TNF inhibits NKCC2 phosphorylation by a calcineurin-dependent pathway

Abstract

We previously demonstrated that TNF inhibits NKCC2 phosphorylation in the thick ascending limb (TAL); however, the underlying mechanism remains unclear. We tested the hypothesis that the induction of calcineurin (CN) activity and the expression of CN isoforms contribute to the mechanism by which TNF inhibits phospho-NKCC2 (pNKCC2) expression. CN activity increased by approximately 2-fold in primary cultures of medullary (m)TAL cells challenged with mouse recombinant TNF. In contrast, silencing TNF production in mTAL cells using lentivirus U6-TNF-ex4 reduced CN activity. pNKCC2 expression decreased in mTAL cells challenged with TNF whereas inhibition of CN activity with cyclosporine A (CsA) increased pNKCC2 expression. Although mTAL cells express both the calcineurin A subunit (CNA) α and β isoforms, only CNA β isoform mRNA increased after mTAL cells were challenged with TNF. In vivo, both TNF and CNA β expression increased in outer medulla (OM) from mice given 1% NaCl in the drinking water for 7 days and intrarenal lentivirus silencing of TNF selectively reduced expression of CNA β. Intrarenal injection of a lentivirus that specifically silenced CNA β (U6-CNAβ-ex6) increased pNKCC2 expression and attenuated the inhibitory effects of TNF on pNKCC2 expression in freshly isolated TAL tubules. Collectively, the study is the first to demonstrate that TNF increases CN activity and specifically induces β-isoform expression in the kidney. Since NKCC2 is a known target of the CNA β isoform, these findings suggest that a CN-dependent signaling pathway involving this isoform contributes to the mechanism by which TNF inhibits pNKCC2 expression.

Graphical Abstract

INTRODUCTION:

Tumor necrosis factor-alpha (TNF) is produced by several renal cell types including proximal tubular (PT), thick ascending limb (TAL), collecting duct (CD), podocytes, and mesangial cells (1–4). It not only exerts significant inflammatory effects but also functions as an immunoregulatory molecule, playing a crucial role in host defense (5–7). TNF also exhibits regulatory effects in various tissues, including the kidney, where active Na⁺ transport plays a critical role in fluid balance and blood pressure regulation (8–10). Previous studies in our laboratory showed that activation of the calcium sensing receptor (CaSR) induced TNF production by mTAL cells (11). Subsequent findings suggested that the inhibitory effects on sodium transport in mTAL cells, mediated via CaSR activation, were TNF-dependent and involved a mechanism that reduced sodium chloride entry into these cells (12, 13). These studies are consistent with data showing that CaSR inhibits the Na⁺-K⁺−2Cl⁻ cotransporter (NKCC2) and suggest TNF may contribute to a mechanism by which CaSR regulates NKCC2, volume homeostasis, and renal function (14, 15).

NKCC2, which is expressed on the apical membrane of the TAL, plays a crucial role in urinary concentration and volume regulation (16, 17). Activity of NKCC2 is regulated by phosphorylation at the cytoplasmic tail at Thr96, Thr101, Ser87, Ser126 (amino-terminus) and Ser874 (carboxy-terminus) via several post-translational mechanisms, including Ste20-related proline-alanine-rich kinase (SPAK), oxidative stress response 1 (OSR1), with-no-lysine kinase (WNK) and calcineurin (CN) (18–21). Calcineurin (CN), a serine/threonine phosphatase that regulates a wide variety of physiological and pathological processes, consists of a catalytic subunit (CNA) and a regulatory subunit (CNB) (22, 23). There also are three different CNA isoforms in mammals, namely CNA α, CNA β, and CNA γ; CNA α and CNA β are ubiquitously expressed while CNA γ is predominantly expressed in brain and testis (22, 23). Elegant studies in the kidney showed that loss of CNA β activity increased the abundance and activity of NKCC2, which may promote electrolyte retention and contribute to increases in blood pressure (18). The CN inhibitor cyclosporine A (CsA) also upregulates NKCC2 in the TAL, thereby increasing Na⁺ reabsorption and contributing to salt-sensitive hypertension (24).

We previously demonstrated that salt-resistant C57BL/6J male mice develop salt-sensitive hypertension when renal cell-derived TNF is silenced (25). Moreover, TNF produced within the kidney and acting on the renal tubular system regulates blood pressure in response to increased sodium intake by inhibiting expression of pNKCC2 and the NKCC2A isoform (25). NKCC2 is a target of CNA β, and inhibition of CN with CsA inhibits TNF production by the TAL, activates NKCC2, and increases blood pressure by several mechanisms (11, 18, 26, 27). These findings suggest that the regulation of the pNKCC2 expression by TNF may involve CN, which contributes to modulation of NaCl reabsorption in the TAL and salt-dependent increases in blood pressure. Hence, the goal of this study was to determine whether TNF-mediated increases in renal CN attenuate pNKCC2 expression in the TAL.

MATERIALS & METHODS:

Mice:

Male and female C57BL/6J mice (8–12 wk) purchased from Jackson Laboratory were maintained on standard diet and given either tap water or 1% NaCl (HS) in the drinking water ad libitum and all healthy animals were eligible for inclusion in the study. Experimental procedures were conducted in accordance with institutional and international guidelines for the welfare of animals (animal welfare assurance number A3362–01 or A5848–01, Office of Laboratory Animal Welfare, Public Health Service, National Institutes of Health). Animal experiments were approved by the New York Medical College Institutional Animal Care and Use Committee (protocol 14036) and all mice completed the protocol and were included in the data analysis.

Materials and reagents:

Lentivirus constructs that specifically silenced CNA β (U6-CNAβ-ex6) and TNF (U6-TNF-ex4) were prepared using the lentiviral vector psiLv-U6 (GeneCopoeia) and the lentivirus purification kit from Takara Bio USA. Lipofectamine reagents, TRIzol, and SDS polyacrylamide gels were from Invitrogen. The pLKO.1, psPAX2, and pMD2.G plasmids were from Addgene (MIT, Cambridge, MA). The lentivirus purification kit (cat. no. 631234) was from Clontech (Mountain View, CA). Tissue culture media were obtained from Life Technologies (Grand Island, NY). DNA shredder was from Qiagen (Valencia, CA). The CN phosphatase activity assay Kit was from Abcam (Waltham, MA). Antibodies for β-actin were obtained from Abcam, Primary antibody for CNA β (PPP3CB, cat. no. 13340–1-AP) was obtained from Proteintech (Rosemont, IL), and anti-phosphorylated (p)NKCC2 was a gift from Drs. Kerim Mutig and James A. McCormick; the latter being an antibody specific to C57BL/6J pT96-NKCC2 with minimal affinity for NCC (28). The Pierce Coomassie (Bradford) Protein Assay Kit and phosphatase inhibitors were from Thermo Scientific (Rockford, IL). The Nonidet P-40 lysis buffer contained protease inhibitors (Roche Diagnostics) for anti-pNKCC2 analysis. All other chemical reagents were purchased from Sigma-Aldrich (St. Louis, MO) unless otherwise specified.

Plasmid and lentivirus constructs:

All constructs and vectors for U6-CNAβ-ex6 or U6-TNF-ex4 were designed and generated using standard cloning procedures and verified by restriction enzyme analysis and DNA sequencing using a protocol similar to that previously described (13). The mouse coding sequence for CNA β was amplified from complementary DNA (cDNA) obtained from primary mTAL cells using a forward primer: 5′- gcaggatccgaattcTGGCCGCCCCGGAGCCG-3′ and a reverse primer: 5′-gcaatggtggtgctcgagTCACTGGGCACTATGGTTGCC-3′ by the PCR. The PCR products were cloned into the pCMV-GFP vector using EcoRI and XhoI sites. The target sequence of the inhibitory construct for CNA β was designed using a short hairpin (sh) RNA-expressing construct targeting exon 6 (CACCTGCATTTGGACCAAT) of murine CNA β (U6-CNAβ-ex6). The target sequence of the silencing construct for TNF (U6-TNF-ex4) was GATGGGTTGTACCTTGTCT; and constructs were designed by targeting exon 4 of the TNF gene. The scrambled U6-shRNA (U6) was used as a negative control. Subcloning of U6 or U6-CNAβ-ex6 into a pLKO.1 vector and cotransfecting HEK293-T cells with pLKO.1 was performed to generate lentivirus encoding U6 or U6-CNAβ-ex6. psPAX2 and pMD2.G plasmids were used for preparation of lentivirus (Addgene MIT, Cambridge, MA).

Murine primary renal mTAL cells:

Murine mTAL cells (90–95% purity) were purified from renal outer medulla (OM) as previously described (29). Briefly, kidneys were perfused with sterile 0.9% saline via retrograde perfusion of the aorta, removed, and cut along the corticopapillary axis; OM was minced, and incubated for 10 min at 37°C in a 0.01% collagenase solution gassed with 95% oxygen. The suspension was sedimented, mixed, then the supernatant containing OM tubules was collected. The collagenase digestion supernatants were filtered through a 52-μm nylon mesh membrane (Fisher Scientific, Springfield, NJ) and filtered tubules were used to establish primary cultures of mouse mTAL cells. Cells were grown with renal epithelial cell growth medium (REGM; Cambrex). After 6–7 days, monolayers of cells were 70–80% confluent; cells were quiesced for 24 h in RPMI before use.

Gene transfection of mTAL cells:

Primary mTAL cells were cultured to 70–80% confluence in 6-well plates with membrane inserts (cell culture inserts; BD Biosciences), the medium was removed, and cells were placed in 1 ml of serum-free OPTI-MEM medium containing different plasmid DNA constructs and 10 μl Lipofectamine 2000 (Invitrogen), for 4 h at 37°C/5% CO₂. After transfection, 1 ml of REGM containing 20% FBS was added, and cells were incubated overnight at 37°C/5% CO₂. The medium was removed, and cells were cultured for an additional 12–48 h in REGM containing 10% FBS.

Gene transfer and TNF treatment in vivo:

A 31-G needle, inserted at the lower pole of each kidney parallel to the long axis, was carefully pushed toward the upper pole in anesthetized mice. As the needle was slowly removed, 50μl filter-purified lentivirus (U6, U6-CNAβ-ex6 or U6-TNF-ex4 ~5 × 10⁷ transducing units) was injected. Previous studies showed that lentiviral-mediated EGFP DNA and protein expression in kidney parenchyma was robust after 72h (30). Similarly, murine recombinant TNF-α (5 ng/g body wt) or saline control was administered by intrarenal injection and primary mTAL cells were pretreated with 1 nM TNF for 30 min before incubation of cells in hypertonic media (400 mOsm/Kg H₂O) or isotonic media for 60 min.

CN activity assay and inhibition with CsA:

CN phosphatase activity was measured using the CN Phosphatase Activity Assay Kit (Abcam, ab139464). Primary mTAL cells were pretreated with 1 μM CsA for 30 min before incubation of cells in hypertonic media (400 mOsm/Kg H₂O) or isotonic media for 60 min (31, 32). Mice were administered CsA (15μg/g body weight) or an equivalent volume of vehicle (castor oil) by intraperitoneal injection for 12 hr, after having received 1% NaCl in the drinking water for 7 days.

Measurement of TNF:

The TNF content in 100 μl of cell-free supernatants was determined in duplicate by ELISA (Pharmingen), as previously described (11). Data were normalized by protein amount, which was determined by a Bradford protein assay (Bio-Rad).

Determination of mRNA levels:

Total RNA was extracted using the RNeasy mini kit (Qiagen, Redwood City, CA) and converted to cDNA using the high-capacity cDNA reverse transcription kit with RNase inhibitor (Thermo Fisher Scientific, MA, USA). The amplification of cDNA fragments, and quantitative real-time PCR analysis (qRT-PCR) were performed using CYBR Green Chemistry. The specific primer pairs for murine CNAα (forward: 5′-GACTGAGATGCTGGTCAATG-3′, reverse: 5′-GAGAACTGAGAACACTCTGG-3′); murine CNAβ (forward: 5′-GGATTCTCTCCACCACATAG-3′, reverse: 5′-GAAGTGATCTGTCCATTTGGG-3′), or murine TNF (forward: 5′-CACCACGCTCTTCTGTCTAC-3′, reverse: 5′-TTGAGATCCATGCCGTTGGC-3′) were designed based on murine accession no NM_008913.6, accession no. NM_008914.3, and accession no. NM_0.13693.3, respectively. Input cDNAs were normalized using housekeeping gene β-actin (forward: 5′-GACCTCTATGCCAACACAGT-3′; reverse: 5′-GACTCATCGTACTCCTGCTT-3′), and the efficiency of primer pair amplification determined using a standard curve generated using serially diluted plasmid DNA. The transcript quantities were compared by using the relative Ct method; relative mRNA expression levels were calculated by the 2(^−ΔΔCT) method (33, 34).

Western blotting analysis:

Briefly, pNKCC2 or CNA β expression was determined by Western blotting of mTAL cells or renal tissue as indicated. These samples were washed three times with ice-cold PBS, then homogenized with RIPA extraction buffer (GBiosciences, MO) containing phosphatase inhibitors on ice, for mTAL cells, or sucrose extraction buffer containing phosphatase inhibitors on ice for the renal tissue (35, 36). The protein extraction solution was vortexed for 10s and centrifuged, then the protein concentration of the supernatants was determined with a Bio-Rad protein assay kit. Equal amounts of protein (30 μg/lane) were separated by SDS-PAGE and electrophoretically transferred to nitrocellulose membranes. Following blocking with 5% milk, membranes were probed at 4°C overnight with appropriate primary antibodies. Membranes were incubated with horseradish peroxidase-conjugated secondary antibodies (1:2000, Bio-Rad, Hercules, CA) for 1 h at room temperature. Visualization was performed using the ECL Prime Western Blotting Detection Reagent (Amersham, Pittsburgh, PA) and bands analyzed using a UVP BioImaging System. Since total NKCC2 levels do not change under the experimental conditions used in this study, pNKCC2 expression was normalized to the houskeeping protein β-actin, except where indicated.

Experimental study design and statistical analysis:

All experiments were repeated at least three times and samples were assayed as duplicates. The data were acquired using randomized procedures for mice selection and samples were coded then analyzed in a blinded manner. Statistical analysis of data was computed using Graphpad Prism. One-way ANOVA (followed by Tukey or Dunnett post hoc test) was used for comparisons among groups, and an unpaired Students t-test was used to compare data between 2 groups, with the aid of Sigma Plot software. Power analysis provided guidance on the number of mice used in each experiment and data are expressed as mean ± SE; a P value of <0.05 was considered statistically significant.

RESULTS:

TNF increases CN phosphatase activity in mTAL cells:

CN activity was determined in cytoplasmic extracts from control and TNF-treated mTAL cells incubated in the absence or presence of excess NaCl. CN activity increased in mTAL cells challenged with 1 nM TNF for 60 min in either isotonic media or hypertonic media (400 mOsm/Kg H₂O), where response of the cells to TNF was enhanced with respect to isotonic media (Fig. 1A). Meanwhile, an shRNA targeting exon 4 of the TNF gene was used to prepare a lentivirus silencing construct (U6-TNF-ex4) (25). ELISA analysis confirmed that U6-TNF-ex4 significantly inhibited TNF production in primary cultures of mouse mTAL cells under high salt conditions (Fig. 1B). Furthermore, silencing TNF in mTAL cells reduced CN activity in cells exposed to hypertonic media for 60 min (Fig. 1C). The effect of TNF on CN activity was then evaluated in the presence of the CN inhibitor cyclosporine A (CsA). Cells were pretreated with CsA (1 μM) in isotonic media for 30 min before they were incubated in hypertonic media with or without 1 nM TNF for 1h. The TNF-mediated increase in CN activity was absent in mTAL cells preincubated with CsA (Fig. 1D). Collectively, these results illustrate that both exogenous and endogenous TNF increase CN activity in mTAL cells.

Figure 1: TNF increases CN phosphatase activity in primary murine mTAL cells.

(A): mTAL cells were incubated in isotonic media or in hypertonic media (400 mOsm/Kg H₂O) with or without TNF for 60 min; (B): TNF was measured by ELISA in cell-free supernatants from mTAL cells challenged with hypertonic media after transduction with U6-TNF-ex4 or control vector (U6) for 24 h (C): mTAL cells were treated in hypertonic media after preincubation with lentivirus (U6-TNF-ex4) for 24 h (D): mTAL cells were incubated in hypertonic media with or without TNF for 60 min after pretreatment in isotonic media for 30 min with CsA (1 μM). CN activity in the lysates was determined by measuring phosphate release using a CN phosphatase assay kit. Data are shown as mean ± SE (n=6).

Effects of TNF and CsA on pNKCC2 expression in mTAL cells:

The effects of TNF on pNKCC2 expression were determined in mTAL cells via stimulation with recombinant TNF and by silencing endogenous TNF production with U6-TNF-ex4. NKCC2 phosphorylation decreased in cells challenged with TNF (1 nM) incubated in hypertonic media for 4hr (Fig. 2A). In contrast, silencing endogenous TNF production in mTAL cells with lentivirus U6-TNF-ex4 increased pNKCC2 expression in mTAL cells (Fig. 2B). Cells were then treated with CsA to determine the effects of CN inhibition on pNKCC2 expression. CsA, which blocks TNF production and TNF-mediated increases in CN activity in mTAL cells, increased pNKCC2 expression (Fig. 2C). Similar to a previous in vivo study (25), additional data confirmed that silencing endogenous TNF production in mTAL cells using the lentivirus U6-TNF-ex4 does not alter total NKCC2 expression (Fig. 2D). These data indicate that TNF attenuates NaCl-dependent increases in pNKCC2 expression.

Figure 2: TNF inhibits pNKCC2 expression in mTAL cells.

mTAL cells were incubated in hypertonic media (400 mOsm/Kg H₂O) for 4 h after (A) challenge with or without TNF in isotonic media for 60 min, (B) preincubation with lentivirus (U6-TNF-ex4) in isotonic media for 24 h, (C) challenge with or without CsA (1 μM) in isotonic media for 30 min, (D) preincubation with lentivirus (U6-TNF-ex4) in isotonic media for 24 h. pNKCC2 or total NKCC2 expression was measured by immunoblot analysis and was normalized using β-actin (mean ± SE). Representative blots are shown (n=6).

Differential effects of TNF on calcineurin A isoforms in mTAL cells:

To explore additional mechanisms by which TNF may influence CN, the effects of this cytokine were evaluated on the expression of CN A isoform mRNA. Expression of both the calcineurin A subunit (CNA) isoforms α (Sequence ID: NM_008913.6) and β (sequence ID: NM_008914.3) was detected in mTAL cells. Treatment of cells with TNF under isotonic conditions did not alter mRNA levels of CNA α isoform compared to untreated cells (Fig. 3A). However, CNA β isoform mRNA levels were increased by TNF (Fig. 3B). A similar differential effect of TNF on CNA isoforms was observed in cells challenged with TNF under hypertonic conditions (Fig. 3C&D). In addition, we confirmed that TNF induced CNA β protein expression in cells under hypertonic conditions (Fig. 3E). These findings indicate that TNF selectively increases CNA β, an isoform previously linked to the regulation of pNKCC2 expression.

Figure 3: Effects of TNF on CNA isoform expression in mTAL cells.

mTAL cells were challenged without or with TNF in isotonic media and qRT-PCR was used to evaluate α isoform (A) or β isoform of CNA mRNA levels (B). CNA α isoform (C) or CNA β isoform (D) expression in mTAL cells in response to high salt (400 mOsmol/kg H₂O) for 4 hours after challenge with or without TNF in isotonic media for 60 min was determined by qRT-PCR; and (E) CNA β protein expression in mTAL cells in response to high salt (400 mOsmol/kg H₂O) for 4 hr after challenge with or without TNF in isotonic media for 60 min was determined by western blot analysis; representative lanes are shown. Data are shown as mean ±SE (n=6).

shTNF treatment in vivo downregulates CNA β isoform expression in the TAL:

To determine whether induction of CNA isoform expression is affected by TNF in vivo, C57BL/6J male and female mice were administered control (U6) or TNF silencing (U6-TNF-ex4) lentivirus via intrarenal injection 3 days before they were given access to 1% NaCl in the drinking water (HS) for 7 days. For in vivo silencing of TNF, qRT-PCR analysis confirmed that TNF mRNA was markedly reduced in the outer medulla (OM) but not the spleen when the purified U6-TNF-ex4 or control lentivirus (U6) was injected directly into each kidney (Fig. 4A), which is consistent with the degree of TNF silencing previously observed using this approach (25). Intrarenal silencing of TNF had no effect on CNA α isoform expression in OM (Fig. 4B) and mTAL tubules (Fig. 4C). However, levels of CNA β mRNA were markedly reduced when TNF was silenced in the kidney (Fig. 4D&E). These data suggest that renal TNF is a specific inducer of the CNA β isoform in the TAL.

Figure 4: Effects of silencing renal TNF in vivo on CNA isoforms.

qRT-PCR was used to evaluate the effect of U6-TNF-ex4 on TNF mRNA levels in OM and spleen (A), CNA α mRNA in OM (B) and mTAL tubules (C), CNA β in OM (D) and mTAL tubules (E) from male and female mice given 1% NaCl in the drinking water for 7 days after pretreatment with intrarenal injection of U6 (control) or lentivirus U6-TNF-ex4 into both kidneys. Data are shown as mean ± SE (n=8).

Upregulation of pNKCC2 following CsA and CNA β silencing in vivo:

Although data from animal models suggest that CNA β is important for the regulation of NKCC2 phosphorylation, no studies have determined the effects of directly silencing this isoform in the kidney on pNKCC2 expression. To corroborate the relationship between dephosphorylation effects on NKCC2 and a CNA β signaling pathway in vivo, mice were treated with CsA using a dose that increases NaCl delivery to the TAL while promoting NKCC2 phosphorylation (24). Administration of CsA decreased CNA β and increased pNKCC2 expression in mTAL tubules compared with controls (Fig. 5A&B). Subsequently, a U6-CNAβ-ex6-specific lentivirus, designed based on the Ppp3cb Gene (ID: 19056), was used to silence CNA β in the kidney. Mice were administered U6 (control) or U6-CNAβ-ex6 (~5 × 10⁷ transducing units) by intrarenal injection and 3 days later were allowed access to 1% NaCl for an additional 7 days. Treatment with U6-CNAβ-ex6 inhibited CNA β mRNA levels in OM (Fig. 5C) but did not affect CNA α isoform mRNA (Fig. 5D) indicating this lentivirus specifically silences the CNA β isoform.

Figure 5: Effects of CsA and CNA β silencing on NKCC2 phosphorylation in vivo.

CNA β mRNA (A) and pNKCC2 expression (B) in mTAL tubules was assessed in control mice and those treated with CsA for 12 hr after 1% NaCl was given in the drinking water for 7 days. CNA β mRNA (C), and CNA α mRNA (D) in OM was measured in mice given 1% NaCl for 7 days after pretreatment with intrarenal injection of lentivirus construct U6-CNAβ-ex6 or U6 into both kidneys. Data are shown as mean ± SE (n=6) and representative blots are shown for pNKCC2.

Inhibition of pNKCC2 expression in vivo via a TNF-CNA β signaling pathway:

To define whether induction of CNA β is part of a CNA-dependent mechanism by which TNF inhibits NKCC2 activity in vivo, mice were treated with TNF (5ng/g body weight) by injection into each kidney for 3 days and then given 1% NaCl in the drinking water for an additional 7 days. Administration of TNF did not alter CNA α isoform mRNA abundance however, it increased CNA β isoform mRNA in mTAL tubules (Fig. 6A&B). Experiments were then performed to analyze the expression of pNKCC2 in mTAL tubules from mice that were given either TNF alone or a mixture of U6-CNAβ-ex6 and TNF for 3 days prior to access to 1% NaCl for 7 days. Silencing of CNA β significantly increased pNKCC2 expression in mTAL tubules indicating this isoform is part of a pathway that attenuates NKCC2 phosphorylation in the mTAL. The data also indicate that pNKCC2 expression was lower in mice that received TNF treatment compared with control mice given saline, and that concomitant silencing of CNA β prevented the inhibitory effects of TNF on pNKCC2 expression (Fig. 6C). Collectively, the results suggest that induction of a CNA β signaling pathway is a key mechanism by which TNF inhibits NKCC2 phosphorylation.

Figure 6: TNF inhibits pNKCC2 expression via a CNA β-dependent mechanism.

CNA α mRNA (A), CNA β mRNA, (B), and pNKCC2 expression (C) were measured in mTAL tubules following intrarenal injection of TNF. For pNKCC2 assessment, mice received intrarenal injections of TNF, U6-CNAβ-ex6, or a combination of both followed by ingestion of 1% NaCl for 7 days. Data are presented as mean ± SE (n=6), with representative blots shown for pNKCC2.

DISCUSSION:

We demonstrated that TNF increases CN activity as well as specific expression of the CNA β isoform in the TAL under both normal and high salt conditions. The data indicate that NKCC2 phosphorylation was significantly decreased by TNF, while inhibition of CN activity with CsA, which inhibits TNF production by the TAL (11), decreased CNA β and increased pNKCC2 expression. Moreover, lentivirus silencing of TNF increased pNKCC2 both in cultured mTAL cells and tubules in vivo. Lentivirus silencing of CNA β also increased pNKCC2 expression in mTAL cells under high salt conditions and in mTAL tubules from mice given 1% NaCl in the drinking water, suggesting that the increase in NKCC2 activity under these conditions is CNA β-dependent. Silencing TNF in the kidney decreased CNA β and increased pNKCC2 expression while silencing CNA β prevented TNF-mediated decreases in pNKCC2. Since NKCC2 has been identified as a target of the CNA β isoform, the present study indicates that a CNA β-dependent pathway is part of the mechanism by which TNF inhibits the phosphorylation of NKCC2. This mechanism likely contributes to an autocrine effect of renal tubular TNF, protecting salt-resistant C57Bl6/J mice from salt-induced blood pressure increases, and may also play a role in CsA-mediated hypertension.

The diversity of CN functions in the kidney, and other tissues, is reflected by the division of its catalytic subunit into the three distinct isoforms α, β, and γ, which subserve different functions (24, 37, 38). For instance, CNA α is the primary isoform expressed in renal cortex, glomeruli, and collecting duct where it likely participates in the regulation of AQP2 expression, phosphorylation, and function (39). On the other hand, genetic knockout of CNA β in a setting of Type I diabetes attenuated renal hypertrophy without affecting the regulation of matrix proteins (40). Further, nuclear translocation of calcium-dependent NFAT isoforms is regulated by CNA β, and a CNA β/NFAT pathway regulates NADPH oxidase expression and activity in response to high glucose (41, 42). We previously showed that TNF inhibits NKCC2 activity in the TAL (25), however, the mechanisms by which TNF inhibits pNKCC2 expression have not yet been defined. The present study was prompted by recent findings showing that NKCC2 is a target of CNA β via interaction with cytoplasmic NKCC2 tails and, to our knowledge, is the first to determine the effects of TNF on CN or its isoforms in the kidney, although TNF was shown to induce CNA α expression in human neuroblastoma cells and CN activity in synoviocytes (18, 43, 44). TNF exhibited short- and long-term regulatory effects in the TAL as both CN activity and specific induction of CNA β mRNA were observed, however the molecular mechanisms that underlie these effects remain to be determined. The mechanism by which TNF differentially regulates CNA β and CNA α mRNA in the kidney also is currently unknown. Cell-type specific control of CN isoforms may enable the regulation of individual transporters by different mechanisms. For instance, while CNA β is expressed in the TAL, tacrolimus, another calcineurin inhibitor, appears not to alter pNKCC2 expression but instead increases phosphorylation of NCC in the DCT by a mechanism involving WNK and SPAK (18, 27). As CNA β may not be expressed in the DCT, perhaps the effects of tacrolimus are mediated indirectly via its interaction with CNA α. Unlike calcineurin inhibitors, direct lentivirus silencing of CNA β in vivo enhanced NKCC2 phosphorylation in the TAL without affecting multiple CN isoforms. Overall, the present data indicate that a TNF/CNA β pathway regulates NKCC2 phosphorylation in the TAL under normal and high salt conditions.

Reabsorption of NaCl along the TAL occurs mainly via NKCC2 expressed in the apical membrane of the TAL (45, 46); ~30% of the total filtered NaCl is reabsorbed along the TAL (47, 48). Mutations of the NKCC2 gene (SLC12A1) that cause reduced NKCC2 activity; and detected by screening >3000 members of the Framingham Heart Study, were associated with a significant reduction in blood pressure and risk of death because of cardiovascular disease (49, 50). Trafficking of NKCC2 to the apical membrane depends on its distal COOH-terminal region, which facilitates cell surface expression, transport activity, and is regulated by phosphorylation of the N-terminus at threonine and serine residues (Thr96 and Thr101) (51, 52). Mutagenic analysis also demonstrated that deleting threonines nearly eliminates baseline NKCC2 activity, which is associated with intracellular chloride depletion (52, 53). Moreover, NKCC2 phosphorylation is increased by protein kinases such as SPAK or OSR1 (19, 20, 52, 54). Further, NKCC2 phosphorylation and activity are regulated by high salt, although the contribution of various autacoid mediators is still being defined in this situation (55). CN also regulates NKCC2 activity by promoting its dephosphorylation (18), which is consistent with the present data in both cultured mTAL cells and freshly isolated TAL tubules. Interestingly, the commonly used inhibitors of CN, CsA and tacrolimus, may differentially regulate NKCC2 phosphorylation as tacrolimus-induced hypertension, which is mainly mediated by NCC activation, was not associated with an increase in NKCC2 phosphorylation (27). Although, in a mouse model of type 4 Bartter syndrome, treatment with tacrolimus increased pNKCC2 levels (56). CsA on the other hand increases NKCC2 phosphorylation in the TAL and chronic treatment with CsA was shown to induce hypertension by upregulating NKCC2 expression and activity, thereby increasing reabsorption of NaCl along the TAL in mice, rats, and humans (24, 26, 57, 58). Collectively, these data agree with the present results and suggest that TNF may be one of several endogenous autacoid, hormone, and cytokine mediators that regulate NKCC2 phosphorylation in various settings. Whether TNF interacts with other autacoids and contributes to findings that the CsA induced stimulation of the renin-angiotensin-aldosterone system (RAAS), or inhibition of COX-2 may synergistically enhance salt reabsorption along the distal nephron thereby contributing to the development of calcineurin inhibitor-induced hypertension, remains to be determined (59–61). Since CsA inhibits TNF production in the TAL and other cell types (11, 62–64), and inhibition of CN induces a salt-sensitive form of hypertension due to increased Na⁺ reabsorption along the TAL [24], the suppression of endogenous TNF produced by TAL cells and its ability to sustain CN activity and CNA β expression in the TAL may contribute to observed increases in NKCC2 phosphorylation and blood pressure.

The in vitro and in vivo effects of TNF on NKCC2 associated with the differential regulation of CNA β isoform expression and increase in CN activity has uncovered an intrinsic relationship between TNF and CN that facilitates a signaling pathway in mice during adaptation to HS intake. This pathway serves to decrease NKCC2 phosphorylation, as well to increase CN-dependent TNF production by the TAL under high salt conditions, which is consistent with data showing that inhibition of CN with CsA increases pNKCC2 expression and blood pressure. These data contribute to understanding the mechanisms by which TNF regulates NKCC2 and identifies a role for the CNA β isoform in the renal response to high salt conditions.