Macula Densa Control of Renin Secretion and Preglomerular Resistance in Mice with Selective Deletion of the B Isoform of the Na,K,2Cl Co-Transporter
Mona Oppermann*,
Diane Mizel*,
George Huang*,
Cuiling Li*,
Chuxia Deng*,
Franziska Theilig,
Sebastian Bachmann,
Josie Briggs*,
Jurgen Schnermann* and
Hayo Castrop*,
* National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland; Anatomy, Charite, Humboldt University, Berlin, Germany; and Institute of Physiology, University of Regensburg, Regensburg, Germany
Address correspondence to: Dr. Hayo Castrop, NIDDK, NIH, Building 10, Room 4 D51, 10 Center Drive MSC-1370, Bethesda, MD 20892-1370; Phone: 301-435-6579; Fax: 301-435-6587; E-mail: hayo{at}castrop.com
Received for publication April 21, 2006.
Accepted for publication May 18, 2006.
Na,K,2Cl co-transporter (NKCC2), the primary NaCl uptake pathwayin the thick ascending limb of Henle, is expressed in threedifferent full-length splice variants, called NKCC2F, NKCC2A,and NKCC2B. These variants, derived by differential splicingof the variable exon 4, show a distinct distribution patternalong the loop of Henle, but the functional significance ofthis organization is unclear. By introduction of premature stopcodons into exon 4B, specific for the B isoform, mice with anexclusive NKCC2B deficiency were generated. Relative expressionlevels and distribution patterns of NKCC2A and NKCC2F were notaltered in the NKCC2B-deficient mice. NKCC2B-deficient micedid not display a salt-losing phenotype; basal plasma reninand aldosterone levels were not different from those of wild-typemice. Ambient urine osmolarities, however, were slightly butsignificantly reduced. Distal Cl concentration was significantlyelevated and loop of Henle Cl absorption was reduced in microperfusedsuperficial loops of Henle of NKCC2B-deficient mice. Becauseof the presence of NKCC2A in the macula densa, maximum tubuloglomerularfeedback responses were normal, but tubuloglomerular feedbackfunction curves were right-shifted, indicating reduced sensitivityin the subnormal flow range. Plasma renin concentration in NKCC2B-deficientmice was reduced under conditions of salt loading compared withthat in wild-type mice. This study shows the feasibility ofgenerating mice with specific deletions of single splice variants.The mild phenotype of mice that are deficient in the B isoformof NKCC2 indicates a limited role for NKCC2B for overall saltretrieval. Nevertheless, the high-affinity NKCC2B contributesto salt absorption and macula densa function in the low NaClconcentration range.
In addition to being affected by numerous systemic factors,substantial evidence supports the concept that renal vascularresistance and renin secretion are the end points of an intrarenalregulatory pathway that resides in the juxtaglomerular apparatus.By connecting tubular epithelial and preglomerular vascularcells, this structure provides an anatomic route along whichinformation about tubular fluid composition can be signaledto vascular effector cells of the same nephron. Specifically,renin secretion and preglomerular resistance have been shownto be strongly affected by changes of luminal NaCl concentrationat the level of the epithelial sensor cells of the macula densa(1–6). Transport activity of the Na,K,2Cl co-transporter(NKCC2; BSC1) in the apical membrane of macula densa cells isthought to be the mechanism by which these cells detect tubularsalt concentration (7–12).
In all mammalian species so far examined, NKCC2 is expressedalong the thick ascending limb of Henle (TAL) in at least threedifferent splice variants, called NKCC2F, NKCC2A, and NKCC2B(13–17). These variants are derived from one distinctgene by differential splicing of the variable exon 4, whichencodes for the second transmembrane domain and parts of theadjacent intracellular loop of the transporter protein (14,15,18).NKCC2F seems to be the dominant isoform of NKCC2, with highestexpression levels in the medullary portion of the TAL (13,19).NKCC2A and NKCC2B seem to be expressed in the more distal orcortical segments of the TAL (13,19). A recent attempt to localizeNKCC2 isoforms by in situ hybridization in the mouse suggestedthat macula densa cells seem to express exclusively and selectivelythe B isoform of the co-transporter (13). Therefore, NKCC2Bseemed to be the most likely candidate for serving as the saltsensor in macula densa control of renin secretion and preglomerularresistance.
Studies of the NKCC2 sensor function in macula densa controlof renin secretion and preglomerular resistance in vivo traditionallyhave been performed by examination of the effects of loop diureticssuch as furosemide or bumetanide (20,21). However, these agentsblock the transport activity of all isoforms of NKCC2. Completeinactivation of NKCC2 by gene targeting results in severe electrolytedisturbances that lead to early postnatal death in most animalsand therefore represents a model that cannot be studied easilyphysiologically (22).
In these experiments, we assessed the effect of a more localizedinactivation of NKCC2 on macula densa function. On the basisof the premise that NKCC2B is the dominant isoform of NKCC2in macula densa cells, we generated a mouse strain that specificallyis deficient in NKCC2B. NKCC2B-deficient mice showed an impairmentof TAL diluting function accompanied by a slight decrease ofosmotic urine concentration and a significant right shift ofthe tubuloglomerular feedback (TGF) function curve. Renin secretionwas suppressed during salt loading. Partial maintenance of maculadensa signaling indicates that another NKCC2 isoform in additionto NKCC2B must be involved in the macula densa sensing function.
Generation of NKCC2B-Deficient Mice
All animal studies were performed according to protocols thatwere examined and approved by the Animal Use and Care Committeeof the National Institute of Diabetes and Digestive and KidneyDiseases and by the Animal Care Committee of the Universityof Regensburg. A targeting vector was generated to introducepremature stop codons into exon 4B of the NKCC2 gene (Slc12a1;Figure 1, A and C). This targeting vector consisted of a 5'homologous arm that was introduced into the vector ploxP (23),followed by the coding sequence for the FLAG peptide, by stopcodons in all reading frames, and by a neomycin-resistance (neo)cassette flanked by loxP sites. 3'-Adjacent to the neo cassette,a 3' homologous arm was inserted (Figure 1B). Homologous armswere generated by long-distance PCR (Roche, Indianapolis, IN).Primers used were 5'-GGAAGGTTATTTGGGCTTGGTCCTG-3' and 5'-ACACCGAGACCTCAGTTGAGAGG-3'for the upstream arm and 5'-GATTATCATCGGCTTAGCCGTGACAG-3' and5'-GGAGTTGCCACACTACATGGGCTC-3' for the downstream arm. The targetingvector was linearized with NotI and transfected into 129SvEvembryonic stem cells. Primers used for PCR screening of theclones were 5'-CGCAGCGCATCGCCTTCTATCGCCTTC-3' (inside the neocassette) and 5'-CCACGGTGATGGAACCGATGATG-3' (3' outside of thedownstream homologous region). Embryonic stem cell clones thathad undergone homologous recombination were injected into blastocystsof C57BL/6 mice and implanted into pseudopregnant foster femalemice. Two male chimeras were crossed to C57BL/6 female miceto test for germline transmission. Genotyping was performedon tail biopsies by PCR using primers for the neo gene (5'-ACAACAGACAATCGGCTGCTCTGATG-3'and 5'-TGCGCGCCTTGAGCCTGGCGAAC-3') and primers flanking thewild-type or the mutated exon 4B, respectively (5'-CGAGGGCAACAGAGCTCTGCACATTCC-3'and 5'-GGTCAGTGGTGAGCTATTGGATGGAGC-3').
Figure 1. (A) Differential splicing of the variable exon 4 results in Na,K,2Cl co-transporter B (NKCC2B), NKCC2A, and NKCC2F transcripts. (B) Targeting strategy in which introduction of premature stop codons into exon 4B leads to premature stop of translation; the vector also contained the FLAG peptide coding sequence upstream of the stop codons (not shown in the scheme). Arrows indicate PCR-based screening strategy for alleles that had undergone homologous recombination. (C) Schematic drawing of the NKCC2 protein with arrow indicating location of translational stop.
Removal of the Neo Cassette
The loxP-flanked neo cassette was excised by crossing homozygousNKCC2B–/–neo+ mice with EIIa-Cre mice (24). OffspringDNA was tested for the absence of neo, and progeny that lackedthe neo gene were intercrossed to obtain homozygous NKCC2B–/–mice and wild-type littermates (Figure 1B).
Analysis of the Mutated Transcripts of NKCC2B–/– Mice
Total RNA from kidneys of NKCC2B+/+ and –/– micewas isolated using Trizol reagent (Life Technologies, Carlsbad,CA). NKCC2B mutated transcripts were analyzed by reverse transcription–PCR(RT-PCR) using primers that spanned from exon 4B to exon 5 andfrom exon 2 to exon 4B. To verify the integrity of the remainingisoforms, we also performed analogous RT-PCR for NKCC2A andNKCC2F transcripts. In addition, cDNA fragments that spannedfrom exon 2 to exon 5 were cloned into TOPO XL (Life Technologies),and all fragments were sequenced.
Localization of NKCC2B, NKCC2A, and NKCC2F Transcripts in Kidney Segments
A localization study for the three different isoforms of NKCC2was performed by real-time RT-PCR on RNA that was derived fromdifferent kidney regions. Kidneys were dissected under a microscope,and cortex, outer medulla, and inner medulla tissue pieces werecollected. Real-time PCR was performed using isoform-specificTaqMan probes in an ABI Prism 7900 Sequence Detection System(Applied Biosystems, Foster City, CA).
Localization of NKCC2B, NKCC2A, and NKCC2F Transcripts by In Situ Hybridization
For in situ hybridization, kidney sections were prepared asdescribed previously (25). The mRNA expression of NKCC2A, -B,and -F was localized using digoxigenin-labeled riboprobes accordingto the manufacturer’s protocol (Roche). Sense and antisenseprobes were generated by in vitro transcription of a 350-bpcDNA of each isoform that spanned from exon 4 into exon 5. Inaddition, a 96-bp antisense probe of each isoform from exon4 was used. Sense probes were applied in parallel with antisenseprobes to test for unspecific hybridization signals.
Immunohistochemistry
Immunolabeling was performed as described (25) using guineapig anti-NKCC2 (directed against the C-terminal region; giftfrom Dr. Ellison, Oregon Health Sciences University, Portland,OR) followed by incubation with donkey anti-guinea pig cy3-Ig(Dianova, Hamburg, Germany).
BP and Heart Rate
BP and heart rate in conscious mice were determined by tail-cuffmanometry (Visitech Systems, Apex, NC) (26).
Blood Collection and Plasma Renin Determination
Plasma renin concentration (PRC) in blood samples from tailveins was measured with a commercial RIA kit (DiaSorin, Stillwater,MN), as described previously (27). For assessment of renin secretionafter inhibition of macula densa salt transport activity, asingle dose of furosemide (40 mg/kg) was injected intraperitoneallyand blood samples were drawn 60 min later. For determinationof macula densa–dependent inhibition of renin release,mice received a single intravenous (tail vein) injection of1 ml of saline, and blood samples were collected 60 min later(27). For chronic modulation of the renin-angiotensin system,mice were subjected to the following treatments: (1) Controlgroup: Mice were fed a standard rodent diet (0.3% NaCl [wt/wt])for 1 wk; 2) salt-deficient group: After injection of a singledose of furosemide (40 mg/kg intraperitoneally), mice receiveda salt-deficient diet (0.03% NaCl [wt/wt]) for 1 wk; 3) andsalt load group: Mice were fed a high-salt diet (4% NaCl [wt/wt])for 1 wk.
Micropuncture Experiments
Measurements of stop flow pressure (PSF) during loop of Henleperfusion were done as described (12). When PSF had stabilized,loop of Henle perfusion rate was increased to 30 nl/min, andmaximum PSF responses were determined. Perfusion rates thenwere decreased to 20, 15, 10, 5, and 0 nl/min and maintaineduntil steady states were achieved at each flow rate. Logisticcurve parameters were determined using GraphPad Prism (GraphPad,San Diego, CA). The perfusion fluid contained 136 mM/L NaCl,4 mM/L NaHCO3, 4 mM/L KCl, 2 mM/L CaCl2, 7.5 mM/L urea, and100 mg/100 ml FD&C green (Keystone, Bellefonte, PA).
For determination of absorptive TAL function, loops of Henlewere perfused as described above. Fluid samples were collectedfrom distal tubular segments of the perfused segments for 1to 3 min at set flow rates of 6 and 15 nl/min. After volumedeterminations, a 1-nl sample was removed for determinationof Cl– concentration using an electrometric Cl–microtitrator (28).
Statistical Analyses
Unpaired t test was used to compare two values between differentanimals. Multiple groups were analyzed with ANOVA followed byBonferroni post test. P < 0.05 was considered significant.
Mutated Transcripts in NKCC2B–/– Mice
NKCC2-deficient mice were viable and showed no gross anatomic,behavioral, or fertility abnormalities (Table 1). Crossing ofheterozygous mice yielded offspring of the different genotypesin near Mendelian ratio. In a first set of experiments, we addressedthe nature of the mutated NKCC2B transcripts and those of theremaining isoforms. As can bee seen in Figure 2, the targetingevent led to the introduction of several premature stop codonsin all reading frames, rendering NKCC2B transcripts nonfunctional.NKCC2A and NKCC2F transcripts, in contrast, were found to beunaltered. Before removal of the bulky neomycin-cassette, splicingof the mutated exon 4B was compromised, leading to two differenttranscripts with partial losses of neo (Figure 2A). In contrast,correct splicing was fully preserved after Cre-mediated removalof the floxed selection cassette (Figure 2B). For assessmentof whether targeting NKCC2B would influence the quantity andthe localization of the remaining NKCC2A and NKCC2F, their respectivemRNA expression was determined by real-time RT-PCR of RNA thatwas derived from the inner medulla, outer medulla, and cortex.As can been seen in Figure 3, targeting NKCC2B did not alterthe expression levels of the remaining NKCC2A and NKCC2F isoforms.NKCC2F was the dominant isoform in the medulla with highestexpression levels in the inner stripe of the outer medulla.NKCC2B was expressed mainly in the cortex and to very low degreesin the outer medulla. NKCC2A was present in both cortex andmedulla. NKCC2A mRNA levels in the outer medulla were roughlytwice those in the cortex. No NKCC2 mRNA was detected in therenal inner medulla.
Figure 2. Sequencing of NKCC2B transcripts before and after removal of the neomycin resistance (neo) cassette. For both strains, the targeting event led to the introduction of several premature stop codons in all reading frames (underlined italic base triplets), rendering NKCC2B transcripts nonfunctional. (A) Presence of the neo cassette affected splicing of the mutated exon 4B and led to two different transcripts with partial losses of neo (* in the annotated sequence). (B) Correct splicing was observed after Cre-mediated removal of the floxed selection cassette; gaps in the annotated sequence refer to remaining vector sequences. Arrows in the schematic drawing of the transcripts (A and B) indicate PCR primers used for cloning and sequencing.
Figure 3. Quantification and localization of NKCC2B, NKCC2A, and NKCC2F mRNA in kidney regions of NKCC2B+/+ and –/– mice; real-time PCR was performed using isoform-specific TaqMan probes and 18s RNA as a reference. Data are given as percentage of the zone with highest expression levels of the respective isoforms. IM, inner medulla; OMIS, outer medulla inner stripe; OMOS, outer medulla outer stripe.
Urine Osmolarity and Concentrating Ability
To assess the effect of the loss of NKCC2B on urine-concentratingability, we determined urine osmolarity both under ambient conditionsand after 48 h of water deprivation (Figure 4). Ambient urineosmolarity averaged 1763 ± 121 in wild-type mice and1192 ± 60 in NKCC2B-deficient mice (n = 30 and 34, respectively;P < 0.0001). Water restriction substantially increased urineosmolarity in both wild-type (n = 19) and NKCC2B knockout (n= 20) mice to 3601 ± 115 and 3311 ± 97 mmol/L,respectively (P = 0.06).
Figure 4. Ambient urine osmolarity and urine osmolarity after water deprivation in NKCC2B+/+ and –/– mice. In water restriction experiments, female mice were used and urine was collected by catheterization of the urethra.
Plasma Renin and Plasma Aldosterone Concentrations
To address the consequences of the loss of the macula densaNKCC2B for the regulation of the renin system, we determinedbaseline PRC in NKCC2B-deficient mice, lacking NKCC2B in themacula densa and upstream parts of the TAL, and wild-type miceunder control conditions and after acute inhibition of NKCC2by administration of furosemide. Basal PRC levels were similarin wild-type (n = 30) and NKCC2B–/– mice (n = 22),averaging 1911± 150 and 1735 ± 160 ng angiotensinI (Ang I)/ml per h (P = 0.53). As can be seen in Figure 5, acuteadministration of furosemide to inhibit all isoforms of NKCC2increased PRC to the same extent in wild-type and knockout mice(7451 ± 1016 [n = 15] and 6213 ± 449 ng Ang I/mlper h [n = 12], respectively; P = 0.315 between genotypes).To investigate the renin system under conditions of chronicmodulation of oral salt intake, we determined PRC in mice thatwere fed either a high-salt diet or a salt-deficient diet for1 wk (Figure 6). After the high-salt diet, PRC was slightlyreduced in NKCC2B wild-type mice from 2218 ± 111 to 1832± 174 ng Ang I/ml per h (n = 12; P = 0.038) versus controldiet (–17%). In NKCC2B-deficient mice, a high-salt dietlowered PRC more efficiently, decreasing it from 2088 ±187 under normal diet to 1249 ± 140 ng Ang I/ml per h(n = 10; P = 0.002 versus control diet and P = 0.019 versuswild-type), a reduction in PRC of 40%. Conversely, a low-saltdiet stimulated PRC to similar levels in both NKCC2B+/+ and–/– mice (3320 ± 284 and 3359 ± 511ng Ang I/ml per h for wild-type [n = 12] and knockout [n = 10],respectively; P = 0.001 and 0.031 versus control diet).
Figure 5. Plasma renin concentration (PRC) before and 60 min after acute administration of furosemide (40 mg/kg, intraperitoneally) in NKCC2B+/+ (left) and –/– mice (right); lines connect data from the same mice.
Figure 6. PRC in NKCC2B+/+ and NKCC2B–/– mice that were fed a standard rodent diet (0.3% NaCl), high-salt diet (4% NaCl), or low-salt diet (a single dose of 40 mg/kg furosemide; 0.03% NaCl). PRC was determined after 1 wk of the respective treatment.
For determination of the effect of an acute salt load on reninsecretion, mice received a single intravenous injection of 1ml of saline, and blood samples were drawn 60 min later. Asshown in Figure 7, acute salt loading suppressed PRC by 42%,from 1032 ± 88 to 603 ± 36 ng Ang I/ml per h,in wild-type (n = 9), and by 64%, from 1139 ± 149 to408 ± 52 ng Ang I/ml per h, in NKCC2B–/–mice (n = 10). PRC was significantly lower after saline injectionin NKCC2B–/– compared with NKCC2B+/+ mice (P = 0.0015).
Figure 7. Effect of a single intravenous injection of 1 ml of saline on PRC in NKCC2B+/+ (n = 9) and –/– mice (n = 10). Blood samples were taken 60 min after the injection. Lines connect values from the same animals; percentages indicate relative decrements of PRC.
Plasma aldosterone concentrations after exposure to high- orlow-salt diet changed in parallel to PRC. As shown in Table 2,no significant differences between genotypes were observed underthese conditions.
Table 2. Plasma aldosterone concentrations (pg/ml) in NKCC2B+/+ and –/– mice after 1 wk on control (0.3% NaCl), high-salt (4% NaCl), or low-salt diet (0.03% NaCl after a single dose of 40 mg/kg furosemide, intraperitoneally)
BP and Heart Rate
Systolic BP measured by the tail-cuff compression method inawake wild-type mice averaged 120 ± 2.8 mmHg, and heartrate was 610 ± 19 bpm (n = 10). In NKCC2B-deficient mice,systolic BP and heart rate were 119 ± 2.0 mmHg and 615± 15 bpm, respectively (n = 10; NS).
Distal Tubular Chloride Concentration
To determine the effect of an NKCC2B deficiency on overall loopof Henle diluting ability, we determined chloride concentrationsin distal tubular fluid of superficial loops of Henle that weremicroperfused in situ at normal and elevated flow rates (Figure 8A).At tubular perfusion rates of 6 or 15 nl/min, distal chlorideconcentrations in NKCC2B–/– exceeded those in wild-typemice by 22.0 and 13.1 mEq/L, averaging 88.8 ± 4.3 versus66.8 ± 6.1 mmol/L at 6 nl/min and 106 ± 6.2 versus92.9 ± 4.0 mEq/L at 15 nl/min in NKCC2B–/–and wild-type mice, respectively (P = 0.008 and P = 0.042, betweengenotypes). Absolute Cl absorption along the loop of Henle wassignificantly higher in wild-type than in NKCC2B–/–mice at the low flow rate (723 ± 19 versus 602 ±26 pEq/min; P = 0.001) but not at the high flow rate (1420 ±72 versus 1269 ± 100 pEq/min; P = 0.22). Similarly, fractionalCl absorption was significantly higher in wild-type than inknockout mice at the low but not at the high flow rate (Figure 8B).Water absorption along the loop of Henle also was significantlyhigher in wild-type than knockout mice at the low flow rate(4.2 ± 0.3 versus 3.3 ± 0.3 nl/min; P = 0.032),whereas differences were NS at the high flow rate (7.9 ±0.6 versus 7.5 ± 0.7 nl/min).
Figure 8. (A) Chloride concentration in individual samples of distal tubular fluid of microperfused superficial loops of Henle in NKCC2B+/+ (•) and NKCC2B–/– mice (); data are shown for perfusion rates of 6 and 15 nl/min. (B) Mean fractional Cl absorption along the loop of Henle perfused at 6 and 15 nl/min. *P < 0.05.
TGF Regulation of GFR
TGF responses were determined by measurement of proximal PSFduring loop perfusion rates of 0, 5, 7.5, 10, 15, 20, and 30nl/min. Experiments were done in seven male NKCC2B wild-typeand four male NKCC2B-deficient mice, with the number of nephronsexamined being 13 and 11, respectively. Body weight averaged33 ± 1.0 g in wild-type and 36 ± 1.2 g in knockoutmice (NS). Mean arterial BP during the period of micropunctureexperiments was 95.1 ± 2.4 and 94.1 ± 3.0 mmHgfor wild-type and knockout mice, respectively (NS). Mean PSFat zero perfusion was not significantly different between genotypes(43.1 ± 2.1 versus 45.5 ± 1.9 mmHg; P = 0.42).PSF at a saturating perfusion rate (30 nl/min) tended to belower in nephrons from wild-type compared with knockout mice(31.9 ± 2.3 and 37.1 ± 2.6 mmHg; P = 0.14). Atintermediate perfusion rates of 5, 7.5, 10, 15, and 20 nl/min,PSF decreased from 43.1 ± 2.1 (zero perfusion) to 41.3± 2.1, 43.6 ± 3.4, 39.5 ± 2.0, 34.5 ±2.4, and 32.5 ± 2.5 mmHg in wild-type mice, whereas itchanged from 45.5 ± 1.9 (zero perfusion) to 47.1 ±3.2, 46.1 ± 2.6, 45.7 ± 1.7, 41.5 ± 1.9,and 38.8 ± 2.1 mmHg in NKCC2B–/– mice (P= 0.42, 0.14, 0.59, 0.04, 0.036, and 0.07 for genotype comparison).Half-maximum TGF responses were calculated to be reached ata perfusion flow of 12.3 ± 0.6 in wild-type and of 15.2± 1.2 nl/min in knockout mice (P = 0.033). Therefore,TGF response curves were right-shifted in NKCC2B-deficient mice(Figure 9).
Figure 9. Change of stop flow pressure (PSF) expressed as percentage reduction from PSFmax (PSF at zero loop flow) in response to stepwise increases of loop of Henle flow rates in wild-type (•) and NKCC2B–/– mice (). Flow rates that caused half-maximum responses (V1/2) are indicated by vertical lines. *P < 0.05.
Localization of the NKCC2 Isoforms along the TAL
In view of the largely preserved macula densa function in NKCC2B-deficientmice, we performed isoform-specific in situ hybridizations todetermine whether an additional isoform of NKCC2 is presentin macula densa cells. Using antisense riboprobes that werespecific for NKCC2B, NKCC2A, and NKCC2F mRNA, the macula densasegment of the TAL showed hybridization signals for both NKCC2Band NKCC2A but not NKCC2F (Figure 10). No signal was detectedusing respective sense probes (data not shown). Consistent withthe presence of the mRNA of an NKCC2 isoform other than NKCC2B,NKCC2 protein was detected in the macula densa of NKCC2B-deficientmice (Figure 11). In line with our localization study on kidneysegments using RT-PCR, NKCC2B and NKCC2A also were detectedin parts of the TAL upstream of the macula densa segment (datanot shown).
Figure 10.In situ hybridization of NKCC2 mRNA isoforms in the macula densa segment of the thick ascending limb (TAL; highlighted by arrows) in wild-type mice showing positive signals for NKCC2B and NKCC2A but not for the medullary F isoform of NKCC2.
Figure 11. NKCC2 protein expression in the macula densa segment of the TAL detected by an antibody directed against the C-terminal end of the NKCC2 protein; staining was observed in both NKCC2B+/+ (top) and NKCC2B–/– mice (bottom).
Movement of NaCl mediated by NKCC2 is the major uptake mechanismin the TAL of the loop of Henle and macula densa cells, andadequate NKCC2 operation is directly required for loop saltabsorption and body Na balance, urine-concentrating ability,renin secretion, and TGF. By alternative splicing, a numberof different NKCC2 isoforms that show distinct spatial distributionalong the loop of Henle and that differ on their affinity forNaCl are generated (13–17).
This report summarizes experiments in which we applied a strategyto target NKCC2B without interfering with the expression ofthe remaining isoforms, NKCC2A and NKCC2F. Several mouse modelswith specific removal of single exons have been reported (23,29–32),but to our knowledge, this is the first report of the inactivationof one of several isoforms derived from differential splicingwithout affecting the mechanism of differential splicing ofthe remaining isoforms. In our approach, we introduced stopcodons into the variable exon 4 (Figure 1), leading to prematuretermination of translation of the B isoform of NKCC2. Our experimentsprove the feasibility of a specific inactivation of a splicevariant of a gene by means of knock-in of premature stop codons.We assume that this strategy can be used as a general approachto obtain mouse models with inactivation of specific splicevariants of proteins, although the generation of a nonfunctionalprotein may be more uncertain if the variable exon is locatedclose to the C-terminal end of the protein.
Whereas inactivation of the entire NKCC2 protein by gene targetingleads to a severe salt-losing nephropathy with most animalsdying within the first 2 wk of postnatal life (22), specificinactivation of the B isoform of NKCC2 was not associated withgross structural or functional abnormalities. This is consonantwith this and previous studies indicating that NKCC2B transcriptsaccount for only approximately 10% of total NKCC2 mRNA expression(19). Ambient urine osmolarity and concentrating ability afterwater restriction was only slightly compromised in NKCC2B-deficientmice. Therefore, NKCC2B does not seem to contribute importantlyto the accumulation of medullary interstitial solute concentrationsthat is the basis of the urine-concentrating ability. In linewith this finding are observations in a single Bartter patientwith a missense mutation in exon 4B who displayed an unusuallymild type I Bartter phenotype (33). Nevertheless, a significantincrease in distal Cl concentration was observed in distal tubularfluid of microperfused loops of Henle, suggesting an impairmentof loop of Henle Cl absorption, presumably along the corticalTAL. It is of note that the effect of NKCC2B deficiency on dilutingability of individual loops of Henle was more pronounced atthe low than at the high flow rate, suggesting that NKCC2B maybe more active in a low Cl concentration range.
In situ hybridization in the mouse kidney has identified a punctatepattern of NKCC2B expression in the glomerular vicinity consistentwith predominant expression of the B isoform in macula densacells (13). NKCC2B has been suggested to represent the transporterisoform that primarily is responsible for detecting variationsof tubular chloride concentration by the macula densa. We thereforeexamined whether specific NKCC2B deficiency affects TGF regulationof afferent arteriolar tone and renin secretion the two endpoints that are known to be altered by macula densa input. Ourdata show that maximum TGF responses to increases in flow ratewere not significantly different between NKCC2B-deficient andwild-type mice. However, TGF responses in the mutant mice werereduced in the subnormal flow range, causing a right shift ofthe TGF curve. The requirement of NKCC2B to maintain a normalTGF responsiveness to subnormal flow changes indicates thatNKCC2B seems to operate most efficiently in the range of lowNaCl concentrations. This observation is in general agreementwith the recent demonstration that NKCC2B has a markedly higherCl affinity than the other two isoforms (13,17). It remainsto be investigated whether a constant exposure of the maculadensa cells to higher Cl concentrations in NKCC2B-deficientmice, as suggested by our distal Cl concentration measurements,leads to some desensitization of the macula densa cells, supportinga right shift of the TGF response curve.
Maintenance of maximum TGF responses in NKCC2B knockout miceclearly indicates that NKCC2B is not the only isoform presentin macula densa cells. In fact, the present in situ hybridizationevidence shows the coexpression of both NKCC2B and NKCC2A inthe macula densa segment and upstream parts of the TAL. In addition,NKCC2 protein was found immunohistochemically in the maculadensa of NKCC2B-deficient mice with an antibody directed againstthe C-terminal part of the protein. Considering the differentCl affinities of the co-transporter isoforms, with a Km forCl– of 9 to 12 and 22 to 45 mM for NKCC2B and NKCC2A,respectively (13,17,34), the A isoform seems to be better suitedto serve as a Cl sensor, with tubular Cl concentrations at themacula densa segment being approximately 35 mM (35).
To explore the effect of NKCC2B deficiency on renin secretion,the second end point of macula densa signaling, we determinedPRC in conscious knockout and wild-type mice. Interference withNKCC2 function by loop diuretics is consistently associatedwith a marked increase in renin secretion, which is mediatedin part by inhibition of macula densa salt transport capacity(36). Consistent with the notion that macula densa functionwas largely intact in NKCC2B-deficient mice as a result of thecoexpression of NKCC2A, PRC values were not significantly elevatedin the knockout animals. In fact, PRC was found to be significantlyreduced in NKCC2B-deficient mice during both chronic and acuteconditions of salt loading. It is conceivable that the increaseof macula densa NaCl concentration caused by a combination ofNKCC2B deficiency in the cortical TAL upstream of the maculadensa segment and salt loading is large enough to increase maculadensa NaCl transport despite NKCC2B absence and that this inhibitsrenin release during conditions of salt loading. Micropuncturestudies have shown that intravenous injections of isotonic salineindeed are accompanied with increased distal NaCl concentrations(2,37).
Overall, NKCC2B-deficient mice showed a mild phenotype comparedwith the massive salt-losing nephropathy observed in mice withinactivation of all NKCC2 isoforms (22). Further investigationswill be necessary to distinguish clearly between a phenotypethat is specifically related to loss of the B isoform of NKCC2and a phenotype derived from the reduction of total NKCC2 expressionas a result of NKCC2B deficiency.
We generated mice that lack the B isoform of NKCC2 by introducingpremature stop codons into exon 4B. NKCC2B-deficient mice didnot display a salt-losing phenotype, although ambient urineosmolarities were reduced. Distal Cl concentration was elevatedand loop of Henle Cl absorption was reduced in microperfusedsuperficial loops of Henle of NKCC2B-deficient mice. PRC inNKCC2B-deficient mice was reduced during salt loading. MaximumTGF responses were not significantly reduced in NKCC2B-deficientmice, indicating the operation of another isoform in the maculadensa, presumably NKCC2A. However, NKCC2B absence caused a rightshift of the TGF function curve consonant with dependence ofTGF responsiveness on NKCC2B in the low NaCl concentration range.
Acknowledgments
This work was supported by intramural funds from the NationalInstitute of Diabetes and Digestive and Kidney Diseases. M.O.is the recipient of a Visiting Fellowship of the National Instituteof Diabetes and Digestive and Kidney Diseases. H.C. was supportedby a grant from the Deutsche Forschungsgemeinschaft (SFB699/A4).
Footnotes
H.C.'s current address is Institute of Physiology, Universityof Regensburg, Germany.
Published online ahead of print. Publication date availableat www.jasn.org.
Itoh S, Carretero OA: Role of the macula densa in renin release. Hypertension 7: I49–I54, 1985[Medline]
Lorenz JN, Kotchen TA, Ott CE: Effect of Na and Cl infusion on loop function and plasma renin activity in rats. Am J Physiol 258: F1328–F1335, 1990
Lorenz JN, Weihprecht H, Schnermann J, Skott O, Briggs JP: Renin release from isolated juxtaglomerular apparatus depends on macula densa chloride transport. Am J Physiol 260: F486–F493, 1991
Persson AE, Salomonsson M, Westerlund P, Greger R, Schlatter E, Gonzalez E: Macula densa cell function. Kidney Int Suppl 32: S39–S44, 1991[Medline]
Schnermann J, Briggs J: Concentration-dependent sodium chloride transport as the signal in feedback control of glomerular filtration rate. Kidney Int Suppl 12: S82–S89, 1982[Medline]
Skott O, Briggs JP: Direct demonstration of macula densa-mediated renin secretion. Science 237: 1618–1620, 1987[Abstract/Free Full Text]
Bell PD, Lapointe JY: Characteristics of membrane transport processes of macula densa cells. Clin Exp Pharmacol Physiol 24: 541–547, 1997[Medline]
Bell PD, Lapointe JY, Cardinal J, Chang YS: Transport pathways in macula densa cells. Kidney Int Suppl 32: S59–S64, 1991[Medline]
Lapointe JY, Bell PD, Cardinal J: Direct evidence for apical Na+:2Cl-:K+ cotransport in macula densa cells. Am J Physiol 258: F1466–F1469, 1990
Schlatter E, Salomonsson M, Persson AE, Greger R: Macula densa cells sense luminal NaCl concentration via furosemide sensitive Na+2Cl-K+ cotransport. Pflugers Arch 414: 286–290, 1989[CrossRef][Medline]
Obermuller N, Kunchaparty S, Ellison DH, Bachmann S: Expression of the Na-K-2Cl cotransporter by macula densa and thick ascending limb cells of rat and rabbit nephron. J Clin Invest 98: 635–640, 1996[Medline]
Wright FS, Schnermann J: Interference with feedback control of glomerular filtration rate by furosemide, triflocin, and cyanide. J Clin Invest 53: 1695–1708, 1974[Medline]
Gimenez I, Isenring P, Forbush B: Spatially distributed alternative splice variants of the renal Na-K-Cl cotransporter exhibit dramatically different affinities for the transported ions. J Biol Chem 277: 8767–8770, 2002[Abstract/Free Full Text]
Igarashi P, Vanden Heuvel GB, Payne JA, Forbush B 3rd: Cloning, embryonic expression, and alternative splicing of a murine kidney-specific Na-K-Cl cotransporter. Am J Physiol 269: F405–F418, 1995
Payne JA, Forbush B 3rd: Alternatively spliced isoforms of the putative renal Na-K-Cl cotransporter are differentially distributed within the rabbit kidney. Proc Natl Acad Sci U S A 91: 4544–4548, 1994[Abstract/Free Full Text]
Plata C, Meade P, Hall A, Welch RC, Vazquez N, Hebert SC, Gamba G: Alternatively spliced isoform of apical Na(+)-K(+)-Cl(-) cotransporter gene encodes a furosemide-sensitive Na(+)-Cl(-)cotransporter. Am J Physiol Renal Physiol 280: F574–F582, 2001[Abstract/Free Full Text]
Plata C, Meade P, Vazquez N, Hebert SC, Gamba G: Functional properties of the apical Na+-K+-2Cl- cotransporter isoforms. J Biol Chem 277: 11004–11012, 2002[Abstract/Free Full Text]
Gagnon E, Bergeron MJ, Brunet GM, Daigle ND, Simard CF, Isenring P: Molecular mechanisms of Cl- transport by the renal Na(+)-K(+)-Cl- cotransporter. Identification of an intracellular locus that may form part of a high affinity Cl(-)-binding site J Biol Chem 279: 5648–5654, 2004[Abstract/Free Full Text]
Yang T, Huang YG, Singh I, Schnermann J, Briggs JP: Localization of bumetanide- and thiazide-sensitive Na-K-Cl cotransporters along the rat nephron. Am J Physiol 271: F931–F939, 1996
Beierwaltes WH: Selective neuronal nitric oxide synthase inhibition blocks furosemide-stimulated renin secretion in vivo. Am J Physiol 269: F134–F139, 1995
Castrop H, Schweda F, Mizel D, Huang Y, Briggs J, Kurtz A, Schnermann J: Permissive role of nitric oxide in macula densa control of renin secretion. Am J Physiol Renal Physiol 286: F848–F857, 2004[Abstract/Free Full Text]
Takahashi N, Chernavvsky DR, Gomez RA, Igarashi P, Gitelman HJ, Smithies O: Uncompensated polyuria in a mouse model of Bartter’s syndrome. Proc Natl Acad Sci U S A 97: 5434–5439, 2000[Abstract/Free Full Text]
Yang X, Li C, Xu X, Deng C: The tumor suppressor SMAD4/DPC4 is essential for epiblast proliferation and mesoderm induction in mice. Proc Natl Acad Sci U S A 95: 3667–3672, 1998[Abstract/Free Full Text]
Xu X, Li C, Garrett-Beal L, Larson D, Wynshaw-Boris A, Deng CX: Direct removal in the mouse of a floxed neo gene from a three-loxP conditional knockout allele by two novel approaches. Genesis 30: 1–6, 2001[CrossRef][Medline]
Theilig F, Bostanjoglo M, Pavenstadt H, Grupp C, Holland G, Slosarek I, Gressner AM, Russwurm M, Koesling D, Bachmann S: Cellular distribution and function of soluble guanylyl cyclase in rat kidney and liver. J Am Soc Nephrol 12: 2209–2220, 2001[Abstract/Free Full Text]
Castrop H, Huang Y, Hashimoto S, Mizel D, Hansen P, Theilig F, Bachmann S, Deng C, Briggs J, Schnermann J: Impairment of tubuloglomerular feedback regulation of GFR in ecto-5'-nucleotidase/CD73-deficient mice. J Clin Invest 114: 634–642, 2004[CrossRef][Medline]
Kim SM, Mizel D, Huang YG, Briggs JP, Schnermann J: Adenosine as a mediator of macula densa-dependent inhibition of renin secretion. Am J Physiol Renal Physiol 290: F1016–F1023, 2005
Ramsey JA, Brown RHJ, Croghan PC: Electrometric titration of chloride in small volumes. J Exp Biol 32: 822, 1955[Abstract]
Abrahamsberg C, Fuchs P, Osmanagic-Myers S, Fischer I, Propst F, Elbe-Burger A, Wiche G: Targeted ablation of plectin isoform 1 uncovers role of cytolinker proteins in leukocyte recruitment. Proc Natl Acad Sci U S A 102: 18449–18454, 2005[Abstract/Free Full Text]
Antoons G, Ver Heyen M, Raeymaekers L, Vangheluwe P, Wuytack F, Sipido KR: Ca2+ uptake by the sarcoplasmic reticulum in ventricular myocytes of the SERCA2b/b mouse is impaired at higher Ca2+ loads only. Circ Res 92: 881–887, 2003[Abstract/Free Full Text]
Singh S, Mishra R, Arango NA, Deng JM, Behringer RR, Saunders GF: Iris hypoplasia in mice that lack the alternatively spliced Pax6(5a) isoform. Proc Natl Acad Sci U S A 99: 6812–6815, 2002[Abstract/Free Full Text]
Wang B, Hu Q, Hearn MG, Shimizu K, Ware CB, Liggitt DH, Jin LW, Cool BH, Storm DR, Martin GM: Isoform-specific knockout of FE65 leads to impaired learning and memory. J Neurosci Res 75: 12–24, 2004[CrossRef][Medline]
Vargas-Poussou R, Feldmann D, Vollmer M, Konrad M, Kelly L, van den Heuvel LP, Tebourbi L, Brandis M, Karolyi L, Hebert SC, Lemmink HH, Deschenes G, Hildebrandt F, Seyberth HW, Guay-Woodford LM, Knoers NV, Antignac C: Novel molecular variants of the Na-K-2Cl cotransporter gene are responsible for antenatal Bartter syndrome. Am J Hum Genet 62: 1332–1340, 1998[CrossRef][Medline]
Gagnon E, Forbush B, Caron L, Isenring P: Functional comparison of renal Na-K-Cl cotransporters between distant species. Am J Physiol Cell Physiol 284: C365–C370, 2003[Abstract/Free Full Text]
Schnermann J, Briggs J, Schubert G: In situ studies of the distal convoluted tubule in the rat. I. Evidence for NaCl secretion. Am J Physiol 243: F160–F166, 1982
Castrop H, Lorenz JN, Hansen PB, Friis U, Mizel D, Oppermann M, Jensen BL, Briggs J, Skott O, Schnermann J: Contribution of the basolateral isoform of the Na-K-2Cl- cotransporter (NKCC1/BSC2) to renin secretion. Am J Physiol Renal Physiol 289: F1185–F1192, 2005[Abstract/Free Full Text]
Kirchner KA, Kotchen TA, Galla JH, Luke RG: Importance of chloride for acute inhibition of renin by sodium chloride. Am J Physiol 235: F444–F450, 1978
This article has been cited by other articles:
H. Castrop and J. Schnermann Isoforms of renal Na-K-2Cl cotransporter NKCC2: expression and functional significance
Am J Physiol Renal Physiol,
October 1, 2008;
295(4):
F859 - F866.
[Abstract][Full Text][PDF]
P. Welker, A. Bohlick, K. Mutig, M. Salanova, T. Kahl, H. Schluter, D. Blottner, J. Ponce-Coria, G. Gamba, and S. Bachmann Renal Na+-K+-Cl- cotransporter activity and vasopressin-induced trafficking are lipid raft-dependent
Am J Physiol Renal Physiol,
September 1, 2008;
295(3):
F789 - F802.
[Abstract][Full Text][PDF]
F. Hanner, R. Chambrey, S. Bourgeois, E. Meer, I. Mucsi, L. Rosivall, G. E. Shull, J. N. Lorenz, D. Eladari, and J. Peti-Peterdi Increased renal renin content in mice lacking the Na+/H+ exchanger NHE2
Am J Physiol Renal Physiol,
April 1, 2008;
294(4):
F937 - F944.
[Abstract][Full Text][PDF]
S. N. Orlov and A. A. Mongin Salt-sensing mechanisms in blood pressure regulation and hypertension
Am J Physiol Heart Circ Physiol,
October 1, 2007;
293(4):
H2039 - H2053.
[Abstract][Full Text][PDF]
T. Moroy and F. Heyd The impact of alternative splicing in vivo: Mouse models show the way
RNA,
August 1, 2007;
13(8):
1155 - 1171.
[Abstract][Full Text][PDF]
M. Oppermann, D. Mizel, S. M. Kim, L. Chen, R. Faulhaber-Walter, Y. Huang, C. Li, C. Deng, J. Briggs, J. Schnermann, et al. Renal Function in Mice with Targeted Disruption of the A Isoform of the Na-K-2Cl Co-Transporter
J. Am. Soc. Nephrol.,
February 1, 2007;
18(2):
440 - 448.
[Abstract][Full Text][PDF]
Top
Abstract
Introduction
); data are shown for perfusion rates of 6 and 15 nl/min. (B) Mean fractional Cl absorption along the loop of Henle perfused at 6 and 15 nl/min. *P < 0.05.
