Late-Onset Manifestation of Antenatal Bartter Syndrome as a Result of Residual Function of the Mutated Renal Na+-K+-2Cl– Co-Transporter
Carsten A. Pressler*,
Jolanta Heinzinger*,
Nikola Jeck*,
Petra Waldegger*,
Ulla Pechmann*,
Stephan Reinalter*,
Martin Konrad,
Rolf Beetz,
Hannsjörg W. Seyberth* and
Siegfried Waldegger*
* Department of Pediatrics, Philipps University of Marburg, Marburg, Germany; Department of Pediatrics, Inselspital, Bern, Switzerland; and Department of Pediatrics, Johannes-Gutenberg University of Mainz, Mainz, Germany
Address correspondence to: Prof. Siegfried Waldegger, Department of Pediatrics, Philipps University of Marburg, Baldingerstrasse, 35033 Marburg, Germany. Phone: +49-6421-2862649; Fax: +49-6421-2865724; E-mail: siegfried.waldegger{at}staff.uni-marburg.de
Received for publication October 17, 2005.
Accepted for publication May 21, 2006.
Genetic defects of the Na+-K+-2Cl– (NKCC2) sodium potassiumchloride co-transporter result in severe, prenatal-onset renalsalt wasting accompanied by polyhydramnios, prematurity, andlife-threatening hypovolemia of the neonate (antenatal Barttersyndrome or hyperprostaglandin E syndrome). Herein are describedtwo brothers who presented with hyperuricemia, mild metabolicalkalosis, low serum potassium levels, and bilateral medullarynephrocalcinosis at the ages of 13 and 15 yr. Impaired functionof sodium chloride reabsorption along the thick ascending limbof Henle’s loop was deduced from a reduced increase indiuresis and urinary chloride excretion upon application offurosemide. Molecular genetic analysis revealed that the brotherswere compound heterozygotes for mutations in the SLC12A1 genecoding for the NKCC2 co-transporter. Functional analysis ofthe mutated rat NKCC2 protein by tracer-flux assays after heterologousexpression in Xenopus oocytes revealed significant residualtransport activity of the NKCC2 p.F177Y mutant construct incontrast to no activity of the NKCC2-D918fs frameshift mutantconstruct. However, coexpression of the two mutants was notsignificantly different from that of NKCC2-F177Y alone or wildtype. Membrane expression of NKCC2-F177Y as determined by luminometricsurface quantification was not significantly different fromwild-type protein, pointing to an intrinsic partial transportdefect caused by the p.F177Y mutation. The partial functionof NKCC2-F177Y, which is not negatively affected by NKCC2-D918fs,therefore explains a mild and late-onset phenotype and for thefirst time establishes a mild phenotype-associated SLC12A1 genemutation.
The central role of the kidney in maintaining body salt andwater homeostasis depends on the integrity of a series of transepithelialtransport processes that are located throughout the nephron.Vectorial transepithelial salt reabsorption along the loop ofHenle is crucial for the generation of a high interstitial osmolality,which draws back water from the tubular fluid that passes themedullary collecting ducts. The key player in the process ofinterstitial salt accumulation is the Na+-K+-2Cl– co-transporter(NKCC2), which catalyzes the apical entry step of sodium chloridefrom the lumen into epithelial cells of the thick ascendinglimb (TAL). Potassium shunts back into the tubular fluid viaapical renal outer medullary potassium channels (ROMK), whereassodium and chloride are extruded basolaterally into the medullaryinterstitium via sodium/potassium ATPases and ClC-K–typechloride channels, respectively. In concerted action, thesetransport processes not only accumulate sodium chloride in themedullary interstitium but also confer positive electrical chargeto the luminal side of the TAL epithelium. Along the TAL, thistransepithelial voltage gradient forces out divalent cationsfrom the lumen via calcium- and magnesium-selective paracellularpathways. The prominent importance of the NKCC2 co-transporterin this machinery is evidenced by the effects of loop diuretics,which, as pharmacologic NKCC2 inhibitors, are the most potentagents available to increase urinary water, salt, and calciumexcretion. Even more impressive, genetic mutations of the NKCC2encoding gene SLC12A1 are associated with antenatal Barttersyndrome (hyperprostaglandin E syndrome [aBS/HPS]) (1). Prenatal-onsetrenal salt and water wasting in this autosomal recessive disorderleads to polyhydramnios entailing extreme prematurity. Afterbirth, life-threatening volume depletion, hypokalemia, and hypochloremicmetabolic alkalosis come along with hypercalciuria and medullarynephrocalcinosis. Without appropriate treatment, patients withaBS/HPS will not survive the early neonatal period (2). In congruencewith the severity of the symptoms and the uniformity of theclinical picture, functional analysis of diverse NKCC2 mutantsconsistently revealed a complete loss-of-function effect ofthe tested mutations (3).
Here we describe the case of two siblings who presented withhypokalemia, hyperuricemia, and medullary nephrocalcinosis atthe ages of 13 and 15 yr. Molecular genetic analyses revealedthat the brothers were compound heterozygotes for mutationsin the SLC12A1 gene. Because one of the mutations only partiallyimpaired NKCC2 function, variable degrees of NKCC2 dysfunctionmay be associated with variations of the clinical picture. Thisis the first report of patients with a mild and late-onset phenotypeassociated with SLC12A1 gene defects and indicates that partiallyimpaired NKCC2 function should be considered even beyond theneonatal period in patients who present with symptoms of renalsalt wasting.
In Vivo Assessment of TAL Function
Diuresis and urinary chloride excretion in response to oralfurosemide application (2 mg/kg body wt) was determined essentiallyas described by Koeckerling et al. (4). In brief, urine volumeand urinary chloride excretion during 3 h were determined atthe same time on 2 consecutive days under standard diet andfluid intake. The first day served as the control period. Onthe second day, urine collection during 3 h was performed immediatelyafter a single oral dose of furosemide. The difference in urinevolume and chloride excretion between day 1 and day 2 (V, Cl)was considered to be induced by furosemide administration. Apossible bias by variable furosemide concentrations was excludedby normalization of V and Cl to urinary furosemide excretion.
SLC12A1 Sequence Analysis
Informed consent was obtained from the patients and their parentsbefore blood samples were taken for genomic DNA isolation. All26 exons together with adjacent intronic sequences of the SLC12A1gene (1) were PCR amplified from genomic DNA according to standardprotocols. PCR products with aberrant electrophoretic mobilitydetermined by single-strand conformational polymorphism analysiswere sequenced in both directions. All nucleotide variationsobserved were confirmed from independent PCR and DNA samplesfrom 50 healthy, unrelated white individuals and served as normalcontrols.
Cloning of the NKCC2 Gene from Rat Kidney
Freshly prepared rat kidney tissue was homogenized in cold Trizolreagent (Invitrogen, Carlsbad, CA; www.invitrogen.com), andtotal RNA was prepared according the manufacturer’s recommendations.After mRNA purification with oligo-dT–coupled beads (OligotexmRNA purification system; Qiagen, Valencia, CA; www.qiagen.com),reverse transcription to cDNA was performed with SuperScriptreverse transcriptase (Invitrogen). We used a proofreading polymerase(pfu-Polymerase; Stratagene, La Jolla, CA; www.stratagene.com)to amplify the complete open reading frame of the NKCC2 cDNA(NCBI GenBank accesion no. XM_579419) with the primers (in 5'-3'direction) TGG AAG ATG TCA GTC AAC ATC CCT TCC(5'-end) and TCCCT CGC GCT TTA AGA GTA AAA TGT C (3'-end; the start ATG andthe reverse complementary stop TTA are indicated in bold). Theresulting PCR product was cloned into pCRII-Topo (Invitrogen)and sequenced in both directions. To optimize translation efficacyin Xenopus oocytes, we then transferred the verified cDNA sequenceto pOGII, which contains the 5'- and 3'- untranslated regionsof the Xenopus-globin gene. Site-directed mutagenesis was performedaccording to the QuickChange protocol (QuickChange Site DirectedMutagenesis Kit; Stratagene).
Expression in Xenopus laevis Oocytes and Tracer Flux Measurements
Ten or 20 ng of in vitro transcribed rat cRNA (mMessage mMachinekit; Ambion, Austin, TX; www.ambion.com) for the NKCC2 constructswere injected in defolliculated Xenopus oocytes, which werekept at 16°C in ND96 storage solution that contained 96.0mmol/L NaCl, 2.0 mmol/L KCl, 1.8 mmol/L CaCl2, 1.0 mmol/L MgCl2,5.0 mmol/L HEPES (pH 7.4), 2.5 mmol/L sodium pyruvate, 0.5 mmol/Ltheophylline, and 20 µg/ml gentamicin. Four days afterinjection, 36Cl uptake was determined in chloride-free ND96uptake solution (which contained [in mmol/L] 96.0 Na-aspartate,80.0 D-Mannitol, 2.0 KOH, 1.8 Ca-acetate, 1.0 Mg-acetate, and5.0 HEPES [pH 7.4]) supplemented with 5.0 mmol/L Na36Cl. Afterequilibration for 15 min in 36Cl-free uptake solution, the tracerwas added and cells were kept in uptake solution at room temperaturefor 60 min, then cooled on ice and washed five times in ice-coldchloride-free ND96 (without tracer). After mechanical lysis,intracellular radioactivity was determined by scintillationcounting for each oocyte. Mean and SEM of the determined countsper minute (cpm) are shown for at least 15 oocytes per datapoint from the same preparation. Experiments were repeated inat least three different batches of oocytes that were derivedfrom different frogs. The error bars in the diagrams were calculatedfrom the SEM. Statistical significance was analyzed with t test(unpaired test with unequal variance) and was assumed at a P 0.05.
Quantification of NKCC2 Surface Expression
For quantification of wild-type and mutated NKCC2 surface expression,a hemagglutinin (HA) epitope was inserted into the extracellularloop between the transmembrane domains VII and VIII (as predictedby hydropathy analysis). As determined by tracer-flux experiments,the extracellular HA epitope did not interfere with 36Cl uptakeactivity (data not shown). Surface quantification of NKCC2-HAconstructs after expression in Xenopus oocytes was performedas described previously (5). In brief, viable NKCC2-HA–expressingoocytes were labeled with rat anti-HA mAb (3F10, 1 µg/ml).After rigorous washing in cold ND96 solution, bound antibodieson individual oocytes were detected with horseradish peroxidase–coupledgoat anti-rat antibodies (Fab fragments, 1:500 dilution; JacksonImmunoResearch, West Grove, PA) in a peroxidase-catalyzed luminescencereaction using SuperSignal ELISA Femto Maximum Sensitivity chemiluminescentsubstrate (Pierce, Rockford, IL). The luminescence signal asdetermined by a luminometer (Berthold Lumat LB9507; BertholdTechnologies, Bad Wildbad, Germany) is given in relative lightunits. Oocytes that expressed wild-type NKCC2 without HA epitopeserved as control. The experiments were repeated in three differentbatches of oocytes that were derived from different frogs. Foreach data point, approximately 20 oocytes that were derivedfrom one single preparation were analyzed individually, andthe relative light units are given as mean ± SEM.
Patient 1 is the first of two sons of healthy nonconsanguineousparents. He was born 3 wk past term after a normal pregnancyand showed normal thriving. The clinical and laboratory dataare summarized in Table 1. A routine check at the age of 15yr in preparation for a tonsillectomy revealed hyperuricemia(0.76 mmol/L; normal range 0.18 to 0.42 mmol/L), hypokalemia(2.5 mmol/L plasma potassium), and metabolic alkalosis (venousblood pH of 7.52) with a bicarbonate concentration of 30.7 mmol/L(normal range 21 to 28 mmol/L). Plasma sodium, calcium, andmagnesium concentrations were normal, as was plasma creatinine.A more elaborate workup that was initiated by these findingsshowed elevated serum renin (50.5 pg/ml) and aldosterone levels(865 pg/ml). Urinalysis after 24-h urine collection revealedmild polyuria (2850 ml/24 h or approximately 30 ml/kg per 24h) with a urine osmolality of 248 mOsmol/kg. Fractional sodiumexcretion was increased (4.7%). Moreover, mild hypercalciuriawith 107 µmol/kg body wt per 24 h calcium excreted anddecreased uric acid clearance (2.7 ml/min) were noted. Urinaryexcretion of prostaglandin E2 (PGE2) was normal (19.4 ng/h per1.73 m2 body surface area); that of prostaglandin EM (PGE-M)was slightly increased (644 ng/h per 1.73 m2). Sonographic evaluationof the kidneys revealed bilateral medullary nephrocalcinosis.Therapy was started with an oral potassium supplementation (1mmol/kg body wt) and allopurinol (3.5 mg/kg body wt).
Table 1. Clinical and laboratory data of both patients (at the initial evaluation) and their parentsa
Patient 2, the younger brother of patient 1, was born at termafter an uncomplicated pregnancy. He thrived well until theage of 5 yr, when for the first time a reduced growth rate wasnoted. A diagnostic evaluation because of overt growth delay(height –2.1 SD score; weight –2.3 SD score) wasperformed at the age of 13 yr and revealed hypokalemia (3.2mmol/L) with normal values for plasma sodium, calcium, and magnesium.Compensated metabolic alkalosis with base excesses ranging from5.1 to 7.7 mmol/L was observed on several occasions in follow-upexaminations. Creatinine was normal, renin was elevated, andaldosterone levels peaked at 232 pg/ml in follow-up examinations.Plasma uric acid was increased, and uric acid clearance wasdecreased. In several follow-up examinations, the fractionalurine excretion of sodium varied between 1.52 and 4.41%, and24-h urinary calcium excretion varied between 64 and 194 µmol/kg.Urine osmolality varied between 245 and 284 mOsmol/kg with 24-hurine amounts ranging between 1050 and 4000 ml (approximately20 to 80 ml/kg per 24 h). Urinary excretion of PGE2 was normal,as was that of PGE-M. Sonography of the kidneys showed bilateralmedullary nephrocalcinosis. Oral potassium chloride (0.6 mmol/kgbody wt) and allopurinol (7.5 mg/kg body wt) were started.
In both siblings, the constellation of hypokalemia, (compensated)metabolic alkalosis, mild polyuria, increased fractional sodiumexcretion, and hypercalciuria with nephrocalcinosis pointedto a sodium chloride transport defect along the TAL of Henle’sloop. To evaluate the function of the TAL in vivo, we appliedthe NKCC inhibitor furosemide to both patients and the effectson diuresis and urinary chloride excretion were determined.In case of impaired TAL sodium chloride reabsorption, additionof the loop diuretic furosemide should exert no or only minoreffects on diuresis and renal chloride excretion. When comparedwith healthy control subjects and patients with typical clinicalfeatures of aBS/HPS (4), the stimulatory effect of furosemideon diuresis and urinary chloride excretion for both siblingswas between the means of both groups (Figure 1). A partiallyimpaired capacity for sodium chloride reabsorption along theTAL therefore could explain the electrolyte disturbances ofboth patients.
Figure 1. Effect of furosemide on diuresis and urinary chloride excretion. Diuresis (top) and urinary chloride excretion (bottom) were determined before and after application of furosemide, and the difference was normalized to urine furosemide excretion as described by Koeckerling et al. (4). The mean values (± SEM) determined for healthy control subjects (n = 13) and patients with antenatal Bartter syndrome/hyperprostaglandin E syndrome (aBS/HPS; n = 8) were taken from Koeckerling et al. (4).
The NKCC2 isoform of the sodium-potassium-chloride co-transportersis the pacemaker of TAL sodium chloride reabsorption. We thereforeanalyzed the sequence of the NKCC2 encoding gene SLC12A1 inboth patients. As shown in Figure 2, mutations were detectedin both SLC12A1 alleles of both patients: A missense mutation(c.530 T>A) resulting in an phenylalanine to tyrosine exchangeat amino acid position 177 (p.F177Y) was inherited from thefather, and a frameshift mutation (c.2751dupT) that resultsin a frameshift with premature stop codon at amino acid position918, deleting a major part of the intracellular C-terminus (p.D918fs),was inherited from the mother. As expected, the parents werefound to be heterozygous for only one mutation. These mutationswere neither described before nor observed in 100 control chromosomesfrom healthy individuals (data not shown).
Figure 2. Na+-K+-2Cl– (NKCC2) mutations and their topology. The p.F177Y and the p.D918fs mutations were identified in both patients. Both mutations affect highly conserved amino acid residues (middle: amino acid alignment of human [Homo sapiens (Hs)], rat [Rattus norvegicus (Rn)], spiny dogfish [Squalus acanthias (Sa)], and fruit fly [Drosophila melanogaster (Dm)] NKCC2 sequences). The p.L196P mutation that was detected in a patient with aBS/HPS served as a nonfunctional negative control. The position of the hemagglutinin (HA) epitope inserted for NKCC2 surface quantification is indicated between transmembrane domains VII and VIII close to two predicted glycosylation sites (protein model according to reference [15]).
Genetic defects of the NKCC2 transporter were the first to beidentified in aBS/HPS (1). Up to now, NKCC2 defects were unambiguouslyassociated with prenatal manifestation and severe, life-threateningvolume depletion in the early neonatal period (6). So far analyzed,complete loss of human NKCC2 transport activity was describedfor the SLC12A1 mutations associated with the severe phenotype(3). The relatively mild phenotype observed in these cases thereforemight be explained by some residual activity of the affectedNKCC2 proteins. To clarify this point, we performed 36Cl-uptakeexperiments after heterologous expression of wild-type rat NKCC2and mutated proteins in Xenopus oocytes. The rat ortholog ofNKCC2, which shows 93% overall sequence identity with the humanisoform, was used for these experiments because it gave morerobust uptake signals. As shown in Figures 3 and 4, top, expressionof wild-type rat NKCC2 (NKCC2-WT) resulted in an approximatelythree-fold increase of 36Cl uptake when compared with noninjectedoocytes. As expected, expression of a mutated NKCC2 proteinthat was identified in a patient with a prenatal onset aBS/HPS(p.L196P) did not induce chloride uptake significantly differentfrom control oocytes. The same was true for the construct withthe p.D918fs mutation detected in one SLC12A1 allele of thetwo brothers. In contrast, expression of the p.F177Y mutationthat was detected in the other allele showed partial chlorideuptake, which amounted to 50% or more of the wild-type NKCC2uptake. Therefore, the p.F177Y mutation in humans may resultin a chloride-transporting protein with approximately half ofthe activity of the wild-type protein.
Figure 3. Functional analysis of the NKCC2 mutations. Xenopus oocytes were injected with 10 ng of rat cRNA for the indicated constructs and 36Cl uptake subsequently was measured. Noninjected oocytes and oocytes that expressed the p.L196P mutation that was identified in a patient with aBS/HPS served as controls. In contrast to the p.D918fs frameshift mutation, the p.F177Y missense mutation induced a 36Cl uptake that was significantly different from noninjected oocytes. The columns represent the mean ± SEM of at least 15 oocytes per injected cRNA. A similar level of expression of wild-type and mutated NKCC2 proteins is demonstrated by Western blot analysis of HA-tagged constructs with a monoclonal anti-HA antibody. Note that the wild-type protein without HA epitope (WT) is not recognized by the antibody and that the p.D918fs construct is shifted to a lower molecular weight because of the deletion of a part of the C-terminal sequence.
Figure 4. Transport activity (top) and surface expression (bottom) of NKCC2 wild-type and mutated proteins. Wild-type and mutated NKCC2 rat cRNA (20 µg for each construct) were injected alone or in combination as indicated. NKCC2 constructs with an extracellular HA epitope were used for luminometric surface quantification after heterologous expression in Xenopus oocytes. Oocytes that expressed NKCC2 without HA epitope (NKCC2-WT) served as a negative control. In contrast to the construct with p.D918fs (p.D918fs-HA) and p.L196P (p.L196P-HA identified in a patient with aBS/HPS) the construct with p.F177Y (p.F177Y-HA), which displayed residual transport activity, showed a surface expression that was not significantly different from the wild-type protein (NKCC2-HA). Coexpression of p.D918fs does not negatively affect transport activity or surface expression of NKCC2 wild type or p.F177Y. The columns represent the mean ± SEM of at least 15 oocytes per injected cRNA. The P values are given as numbers.
Functional NKCC2 transporters recently were shown to consistof two NKCC2 subunits (7). In view of the heterozygosity forboth mutations, as observed in our patients, this dimeric architecturemight implicate a negative impact of one mutated transporteron the function of the other. To rule out such a dominant negativeeffect of the p.D918fs mutation on the function of the constructwith p.F177Y, we performed coexpression experiments with co-injectionof both mutants as well as co-injection of each of the mutantsplus the wild type. As shown in Figure 4, neither of the constructswith p.F177Y or p.D918fs impaired transport activity of NKCC2wild type. Moreover, the activity of the construct with p.F177Yremained unchanged in the presence of the p.D918fs mutant. Thesein vitro findings argue against a dominant negative effect ofthe p.D918fs mutation. To support this notion in vivo, we determinedthe osmolality of a morning urine sample of the patients’mother, who is heterozygous for the p.D918fs mutation. In linewith our experimental data, the measured urine osmolality of930 mOsm/kg pointed to a normal renal concentrating capabilityand argues against an inhibitory effect of the p.D918fs mutationon NKCC2 wild-type function.
Impaired NKCC2 transport activity in case of the p.F177Y mutationmight result from (1) decreased protein stability, (2) decreasedexpression in the plasma membrane via either impaired insertionin or increased clearance from the membrane, or (3) disturbedfunction of an otherwise normally processed protein. To revealthe underlying mechanism compromising transport activity ofthe construct with the p.F177Y mutation, we compared surfaceexpression of wild-type and mutated proteins expressed in Xenopusoocytes. To this end, we inserted an HA epitope into an extracellularloop of the NKCC2 proteins (NKCC2-HA; the transport activityas determined by 36Cl-uptake measurements was not affected bythis protein modification; data not shown). Quantification ofHA expression on viable oocytes via an enzyme-linked luminometricassay then allowed determination of NKCC2 surface expression.As shown in Figure 4, a strong HA-dependent luminescence signalappeared after expression of wild-type NKCC2-HA. No surfaceexpression was detected for the loss-of-function mutation p.L196Pdetected in a patient with aBS/HPS. Similarly, a severely reducedcell membrane expression was observed for the construct withthe p.D918fs (p.D918fs-HA). In contrast, a luminescence signalsimilar to that of the wild-type protein appeared after expressionof the p.F177Y mutation (p.F177Y-HA). The p.D918fs mutationtherefore seemed to affect cell membrane insertion of NKCC2,whereas the p.F177Y mutation impaired directly the transportfunction of the NKCC2 protein without affecting protein stabilityor membrane insertion. Moreover, in co-injection experiments,neither p.F177Y nor p.D918fs affected the surface expressionof the HA-tagged wild-type protein (NKCC2-HA + p.F177Y and NKCC2-HA+ p.D918fs, respectively), and co-injection of both mutants(p.F177Y-HA + p.D918fs) showed no effect of the p.D918fs mutanton surface expression of p.F177Y-HA.
In this report, we describe an unusually mild phenotype thatresulted from a particular combination of mutations of the SLC12A1gene coding for the NKCC2 sodium potassium chloride co-transporter.The clinical picture with mild polyuria and borderline hypercalciuriaclearly contrasts with the prenatal-onset disease with polyhydramnios,prematurity, and severe neonatal volume depletion hitherto knownto be associated with SLC12A1 gene mutations (aBS/HPS). Consistentwith the in vivo finding of residual furosemide-sensitive chloridetransport along the TAL, our functional results in vitro suggestsignificant residual transport activity in humans of the p.F177Ymutated NKCC2 protein, which may well account for the clearlyattenuated phenotype of both patients. Their laboratory findingsperfectly mirror the side effects of NKCC inhibition by chronicfurosemide application (8): Decreased urine-concentrating ability,increased fractional sodium excretion, and hypercalciuria. Saltwasting and volume depletion in addition lead to secondary hyperaldosteronism(resulting in hypokalemia and metabolic alkalosis) and increaseproximal tubular urate reabsorption entailing hyperuricemia.
According to our functional data, only approximately half ofthe normal NKCC2 activity would be expected along the TAL ofthe described patients. Although this extent of functional NKCC2impairment plausibly reconciles with the mild clinical pictureand the reduced renal tubular sensitivity for furosemide, aone-to-one translation from our heterologous expression datato in vivo conditions might not necessarily be warranted. Additionalfactors in TAL cells, which might not be relevant in the settingof heterologous overexpression of cRNA in Xenopus oocytes, mightinfluence the activity of the mutated proteins. Despite theselimitations of a heterologous expression system using rat cRNA,our data suggest that the investigated mutations affect differentaspects of NKCC2 protein function. The p.D918fs mutation clearlyinterferes with membrane trafficking of the NKCC2 protein, anda patient who is homozygous for this mutation is likely to presentwith severe aBS/HPS, whereas the p.F177Y mutation directly impairstransport function of NKCC2, whose abundance in the cell membraneseems not to be affected. One might speculate that homozygosityfor this mutation would result in an even milder phenotype asobserved for the described patients. Unfortunately, comprehensivestructure-function data (i.e., crystallization data) are notyet on hand for this family of transport proteins. The availabledata mainly are inferred indirectly from domain-swapping approachesbetween shark and mammalian NKCC and point to the importanceof the second, fourth, and seventh predicted -helical domainsfor ion transport (9). Therefore, we cannot predict the roleof the affected amino acid residue at the beginning of the first-helical domain in the transport cycle of the NKCC2 protein.With respect to the results from our coexpression experimentsand the normal urine-concentrating capability of the mother,who is heterozygous for the p.D918fs mutation, we can exclude,however, a dominant negative effect of this mutation on NKCC2wild-type or p.F177Y activity.
In line with the mild clinical presentation, urinary excretionof PGE2, which is increased heavily in the prenatal-onset disease(2), was close to normal in our patients. Renal PGE2 overproductionin aBS/HPS is thought to derive from impaired entry of chlorideinto the macula densa cells of the distal tubule, which in turnincreases expression of cyclooxygenase 2, a key enzyme of theprostanoid pathway (2,10,11). In contrast to epithelial cellsof the TAL, which express the isoforms A and F of NKCC2, apicalsodium chloride entry in macula densa cells is mediated by theisoform B. These isoforms result from alternative splicing ofan exon encoding a 96-bp region of NKCC2, giving rise to threedifferent protein sequences in the second transmembrane domain.Because the p.F177Y mutation affects the beginning of the firsttransmembrane domain, it is expected to occur in any of thedescribed NKCC2 splice variants. Residual transport activityof the macula densa isoform B hence might explain normal urinaryPGE2 excretion as observed in our patients and thereby in additionwould support the concept of residual NKCC2 transport activity.
Despite this strong evidence for residual NKCC2 function, therenal concentrating capability of our patients, as deduced fromtheir urine osmolalities, surprisingly was not substantiallyhigher than that of patients with untreated aBS/HPS (2). Animpairment of the renal concentrating capability by the severemedullary nephrocalcinosis in addition to the partial NKCC2defect might explain this discrepancy. Moreover, we cannot completelyrule out the possibility of another mutation within unsequencedregions of the SLC12A1 gene, which might interfere with NKCC2expression or function in vivo.
Although exceptional for NKCC2 mutations, mild and late-onsetclinical manifestation of Bartter syndrome has been describedfor mutations that affect the CLCNKB gene coding for the ClC-Kbchloride channel (6,12,13). Distinct differences in diseaseseverities hitherto hampered a direct comparison of severe renalsalt wasting as observed in aBS/HPS caused by NKCC2 malfunctionwith the milder Bartter syndrome as a result of impaired ClC-Kbfunction. Whereas plasma electrolyte abnormalities of patientswho have CLCNKB gene defects resemble those of the here describedmild NKCC2 defect, the former normally show milder impairmentin urine-concentrating capability and normal urinary calciumexcretion so that nephrocalcinosis usually does not develop(6). How could this discrepancy be explained? ClC-Kb togetherwith the -subunit barttin forms a basolateral chloride exitpathway not only in the TAL but also along the distal convolutedtubule (DCT), where luminal uptake of sodium chloride is mediatedvia the thiazide-sensitive sodium chloride co-transporter NCCT(2,14). In contrast to a NKCC2 defect, which affects sodiumchloride reabsorption exclusively along the TAL, a ClC-Kb defectthus would impair salt reabsorption along the TAL and the DCT.In contrast to the TAL, disturbed sodium chloride reabsorptionalong the DCT results in hypocalciuria. In case of a combinedimpairment of TAL and DCT, this calcium-saving effect of disturbedDCT sodium chloride reabsorption could counteract the reducedcalcium reabsorption along the TAL, eventually resulting innormo- or even hypocalciuria (12). As learned from our two patients,the phenotypic differences observed in NKCC2 and ClC-Kb malfunctiontherefore may reflect not only differences in disease severitybut also the different nephron segments affected by the mutations.
We have shown that a particular combination of NKCC2 mutationsis associated with variations in disease severity. An incompleteloss-of-function effect of the described p.F177Y mutation mightexplain the mild phenotype. Reduced NKCC2 function thereforemight be considered in patients who present with hypokalemia,metabolic alkalosis, hyperuricemia, and nephrocalcinosis evenbeyond the neonatal period.
Acknowledgments
This study was supported by the Deutsche Forschungsgemeinschaft(WA1088/3), the Kempkes Foundation of the University of Marburg,and the German Kidney Foundation.
Footnotes
Published online ahead of print. Publication date availableat www.jasn.org.
Simon DB, Karet FE, Hamdan JM, DiPietro A, Sanjad SA, Lifton RP: Bartter’s syndrome, hypokalaemic alkalosis with hypercalciuria, is caused by mutations in the Na-K-2Cl cotransporter NKCC2. Nat Genet 13: 183–188, 1996[CrossRef][Medline]
Jeck N, Schlingmann KP, Reinalter SC, Komhoff M, Peters M, Waldegger S, Seyberth HW: Salt handling in the distal nephron: Lessons learned from inherited human disorders. Am J Physiol Regul Integr Comp Physiol 288: R782–R795, 2005[Abstract/Free Full Text]
Starremans PG, Kersten FF, Knoers NV, van den Heuvel LP, Bindels RJ: Mutations in the human Na-K-2Cl cotransporter (NKCC2) identified in Bartter syndrome type I consistently result in nonfunctional transporters. J Am Soc Nephrol 14: 1419–1426, 2003[Abstract/Free Full Text]
Kockerling A, Reinalter SC, Seyberth HW: Impaired response to furosemide in hyperprostaglandin E syndrome: Evidence for a tubular defect in the loop of Henle. J Pediatr 129: 519–528, 1996[CrossRef][Medline]
Zerangue N, Schwappach B, Jan YN, Jan LY: A new ER trafficking signal regulates the subunit stoichiometry of plasma membrane K(ATP) channels. Neuron 22: 537–548, 1999[CrossRef][Medline]
Peters M, Jeck N, Reinalter S, Leonhardt A, Tonshoff B, Klaus GG, Konrad M, Seyberth HW: Clinical presentation of genetically defined patients with hypokalemic salt-losing tubulopathies. Am J Med 112: 183–190, 2002[CrossRef][Medline]
Starremans PG, Kersten FF, Van Den Heuvel LP, Knoers NV, Bindels RJ: Dimeric architecture of the human bumetanide-sensitive Na-K-Cl co-transporter. J Am Soc Nephrol 14: 3039–3046, 2003[Abstract/Free Full Text]
Greenberg A: Diuretic complications. Am J Med Sci 319: 10–24, 2000[CrossRef][Medline]
Isenring P, Forbush B: Ion transport and ligand binding by the Na-K-Cl cotransporter, structure-function studies. Comp Biochem Physiol A Mol Integr Physiol 130: 487–497, 2001[CrossRef][Medline]
Cheng HF, Wang JL, Zhang MZ, McKanna JA, Harris RC: Role of p38 in the regulation of renal cortical cyclooxygenase-2 expression by extracellular chloride. J Clin Invest 106: 681–688, 2000[Medline]
Komhoff M, Jeck ND, Seyberth HW, Grone HJ, Nusing RM, Breyer MD: Cyclooxygenase-2 expression is associated with the renal macula densa of patients with Bartter-like syndrome. Kidney Int 58: 2420–2424, 2000[CrossRef][Medline]
Jeck N, Konrad M, Peters M, Weber S, Bonzel KE, Seyberth HW: Mutations in the chloride channel gene, CLCNKB, leading to a mixed Bartter-Gitelman phenotype. Pediatr Res 48: 754–758, 2000[Medline]
Konrad M, Vollmer M, Lemmink HH, van den Heuvel LP, Jeck N, Vargas-Poussou R, Lakings A, Ruf R, Deschenes G, Antignac C, Guay-Woodford L, Knoers NV, Seyberth HW, Feldmann D, Hildebrandt F: Mutations in the chloride channel gene CLCNKB as a cause of classic Bartter syndrome. J Am Soc Nephrol 11: 1449–1459, 2000[Abstract/Free Full Text]
Jentsch TJ: Chloride transport in the kidney: Lessons from human disease and knockout mice. J Am Soc Nephrol 16: 1549–1561, 2005[Abstract/Free Full Text]
Haas M, Forbush B 3rd: The Na-K-Cl cotransporter of secretory epithelia. Annu Rev Physiol 62: 515–534, 2000[CrossRef][Medline]
Related Article
This Month’s Highlights
J. Am. Soc. Nephrol. 2006 17: 2075-2076.
[Full Text][PDF]
This article has been cited by other articles:
F. Lang, V. Vallon, M. Knipper, and P. Wangemann Functional significance of channels and transporters expressed in the inner ear and kidney
Am J Physiol Cell Physiol,
October 1, 2007;
293(4):
C1187 - C1208.
[Abstract][Full Text][PDF]
Top
Abstract
Introduction
V, 
-globin gene. Site-directed mutagenesis was performed according to the QuickChange protocol (QuickChange Site Directed Mutagenesis Kit; Stratagene). 
0.05. 
-helical domains for ion transport (9). Therefore, we cannot predict the role of the affected amino acid residue at the beginning of the first 
