bothportions is insensitive to ADH and is thus relativelyimpermeable to water but the ascending limb doespossess the ability to reabsorb salt actively. Because,in mammals, the diluting segments of the loop ofHenle do not reach the surface of the kidney,their functional characteristics have had to bededuced from micropuncture of distal convolutedtubules. Recent experiments employing techniques

PROXIMAL

1269

of tubular microperfusion have shown that thetransepithelial voltage in this part of the nephronis positive within the lumen and that chloride ionis absorbed against an electrochemical gradient.Ion substitution studies have lent support to theview that active chloride transport is the pri-mary event and that sodium moves secondarilyin this part of the nephron. Electrogenic sodiumreabsorption from the tubular fluid may coexistwith a separate neutral NaCI path (see later).

Hypertonicity of the inner medulla is the resultof combined functions of the hairpin structure ofthe long loops of Henle acting as counter currentmultiplier systems and of their associated vasa rectaacting as counter current exchange systems. Ithas been proposed that recycling of urea between thetwo limbs of the vasa recta and between the loopof Henle and the collecting duct contributes to thehypertonicity of the inner medulla on which concen-tration of the urine depends. In the presence ofADH16 the collecting ducts, which traverse thehypertonic medullary interstitium on their way tothe renal pelvis, become permeable to water

DISTAL

SiteIV

-____ NQv

Aldosterone A

NcCI

NaCl

II3TCH20ASCENDING LIMB(water impermeable)

Fig. I A diagram of the nephron showing the four tubular sites where diuretics act to block sodium chloridereabsorption.Urinary dilution occurs as a result ofsodium chloride reabsorption at two water-impermeable sites, the ascending limbof Henle's loop (site MI; medullary diluting site) and the early part of the distal tubule (site II; cortical diluting site).During water diuresis, total urine volume can be divided into two moieties, the volume of urine required to excreteurinary solutes at plasma tonicity-that is, the osmolal clearance Cosm, and the volume of solute-free water generatedat diluting sites II and III-that is, the solute-free water clearance CH,O. When fluid intake is restricted and hypertonicurine is formed, CH2O becomes negative and is referred to as TCH2o. TCH2O reflects the passive reabsorption into thehypertonic medulla of solute-free water from the collecting ducts, stimulated by ADH (see footnote on page 1270).



j.com/

J Clin P


ber 1981. Dow

nloaded from

http://jcp.bmj.com/

1270

Progressive extraction of solute-free water from thecollecting ducts into the hypertonic medulla rendersthe residual tubular urine hypertonic.

CORTICAL DILUTING SEGMENTAs the diluting segment of the loop of Henleascends out of the medulla to reach the cortex,continued reabsorption of sodium without waterfurther dilutes the tubular fluid and enables a diluteurine to be excreted in the absence of ADH. Thesodium that is reabsorbed in the cortex does notcontribute to medullary hypertonicity, so that,whereas a reduction of sodium transport in thiscortical segment will be reflected in a reduction ofsolute-free water clearance CH2o,* it will not reducesolute-free water reabsorption TCH2O,* by the col-lecting duct which depends on the hypertonicityof the medulla.

DISTAL CONVOLUTED TUBULE ANDCOLLECTING DUCTSodium reabsorption in this part of the nephronis also an active process and may be accompaniedby chloride reabsorption or alternatively coupledto potassium and hydrogen ion secretion. Theactivity of these distal tubular exchange mechanismsis controlled to a large extent by aldosterone, whosemodulating action involves the synthesis of mineralo-corticoid-induced proteins capable of stimulatingthe sodium pump located at the peritubular borderof distal tubular cells. Permeability to sodium atthe luminal border of these cells and those of thecortical collecting ducts may also be increased.The trans-tubular electrical potential difference

in the distal tubule is negative within the lumen andlargely generated by the active reabsorption ofsodium. It amounts to about -10 mV in the earlypart ofthe tubule and reaches -45 mV more distally;it is increased further by the presence of poorlyreabsorbable anions within the tubule.

In common with most other cells distal tubularcells are rich in potassium, and a chemical gradientexists for diffusion out of the cells across both the

*The free water clearance, CH,O, is defined as the volume ofurine excreted in excess of that required to excrete the soluteiso-osmotically with plasma. If the urinary solute concentra-tion and volume are Uosm and V respectively, the totalsolute content = Uosm x V, and the volume required toexcrete this iso-osmotically is

Uosm x V= the osmolal clearance Cosm.POSM

.CH. =O V- CosmIf there is net reabsorption of solute-free water, CH,O isnegative; it is therefore expressed as

TcH,o = -CH,o = Cosm - V(Eds)

Lant

luminal and peritubular membranes. Distal tubularpotassium secretion is largely passive and followsa net electrochemical gradient generated by theactive uptake of sodium, and probably also ofpotassium, from the luminal border of the cell.The luminal membrane possesses an active secretorypump for hydrogen ions which is fed by the intra-cellular hydrogen ion pool. This pool is replenishedby several processes working in parallel, includingcatalysed and uncatalysed hydration of C02 withinthe cell, backdiffusion of CO2 from the lumenand uptake of hydrogen ions from the peritubularspace. Any diuretic acting proximal to the aldoster-one-sensitive ion-exchange sites causes an increaseddelivery of sodium to the distal tubule and therebyleads to an increase in urinary loss of potassium.The lining epitheliumofthelater part of the distal

tubule gradually merges with that of the corticalpart of the collecting duct where ADH-responsive-ness becomes a feature. The latter mechanisminvolves a stereospecific receptor at the basal orblood side of the tubular cell, a modulator whichreceives positive or negative signals-for example,PGE1, to be passed on to the catalytic step, phospho-diesterase, which converts cyclic AMP to 5'AMP.The cyclic nucleotide diffuses through the cyto-plasm to reach the effector site-the luminal plasmamembrane.

Sites of action of diuretic drugs in the nephron17 18 19

Many techniques have been employed to localisethe sites of action of diuretics in the nephron.These include various in vitro approaches such asstop-flow, free-flow micropuncture, autoradiographyand single nephron microperfusion. The use ofdifferent species has made it difficult to interpretthe results of such experiments and evidence fromin vitro investigation may not be directly applica-ble to man. Understandably in vitro methods havebeen much more limited in scope but have neverthe-less provided valuable evidence on the possiblesites of action in the human nephron.

If it is recalled that of total sodium reabsorptionin the nephron, approximately 65% takes place inthe proximal tubule, 25% in the loop of Henle,8-9% in the distal tubule and the remaining 1-2%in the collecting duct, the maximal natriureticresponse to a diuretic can give a clue to its site ofaction (Table). Similarly the pattern of excretionof anions and cations evoked by diuretics can berelated to the known functional characteristicsof specific portions of the tubule.

Probably the most informative in vivo techniquehas been the study of diuretic effects on the mech-



j.com/

J Clin P


ber 1981. Dow

nloaded from

http://jcp.bmj.com/


Table Classification of diuretics

Group Predominant site ofaction*

High efficacy diuretics ( > 15%)Organomercurials Medullary dilutingEthacrynic acid segmentFrusemide (Furosemide) (site 11)HumetanidePiretanideMuzolimineMedium efficacy diuretics (5-10%)Chlorothiazide and Thia7ide Family Cortical diluting

segment(site III)

PhthalimidinesChlorthalidone, Clorexolone

QuinazolinonesQuinethazoneMetolazone

IlenzenesulphonamidesMefruside

ChlorobenzamidesClopamide

SalicylamidesXipamide

Weak or adjunct diuretics (< 5 %)Xanthines Glomerular

Aminophylline arteriolardilatation

Carbonic anhydrase inhibitors Proximal tubuleAcetazolamide (site I)

Osmotic agents Proximal tubuleMannitol (site I)

Potassium-sparing compounds(a) Aldosterone antagonists Distal

Spironolactone and canrenoate Na+/K+H+(b) Pteridines and pyrazine- Exchange

carboxamides-Triamterene and (site IV)Amiloride

Numbers in parentheses are the maximal fractional excretion ofsodium (sodium excretion expressed as a percentage of the sodiumfiltered).*In certain instances, experimental evidence has been adduced forsecondary sites of diuretic action additional to the major locus withinthe nephron-for example, frusemide and piretanide may also exertsome effects upon the proximal site I, as may the thiazide-like com-pound, metolazone.

anisms for urinary concentration and dilution. Theresults of such investigations have permitted in-ferences to be drawn on the inhibitory effects ofvarious diuretics on electrolyte transport withinthe medullary and cortical diluting segments of theloop of Henle. Four main tubular sites can beidentified as of importance in the action of diureticsin the kidney (Fig. 1).

In view of the substantial reabsorptive capacityof the proximal tubule, it might at first sight appearto be useful to produce diuretics whose main locusof action lay proximally. However, those agentswhich do seem to work in the proximal tubuleare either too weak to be effective alone-forexample, inhibitors of carbonic anhydrase, orinconvenient, because of the need for intravenousadministration-forexample, mannitol. Furthermore,a major disadvantage of carbonic anhydrase block-ade is that by inhibiting proximal bicarbonate

reabsorption, a metabolic acidosis is engenderedwhich has been found to limit the diuretic effect.There are other reasons why direct effects of diureticdrugs on the proximal tubule may be obscured oreven annulled. When proximal tubular reabsorptionof sodium chloride is blocked, compensatoryincreases in reabsorption further down the nephronmay follow so that little additional saluresis occursin the final urine. The reserve reabsorptive capacityof the diluting segments is considerable and canovershadow effects occurring in more proximalparts of the nephron. By contrast, a diuretic havinga primary effect on the medullary diluting segmentmay lead to the excretion of a substantial amountof excess salt. The main reason for this is the limitedreabsorptive capacity for salt of the distal tubuleand collecting duct. The latter explains whydiuretics acting primarily in the distal tubule suchas the potassium-sparing compounds, evoke onlya modest saluretic effect amounting, at most, toonly 5% of the filtered load of sodium (see Table).

Subcellular mechanisms of diuretic drug action'9 20

With the exception of carbonic anhydrase inhibitors,the biochemical basis for the action of diureticsis still largely unknown. There is no doubt that theavailability of diuretically active substances hasstimulated considerable interest in the physiologyand biochemistry of nephron function and haspermitted the characterisation of intricacies oftubular function which had hitherto been un-recognised. Such studies have also emphasised,however, the unique anatomical and functionalrelations within the kidney which make it such adifficult organ to investigate experimentally.

Because tubular epithelium shares certain trans-port characteristics with other cellular systems,simpler tissues have been utilised to study themechanisms of diuretic action with a view toexplaining the latter in biochemical terms. This isthe reason why such an assortment of non-renaltissues as-for example, amphibian skin, urinarybladder, blastocyst and corneal membranes havebeen used. Whilst undoubted parallels can be drawnfrom the findings in such diverse tissues, it isimportant that direct extrapolation to the functionof the intact human kidney be made with caution.What may be demonstrable in isolated tissues inone species may be drastically altered by the manyextra- and intrarenal mechanisms known to beoperative in man. Despite these criticisms, the useof diuretics as "membrane probes" has yieldedfascinating insights into some of the mechanismswhereby solutes are transferred across cell mem-branes. Numerous questions about the nature of

1271



j.com/

J Clin P


ber 1981. Dow

nloaded from

http://jcp.bmj.com/

1272

the "cell receptors" for diuretic drugs remainunanswered and await further exploration.

STRUCTURE :ACTIVITY RELATIONS21 22The history of the evolution of diuretic drugs revealsan interesting sequence based on chance observa-tions of unwanted effects. Although the diureticeffects of mercury compounds were known toParacelsus in the sixteenth century, it was Voglwho, as a medical student in Vienna in 1921,observed diuresis in a patient being treated with anorgano-mercurial for syphilis and paved the wayfor the development of mercurial diuretics. By theearly 1940s, it was known that various phenyl-and heterocyclic- substituted sulphonamides in-hibited carbonic anhydrase and caused a natriuresiswith the passage of alkaline urine but at the expenseof inducing a metabolic acidosis. A major milestonein the development of the sulphonamide diureticswas the discovery of chlorothiazide in 1957. Subse-quent molecular manipulation of the hydrogenatedbenzothiadiazine structure yielded many differentthiazide and thiazide-like agents with progressivelydwindling capacity to inhibit carbonic anhydrase(Fig. 2). Then, by opening the thiadiazine ring, and

HAcCHN sS.O/S02N

0 IIN-N

Acetazolamide

Rs NH,3}6#fCH-R3

H2N-02S 2

Thiazide group

Fig. 2 Structural relation between a classical carbonicanhydrase inhibitor, such as acetazolamide, and the]benzothiadiazine (thiazide) group of diuretics. Thethree main positions for substitution in the thiazidemolecule are depicted as R2, R3 and R6, the latterinvariably consisting ofa halogen (eg, Cl) ortrifluoromethyl (eg, CF3) radical.

exploring derivatives of anthranilic acid, frusemideemerged in the early 1960s (Fig. 3). The discoveryof ethacrynic acid, on the other hand, followed aquite separate avenue of research, in that it rep-resented an orally effective "high ceiling" diuretic*which bound to sulphydryl groups in renal tissuebut did not contain any mercury.

It might, at first sight, appear that the biochemicalmechanisms of actions of diuretics would be

*The term "high ceiling" is applied to diuretics which acton the loop of Henle because the dose/response curvescontinue to rise, even with very large doses, unlike theesponse to thiazides, which reaches a plateau.

CH-CH

CL NH-, C1

CCH2-C CH

~~~~~0H2N 02S COOH

Frusemide

NH-(CH CH3

H2NB02 d COOH

Bumetanide

Fig. 3 Comparison of two sulphamoyl loop diuretics,frusemide and bumetanide. Both are 5-sulphamoyl-carboxylic acids, but the change from the 4-chlorosubstituent in frusemide to a phenoxy group, as inbumetanide, results in a substantial increase in diureticpotency on a weight for weight basis.

especially straightforward in the case of the sul-phamoyl radical (SO2-NH2). The converse is thecase. Without doubt, the target enzyme for drugslike acetazolamide or dichlorphenamide is carbonicanhydrase, but precisely how the natriuretic effectsof carbonic anhydrase inhibition are achieved isnot clear. The evidence for involvement of a bi-carbonate-activated ATPase coupled to sodiumtransport remains uncertain. Although all thiazideand thiazide-like diuretics possess a free sulphamoylgroup, the hydrogenated derivatives which followedchlorothiazide have negligible carbonic anhydraseinhibitory capacities yet are potent saluretic agents.To date, no consistent biochemical mechanism ofaction has been worked out for this large group ofdiuretics. When high ceiling compounds such asfrusemide and ethacrynic acid are compared interms of their inhibitory effects on respiration andglycolysis in renal tissue it is again disappointingto find that natriuretic activity does not necessarilycorrelate with metabolic effects. For example, theconcentration of diuretic needed to inhibit glycolysisin vitro may be higher than that which could beachieved in vivo within the renal tubular cells.

RENAL ION PUMPS AND DIURETIC ACTION1018An impressive body of evidence now exists to showthat the Na, K-ATPase of most cells representsor is an integral part of the sodium pump. Theenzyme is found in high concentrations in the kidneyand especially in the ascending limb of the loop ofHenle. Not surprisingly, Na, K-ATPase has beenconsidered a potential candidate as the cell "re-ceptor" for diuretics, especially high ceiling "loop"compounds. Yet a number of important observationsconflict with such a view. Although loop diureticsinhibit glycoside-sensitive Na, K-ATPase, they alsoinhibit that part of sodium transport that persistseven in the presence of full doses of the cardiacglycoside, ouabain. The diuretic effects of "loop"compounds has been ascribed to blockade of active



j.com/

J Clin P


ber 1981. Dow

nloaded from

http://jcp.bmj.com/


chloride reabsorption in the medullary dilutingsegment, an action shared with organomercurials.The driving force for active chloride transport inrenal tissue has not been fully defined nor has therelation of Na, K-ATPase or other Mg-dependentATPases to chloride as opposed to sodium transport.A cyclic AMP-activated Na, K cotransport systemhas been described in avian erythrocytes which isinhibited by loop compounds, whilst a chlorideself-exchange system in human erythrocytes hasalso been found to be inhibited by these diuretics.The relation between these transport systems inhuman red cells and the renal tubule remains to beclarified.23 24

It is of considerable interest to note that aboutthirty years ago in the era when organomercurialdiuretics reigned supreme, it was proposed thatthese compounds worked primarily by inhibitingchloride transport. This conclusion was encompassedin the view that "the relationship to plasma chlorideis the significant one, and that if mercury specificallyblocks one ion absorptive mechanism, that mech-anism is the one for chloride absorption. Increasedsodium excretion following mercurial diuretics isthus a more or less passive consequence of increasedchloride elimination."25 In retrospect, clearly themajor role of chloride transport in the action ofdiuretics has been overshadowed over the years.Yet, it remains likely that the "uphill" reabsorptivetransport of chloride in the loop of Henle may rep-resent a transport process tied to the Na, K-ATPase.This means that translocation of chloride acrossmembranes occurs by utilising the energy of thesodium gradient. This would still imply a degree ofprimacy in the operation of the electrogenic sodiumpump.26 Vanadate occurs naturally in cells throughoutthe body and has been found to inhibit the sodiumpump reversibly from the cytoplasmic side of thecell membrane, in contrast with cardiac glyco-sides such as ouabain which bind to the exteriorof cells.27 Infusion of vanadate can induce saluresisby inhibition of tubular reabsorption and it hasbeen proposed that vanadate may have a modulatingrole on the performance of renal Na, K-ATPase innormal as well as pathological states. The relationif any, between pump modification by vanadateand the cellular actions of diuretics remains obscure.

POTASSIUM-SPARING DIURETICS ANDPASSIVE SODIUM-ENTRY CHANNELS2829The use of non-renal tissues in the study of diureticdrug action has paid off most handsomely in thecase of the potassium-sparing drugs, where mostwork on the mechanisms has been carried out onisolated amphibian epithelia. In the case of spiro-nolactone, the close chemical similarity with

24lAmiloride 12jugwash with Ringer I(x2)

Percentagechange inshort circuitcurrent

20

0.

-20

-40*

-60

-80

0 10 20 30 40 50 60Time ( min)

7fl 80

Fig. 4 Effects of 12 and 24 pzg Amiloride (8 and16 x 10-7M) added to the Ringer solution bathing theexternal surface of isolatedfrog skin. The immediatefall in short circuit current is readily reversed by washingthe skin surface with fresh Ringer solution (based onBaba et a128).

aldosterone made it likely that diuretic effects aremediated by competitive binding to the mineralo-corticoid-sensitive binding protein in cells locatedwithin the distal convoluted and cortical collectingtubules. Triamterene and amiloride act by blockingthe passive entry of sodium into transporting cells inthese same regions of the nephron. The evidencefor this has been obtained for example in experi-ments where diuretics have been added to the outsidesurface of frog skin (Fig. 4) or the epithelial surface ofthe apex of the toad bladder. Triamterene and amilo-ride are unique among diuretics in blocking sodiumentry mechanisms in a rapid and easily reversiblemanner. Both compounds have been extensivelyemployed as membrane probes to investigatesodium-selective channels in a variety of transport-ing epithelia.

NEW GENERATION DIURETICS3031The lack of any easily definable relation betweenchemical structure and mechanism of action hasbeen emphasised clearly in the sulphonamideseries of diuretics. At least three different modes ofoperation occur: classical carbonic anhydraseinhibition, benzothiadiazine action and high ceilingaction. It is often questioned whether there is anymerit in further work involving molecular manipula-tion within a drug family, when apparently all theuseful variations in substitution have been carriedout. Experience teaches us that the unpredictablecan always emerge even from such mundaneexercises. Thus, for example, the unexpected diureticprofile of frusemide emerged from molecular play

-luu

1273



j.com/

J Clin P


ber 1981. Dow

nloaded from

http://jcp.bmj.com/

1274

05,

C2H,--C C O-CH,-COOHCHP0

Ethocryr ic cicid

C O-CH COOH

Tienilic acid fTi 'ynrfen)

CHOO-CH - 16H,

Indocrinone (MK 196i

Fig. 5 Structural relation between ethacrs'nic acidand two polyvalent phenoxyacetic acid derivatives,tienilic acid and indacrinone, both capable of causingsimultaneous saluresis and uricosuria. Indacrinone exi.st.sas a racemic mixture, isomerism occurlring at the 2position of the indanone rting.

with the well-explored benzothiadiazine series.Subsequent high ceiling developments which havefollowed include bumetanide and piretanide. Mole-cular manipulation of the ethacrynic acid moleculehas led to the emergence of a new class of phenoxy-acetic acid derivatives possessing an unusualcombination of saluretic and uricosuric properties.The first of these so-called polyvalent diureticswas tienilic acid (tricrynafen) which unfortunatelyran into major toxicological problems early in itsclinical application to antihypertensive treatment.Development of human hepatotoxicity led to thewithdrawal of tienilic acid from clinical use by itsmanufacturers in 1980. A further phenoxyaceticacid derivative, indacrinone lacks the thienylsubstituent and has the unusual feature of stereo-isomerism, each of its constituent enantiomerspossessing different pharmacological profiles inrelation to saluresis and uricosuria.32

Availability of this new generation of compoundshas offered a means of investigating in greaterdepth the mechanisms of uric acid handling by thekidney and how these are affected by diuretics.

Current concepts of renal urate handling supportthe view that uric acid may undergo bidirectionaltransfer via one single transport system located atvarious sites along the nephron.33 Such a mechanismis analogous to the forwards and backwards move-

ment of sodium and potassium ions which has beendemonstrated to occur through the conventionalsodium pump.34 Disturbance in the relative balancebetween the two processes of secretion and absorp-tion of uric acid would be responsible for the pre-

dominance of either hyperuricaemia or uricosuria.Renal urate regulation probably involves a "four-component" system, namely, filtration, bidirectional

-6E

0 04

Lno2

0310

201

2 4 6Collection time(h)

-1 1-2 2-4 4-6 6-8 8-12 12-24

Collection periods (h)

Fig. 6 Changes in handling of luic acid in iiorinal ,,,aliaifter a single close of iacemic indacrinone. nig.The top half of the figure shows the plasma urcateconcentrations on control da -)anid dr}ug daX

A significant fall occur.s at 1, 2 and 4 h afterindacrinone conipared to control. This parallel.s a

Significant increase in fractional Ienal uric acid clearanicmeasured as Eurate Cturate/Cre(itwinei x1X/00. At tim7eofpeak uiricosuria (2-4h) Es a increased frotn 0.55 + 0.09

(control) to 2.92 + 0.30 (p < 0.001)L control day; dr-ug dayResults are expr-essed as Iiieani values + SEM (ti ')(based on Brooks et a132).

transtubular fluxes and also postsecretory reabsorp-tion. The latter process is of particular importancein accounting for renal urate retention associatedwith chronic thiazide administration and occurs

secondarily to diuretic-induced ECF volume con-

traction.

References

' Levinsky NG. Some aspects of the regulation of sodiuln.excretion. In: Fisher JW, ed. Renal pharmacologi.London: Butterworths, 1971:85-98.

2 Blythe WB. Regulation of sodium excretioni. In: Jones NF,ed. Recent advanccs in renal disease. London: ChurchillLivingstone, 1975:322-49.

Earley LE. Hemodynamic regulation of sodium andwater absorption by epithelial membranes. In: Lant AF,Wilson GM, eds. Modern diuiretic therapy, in the treat-

mnent of cardiovascular and renal dlisease. Amsterdam:

'Thr

8 24rLant



j.com/

J Clin P


ber 1981. Dow

nloaded from

http://jcp.bmj.com/


Excerpta Medica, 1973:11-25.4 Laragh JL, Sealey JE, Brunner HR. Humoral control of

renal sodium excretion. In: Lant AF, Wilson GM, eds.Modern diuretic therapy in the treatment ofcardiovascularand renal disease. Amsterdam: Excerpta Medica,1973:26-39.

Dzau VJ, Colucci WS, Hollenberg NK, Williams GH.Relation of the renin-angiotensin-aldosterone systemto clinical state in congestive heart failure. Circulation1981 ;63:645-51.

6de Wardener H. The control of sodium excretion.Am J Physiol 1978;235:163-73.

Hook JB, Bailie MD. Release of vasoactive materialsfrom the kidney by diuretics. J Clin Pharmacol 1977:17 :673-80.

8 McGiff JC, Wong P Y-K. Prostaglandins and renalfunction: implications for the activity of diureticagents. In: Cragoe EJ, ed. Diuretic agents. Washington:American Chemical Society, 1978:1-11.

9 Thurau K, Mason J. The intrarenal function of the juxta-glomerular apparatus. In: Thurau K, ed. MTP Inter-national Review of Science Vol 6, Kidney and urinarytract physiology. London: Butterworths, 1974:357-89.

10 Sachs G. Ion pumps in the renal tubule. Am J Physiol1977 ;233:F359-65.

Maack T, Windhager EH. Electrolyte transport in thenephron. In: Black D, Jones NF, eds. Renal disease4th ed. Oxford: Blackwell Scientific Publications,1979:107-138.

12 Kokko JP, Rector FC. Countercurrent multiplicationsystem without active transport in inner medulla.Kidney Int 1971 ;2:214-23.

13 Valtin H. Structural and functional heterogeneity ofmammalian nephrons. Am J Physiol 1977;233:F491-501

14 Kuhn W, Ryffel K. Herstellung konzentrierter Losungenans verdunnten durch blosse Membranewirkung:ein Modellversuch zur Funktion der Niere. Z PhysiolChem 1942 ;276:145-78.

15 Wirz H, Hargitay B, Kuhn W. Lokalisation des konzen-trierungsprozesses in der Niere durch direkte kryoscopie.Helv Physiol Pharmacol Acta 1951 ;9:196-207.

16 Bisset GW, Jones NF. Antidiuretic hormone. In: Jones NF,ed. Recent advances in renal disease. London: ChurchillLivingstone, 1975:350-416.

17 Lant AF, Wilson GM. Diuretics. In: Black DAK, ed.Renal disease 2nd ed. Oxford: Blackwell ScientificPublications, 1972:594-637.

18 Burg MB. Tubular chloride transport and the mode ofaction of some diuretics. Kidney Int 1976;9:189-97.

19 Lant AF. Relief of oedema and the action of diuretics.In: Vere DW, ed. Topics in therapeutics. London: PitmanPublishing, 1978:150-66.

20 Jacobson HR, Kokko JP. Diuretics: sites and mechanismsof action. Ann Rev Pharmacol 1976;38:201-14.

21 Cragoe EJ, Jr. ed. Diuretic agents. American ChemicalSociety Symposium 83. Washington: American ChemicalSociety, 1978.

22 Feit PW. Diuretics. In: Encyclopaedia of chemical tech-nology 3rd ed. Vol 8. London: John Wiley & Sons, 1979.

23 Brooks BA, Lant AF. Use of human erythrocyte as amodel for studying the action of diuretics on Na+and Cl- transport. Clin Sci 1978;54:679-83.

24 Palfrey HC, Feit PW, Greengard P. cAMP-stimulatedcation cotransport in avian erythrocytes: inhibitionby "loop" diuretics. Am J Physiol 1980;238:C139-48.

21 Axelrod DR, Pitts RF. The relationship of plasma pHand anion pattern to mercurial diuresis. J Clin Invest1952;31 :171-9.

26 Epstein FH, Silva, P, Kormanik G. Role of Na-K-ATPasein chloride cell function. Am J Physiol 1980;238:R246-50.

27 Grantham JJ. The renal sodium pump and vanadate.Am J Physiol 1980;239:F97-106.

28 Baba WI, Lant AF, Smith AJ, Townshend MM, WilsonGM. Pharmacological effects in animals and normalhuman subjects of the diuretic, amiloride hydrochloride(MK 870). Clin Pharmacol Ther 1968;9:318-27.

29 Cuthbert AW. Aspects of the pharmacology of passiveion transfer across cell membranes. In: Ellis GP,West GB, eds. Progress in medicinal chemistry 14.Amsterdam: Elsevier, 1977:2-50.

30 Woltersdorf OW, Jr, de Solms SJ, Cragoe EJ, Jr. Theevolution of the (aryloxy) acetic acid diuretics. In:Cragoe EJ Jr, ed. Diuretic agents. Washington: AmericanChemical Society,1978:190-230.

31 Lemieux G, Steele TU, eds. Symposium on antihyperten-sive uricosuric diuretics. A new class of renally activecompounds with antihypertensive, diuretic, uricosuricproperties. Nephron 1979;23 Suppl 1.

32 Brooks BA, Blair EM, Finch R, Lant AF. Studies on themechanism and characteristics of action of a uricosuricdiuretic, indacrinone (MK-196). Br J Clin Pharmacol1980;10:249-58.

33 Weiner IM. Urate transport in the nephron. Am J Physiol1979;237:F 85-92.

34 Lant AF, Priestland RN, Whittam R. The coupling ofdownhill ion movements associated with reversal ofthe sodium pump in human red cells. J Physlol (Lond)1970;207:291-301.

Requests for reprints to: Prof AF Lant, Department ofTherapeutics, Page Street Wing, Westminster Hospital,London SWIP 2AP, England.

1275



j.com/

J Clin P


ber 1981. Dow

nloaded from

http://jcp.bmj.com/

modern diuretics and the kidney - journal of clinical ... · changes in physical forces within the...

Documents