Introduction

20-Hydroxyeicosatetraenoic acid (20-HETE) is a potent vasoconstrictor that regulates vascular tone in arterioles through blocking Ca2+-activated K+ channel; however, it is also an inhibitor of tubular sodium reabsorption in the proximal tubule and in the thick ascending limb of Henle’s loop (TALH) via inhibiting the activities of Na+-K+-ATPase, Na+/H+ exchanger (NHE3) and Na+-K+-2Cl− cotransporter, isoform 2 (SLC12A1; NKCC2) (Imig 2004; Roman 2002). Therefore, 20-HETE plays a pivotal role in the development of hypertension via either vascular reactivity or ion transport determined by other risk factors for hypertension including high salt intake, alcohol overuse, and excessive body weight. Considerable human studies have indicated that 20-HETE level varies in individuals depending on some functional polymorphisms of cytochrome P450 4F2 (CYP4F2) and cytochrome P450 4A11 (CYP4A11) which are the major genes involving in the ω-hydroxylation of arachidonic acid to 20-HETE (Roman 2002). We have reported that a functional haplotype of the CYP4F2 promoter with increased transcriptional activity is associated with elevated urinary 20-HETE and hypertension in a Chinese population (Liu et al. 2008). Another variant V433M in CYP4F2 is identified to decrease 20-HETE production (Stec et al. 2007) and is associated with the increase of urinary 20-HETE excretion and systolic blood pressure (SBP) (Hu et al. 2011; Ward et al. 2008). As to the CYP4A11 gene, the F434S variant is found to decrease the formation of 20-HETE and is linked with lower urinary 20-HETE excretion, but not with blood pressure (Gainer et al. 2005; Ward et al. 2008). Despite the fact that 20-HETE per se has dual role on blood pressure and elicits discrepant association with hypertension, population could be classified into high and low 20-HETE groups so as to attribute 20-HETE to different types of hypertension. To better understand the overall effect of diverse levels of 20-HETE on blood pressure in vivo, we have established a CYP4F2 transgenic mouse model with increased expression and catalytic activity of CYP4F2 as well as elevated 20-HETE production, and further uncovered that high 20-HETE positively correlated with elevated SBP in the CYP4F2 transgenic mice with normal salt intake (Liu et al. 2009, 2012). Hence, we could use the CYP4F2 transgenic mice with high 20-HETE level and wild-type mice with low 20-HETE level to address the interplay of 20-HETE with dietary salt in the development of hypertension. Such simulation could help decipher the molecular mechanism of salt-related hypertension underlying the genetic determinants that interact with environmental factors.

The Na+ transport molecules targeted by 20-HETE are Na+-K+-ATPase, NHE3, and NKCC2 in kidney. In proximal tubules, 20-HETE inhibits renal Na+-K+-ATPase activity by activating protein kinase C to phosphorylate its α-subunit (Imig 2004; Roman 2002) and also inhibits NHE3 activity by internalizing its protein in this segment (Roman 2002). In TALH, NKCC2 is localized in the apical membrane domains and accounts for about 80 % of the absorbed Na+ that enters TALH. 20-HETE has been shown to block apical membrane 70-pS K+ channel (ROMK) that maintains the K+ gradient for NKCC2, so as to suspect that 20-HETE inhibits the NKCC2 activity by the limit of ROMK (Wang and Lu 1995). So far, the molecular mechanism whereby 20-HETE inhibits NKCC2 activity of the outer medullary and cortical TALH remains ambiguous. Therefore, in the present study, we aim to investigate the interplay between 20-HETE and salt on NKCC2 and natriuresis in the CYP4F2 transgenic mice comparing with wild-type mice and to elucidate the potential pathway by which 20-HETE and salt modulate NKCC2 expression and blood pressure.

Materials and methods

Animals

The CYP4F2 transgenic mice overexpressing CYP4F2 was a FVB strain (Liu et al. 2009). Experiments were performed on 12- to 16-week-old male transgenic mice weighing between 24 and 30 g. All mice were weight- and age-matched with littermate wild-type FVB mice as controls. Mice were fed with a diet containing either normal salt (0.4 % NaCl) or high salt (4 % NaCl) for 2 weeks. All animal experimental protocols were conducted in accordance with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996). For 20-HETE inhibition, mice were administered intraperitoneally either HET0016 (Cayman Chemical, USA) by 10 μg/g body weight daily or lecithin (Roche Applied Science, Switzerland) vehicle (10 % weight per volume lecithin in saline) in the second week of 2-week high salt intake. For proteasome inhibition, mice were administered intraperitoneally either MG132 (Sigma-Aldrich, USA) by 5 μg/g body weight daily or DMSO (Sigma-Aldrich, St Louis, USA) vehicle (2 % volume per volume DMSO in saline) for the last 3 days of 2-week high salt intake.

Measurement of blood pressure

The blood pressure was measured by the tail-cuff method, using IITC Life Science Model 1631 tail pulse detection system (IITC Life Science Model 1631, USA) according to the manufacturer’s instructions. After 5 days of training, the blood pressure was recorded from conscious mice that were fed with either a normal salt or a high salt diet. One measurement session involved ten repetitions, and at least three sessions were performed on each mouse to calculate the average.

Urine collection and analysis of urinary sodium concentration

Mice were housed in stainless steel metabolic cages and a 24-h urine sample was collected in bottles cooled with ice to prevent the breakdown of 20-HETE. Urinary sodium levels were measured on a Hitachi 7600-110 analyzer (Hitachi High-Technologies Corp., Japan).

Analysis of urinary 20-HETE excretion

Urinary 20-HETE excretion was quantified by ABI 3200 Q-trap LC–MS/MS System (Applied Biosystems, USA). After adding 20-HETE-d6 (Cayman Chemical, USA) as internal standard, the urine samples were incubated at 37 °C for 2 h with 0.1 mg/ml β-glucuronidase (Sigma-Aldrich, St. Louis, USA), and then urinary 20-HETE was extracted with ethyl acetate. The ethyl acetate was evaporated under nitrogen, and the metabolites were resuspended in methanol. Samples were separated on a reversed-phase Symmetry C18 column (3.5 μm, 2.1 × 150 mm; Waters, Milford, MA) at a flow rate of 0.2 ml/min using solvent A (water, 0.1 % formic acid) and solvent B (acetonitrile:methanol = 6:1, 0.1 % formic acid) (0–2 min 25 % B, 2–10 min 25–75 % B, 10–18 min 75–95 % B, 18–30 min 95 % B, 30–30.5 min 95–25 % B, 30.5–40 in 25 % B). The effluent was ionized using negative ion electrospray and quantified by multiple reaction monitoring. The ratio of ion abundance in the peaks versus that seen in the internal standard was determined and compared with standard curves generated over the range from 0.2 to 10 ng for 20-HETE.

Real-time PCR

Total RNA was extracted from kidneys using TRIzol reagent (Invitrogen, Carlsbad, USA), and reverse transcribed into cDNA with Reverse Transcription Reagent Kit (Promega, Madison, USA). Real-time PCR was performed on the ABI 7500 System (Applied Biosystems, Foster City, USA) in a 20 μl SYBR Green PCR containing 1× SYBR Green PCR master mix (Applied Biosystems, Foster City, USA), 10 ng cDNA, and 100 nM forward and reverse primers. The sequences of the used primers were 5′-GAGATTGGCGTGGTCATAGTCAGAA-3′ (forward) and 5′-TGCTGCTGATGTTGCCGTCTTT-3′ (reverse) for NKCC2; 5′-TACCAACTGGGACGACATGG-3′ (forward) and 5′-GGAGTCCATCACAATGCCTG-3′ (reverse) for β-actin. Samples were subjected to 40 cycles of two temperature steps as follows: 95 °C for 15 s, 60 °C for 1 min. Dissociation curves were generated to insure that a single and specific product was amplified. Cycle threshold values (Ct) were analyzed by the SDS2.4 software (Applied Biosystems, Foster City, USA), and relative quantification of NKCC2 expression was determined using the comparative Ct method with the β-actin transcript as an internal control.

Immunoprecipitation (IP) and Western blot (WB)

Renal protein samples were prepared by homogenizing the frozen tissues in lysis buffer containing protease inhibitors, and the concentration was determined by the Bradford method. The protein samples were pre-incubated with primary antibody by rotating at 4 °C overnight, followed by the addition of 20 μl equivalents of protein A/G PLUS-Agarose (Santa Cruz Biotechnology, Santa Cruz, USA) and rotating for 1 h. Protein A/G beads were collected and washed with lysis buffer three times. Immunoprecipitates were resolved by SDS-polyacrylamide gel electrophoresis (PAGE) and analyzed by WB. In WB analysis, the protein samples were subjected to 8, 10 or 12 % SDS-PAGE and transferred onto polyvinylidene difluoride (PVDF, Bio-Rad, Hercules, USA) membranes. The membranes were blocked with 5 % nonfat dry milk in TBS containing 0.1 % Tween-20 and incubated with primary antibodies and then with horseradish peroxidase-conjugated secondary antibodies, according to the manufacturer’s instructions. WB analysis was visualized with the enhanced chemiluminescence (ECL) kit obtained from Thermo Scientific. Primary antibodies included NKCC2 (Upstate Biotech Millipore, Lake Placid, USA), ubiquitin (Cell Signaling Technology, Danvers, USA), Nedd4-2 (Abcam, Cambridge, USA) and β-actin (Santa Cruz Biotechnology, Santa Cruz, USA).

Immunofluorescence (IF)

Paraffin embedded renal tissue was blocked with 0.3 % hydrogen peroxide and 10 % BSA/PBS before incubation with the NKCC2 antibody overnight. The sections were washed and incubated with biotinylated FITC-secondary antibody (ZSGB-BIO) for 1 h, according to the manufacturer’s protocol. The sections were viewed with Nikon Eclipse 80i fluorescence microscope (Nikon corp., Japan), at a magnification of 100×.

Proteasome activity measurement

Peptidase activity of the proteasome was measured by mixing tissue homogenate protein (100 μg) with 100 μmol/l fluorogenic peptide Suc-LLVY-AMC (succinyl-Leu-Leu-Val-Tyr-7-amino-4-methylcoumarin, chymotrypsin-like; Enzo Life Sciences, Farmingdale, USA) to a final volume of 100 μl. The reaction buffer consisted of 50 mmol/l Tris–HCl (pH 7.5), 20 mmol/l KCl, 5 mmol/l MgCl2 and 5 mmol/l ATP. The mixture was incubated at 37 °C for 20 min, and then the reaction was stopped by adding an equal volume of 125 mmol/l sodium borate buffer (pH 9.0) containing 7.5 % ethanol. The released fluorogenic AMC was measured in a Tecan, M200pro fluorescence plate reader, with the excitation wavelength set at 360 nm and the emission wavelength at 460 nm (Tecan Trading AG, CH).

Statistical analysis

Data are expressed as mean ± SEM from at least three independent experiments and analyzed by SPSS (Version 17.0 for Windows; SPSS, Inc., Chicago, IL, USA). To examine the effects of two factors, genetic background of 20-HETE and dietary salt, mean values were calculated for each animal and were subjected to 2-way ANOVA. Statistical significance was set at p < 0.05.

Results

Synergy of 20-HETE and high salt intake in the regulation of NKCC2 expression

To gain insights into the mechanism for adaptive modulation of 20-HETE to high salt intake in the transgenic mice, we fed the wild-type and CYP4F2 transgenic mice a high salt diet for 2 weeks and compared the expression of renal NKCC2 protein, which is critical for Na+ reabsorption. Figure 1a summarizes the expression of renal NKCC2 in response to either normal salt (0.4 % NaCl) or high salt (4 % NaCl) intake in the wild-type and transgenic mice. The abundance of NKCC2 protein was lower in the transgenic mice than in the wild-type mice with normal salt intake. When mice were fed a high salt diet, NKCC2 abundance was reduced about 85 % in the transgenic mice and 24 % in the wild-type mice, compared with the normal salt diet. These results suggest that 20-HETE or high salt intake inhibits NKCC2 expression, and their synergistic actions dramatically exaggerate the reduction of NKCC2.

Fig. 1
figure 1

20-HETE and high salt intake regulate renal NKCC2 protein expression. Renal NKCC2 protein expression is detected by Western blot from a wild-type (WT) and transgenic (TG) mice fed either a normal salt (NS) or a high salt (HS) diet for 2 weeks, and b TG mice treated with HET0016 in the second week of 2-week NS or HS intake. Representative Western blot of NKCC2 is shown. The bar chart denotes mean ± SEM of quantitative densitometric scan from three independent experiments of three mice in each group. c Immunofluorescence analysis: micrographs are taken at a magnification of ×100, and the bar represents 100 μm. *p < 0.05 versus untreated value of the respective group of mice; #p < 0.05 versus values of corresponding WT mice; **p < 0.05 versus values of corresponding NS diet mice

Full size image

We infused the transgenic mice with HET0016, a selective inhibitor of the synthesis of 20-HETE, in the second week of high salt intake and further confirmed that the interaction of 20-HETE with high salt intake led to the fall of NKCC2 protein in the kidneys of the transgenic mice. HET0016 significantly reversed the decrease of NKCC2 protein resulting from high salt intake (Fig. 1b).

These findings are consistent with immunofluorescence images obtained from the kidneys of the wild-type and transgenic mice. Representative examples of the NKCC2 protein distribution and abundance seen in the renal cortex and outer medulla are presented in Fig. 1c. NKCC2 protein in the transgenic mice with high salt intake showed a marked reduction, and the reduction could be reversed by HET0016. The obvious change of NKCC2 expression in the kidneys of the transgenic mice with high salt intake indicates a synergy between 20-HETE and high salt intake in the regulation of NKCC2 protein expression.

mRNA level of NKCC2 in the CYP4F2 transgenic mice with high salt intake

Next, we investigated whether the reduction of NKCC2 protein in the transgenic mice with high salt intake resulted from transcriptional regulation. We examined renal NKCC2 mRNA levels by real-time PCR. As shown in Fig. 2, mRNA levels of NKCC2 were significantly lower in the transgenic than in the wild-type mice with normal salt intake, in accordance with the change of NKCC2 protein. However, high salt intake did not affect the NKCC2 mRNA in the transgenic mice, although it increased moderately the NKCC2 mRNA in the wild-type mice; this observation was inconsistent with the change of NKCC2 protein. We hypothesize that a protein modification might be responsible for high salt intake, leading to the reduction of NKCC2 protein in the transgenic mice.

Fig. 2
figure 2

Renal NKCC2 mRNA expression responses to high salt intake. NKCC2 mRNA levels are quantified by real-time PCR in the kidneys of the wild-type (WT) and transgenic (TG) mice fed either a normal salt (NS) diet or a high salt (HS) diet for 2 weeks, and are normalized by the level of β-actin mRNA. The experiments are performed three times independently from four mice in each group

Full size image

Ubiquitination of NKCC2 in the CYP4F2 transgenic mice with high salt intake

Protein ubiquitination is a key regulatory process essential to cellular metabolism in response to injurious stimuli. To investigate whether NKCC2 is ubiquitinated and reduced by 20-HETE and high salt intake, we examine the ubiquitination of NKCC2 by immunoprecipitation (IP) in mice with high salt intake compared with those with normal salt intake. Ubiquitinated NKCC2 protein was induced by high salt intake and was significantly higher in the transgenic mice than in the wild-type mice (Fig. 3). This result suggests that high salt intake facilitates ubiquitination of NKCC2 and that 20-HETE enhances the ubiquitination level of NKCC2 protein.

Fig. 3
figure 3

Ubiquitination of renal NKCC2 by immunoprecipitation (IP). IP with NKCC2 antibody, followed by Western blot (WB) with ubiquitin/NKCC2 antibody detect ubiquitinated NKCC2 protein of wild-type (WT) and transgenic (TG) mice fed with either a normal salt (NS) or a high salt (HS) diet. A representative Western bolt is shown. The experiments are performed three times independently from three mice in each group

Full size image

Reduction of NKCC2 protein via the ubiquitin–proteasome pathway in the CYP4F2 transgenic mice with high salt intake

Ubiquitinated protein usually undergoes proteasomal degradation. To investigate the possible pathway of the degradation of NKCC2 protein resulting from high salt intake in the transgenic mice, we detected proteasome activity in kidneys. The proteasome activity was significantly increased in the wild-type and transgenic mice with high salt intake compared with those with normal salt intake and was also significantly higher in the transgenic mice than in the wild-type mice (Fig. 4a). To determine whether the reduction of NKCC2 protein is the result of proteasomal degradation in the transgenic mice with high salt intake, we treated mice with a specific proteasome inhibitor MG132, in the last 3 days of 2-week high salt intake. MG132 abolished the reduction of NKCC2 protein in mice with high salt intake, showing an increase of NKCC2 protein by 47.5 % in the wild-type and 119.6 % in the transgenic mice compared with the MG132 untreated respective mice (Fig. 4b). Moreover, MG132 resulted in an increase of ubiquitinated NKCC2 protein in the wild-type and transgenic mice with high salt intake, but more in the transgenic mice than in the wild-type mice (Fig. 4c). These results demonstrate that the ubiquitinated NKCC2 protein is chiefly degraded by proteasome. Taken together, NKCC2 protein degradation in the transgenic mice is enhanced by the synergy of 20-HETE with high salt intake through ubiquitin–proteasome pathway.

Fig. 4
figure 4

Ubiquitinated NKCC2 is degraded through ubiquitin–proteasome pathway. a Proteasome activity in the kidneys of wild-type (WT) and transgenic (TG) mice fed with either a normal salt (NS) or a high salt (HS) diet. n = 4. b Renal NKCC2 protein expression in WT and TG mice treated with MG132 in the last three days of two weeks of HS diet. n = 3. c Ubiquitinated NKCC2 protein is detected by immunoprecipitation (IP) and subsequent Western blot (WB) in WT and TG mice with 2 weeks of HS diet and co-administration of MG132. n = 3. Representative Western blot is shown. All experiments are performed three times independently. *p < 0.05 versus untreated value of the respective group of mice; # p < 0.05 versus values of corresponding WT mice

Full size image

Nedd4-2-mediated ubiquitination of NKCC2 in the CYP4F2 transgenic mice with high salt intake

As Nedd4-2 is an important ubiquitin E3 ligase and involves in ubiquitination of numerous ion channels, we wonder whether Nedd4-2 also mediates the ubiquitination of NKCC2. Hereby, renal Nedd4-2 expression was assessed, which was significantly higher in the transgenic mice than in the wild-type mice with high salt intake (Fig. 5a). Furthermore, an interaction between NKCC2 and Nedd4-2 was detected, and the transgenic mice showed an increase of their abundance in IP (Fig. 5b). These observations suggest that Nedd4-2 might be involved in the ubiquitination of NKCC2, and 20-HETE activated this process partly by increasing Nedd4-2.

Fig. 5
figure 5

Nedd4-2 mediated the ubiquitination of NKCC2 in HS intake. a Nedd4-2 protein expression in the kidneys of wild-type (WT) and transgenic (TG) mice fed with high salt (HS) diet for 2 weeks. b The interaction of NKCC2 and Nedd4-2 in WT and TG mice with 2 weeks HS diet is detected by co-immunoprecipitation: immunoprecipitation (IP) with NKCC2 or Nedd4-2 antibody, followed by Western blot (WB) with Nedd4-2 or NKCC2 antibody. Representative Western blot is shown. All experiments are performed three times independently from three mice in each group

Full size image

Effect of high salt intake on sodium excretion and blood pressure in the CYP4F2 transgenic mice

Finally, we compared urinary 20-HETE, urinary sodium excretion and urine volume for 24 h, as well as blood pressure of the wild-type and transgenic mice with and without high salt intake to evaluate the synergy of 20-HETE and high salt intake in modulation of blood pressure. Urinary 20-HETE excretion was detected in the wild-type and transgenic mice by LC/MS techniques. As shown in Fig. 6a, the transgenic mice had a consistently higher urinary 20-HETE than the wild-type mice, although 20-HETE excretion was increased in both the wild-type and transgenic mice with high salt intake. Additionally, we measured urine sodium excretion (Fig. 6b) and urine volume (Fig. 6c) of the wild-type and transgenic mice, with and without high salt intake. The baseline sodium excretion and urine flow for 24 h in the transgenic mice were not significantly different from that in the wild-type mice. High salt intake resulted in the increase of sodium excretion and urine flow in both the wild-type and transgenic mice, and the transgenic mice had higher values than the wild-type mice.

Fig. 6
figure 6

The synergy of 20-HETE and high salt intake in modulation of blood pressure. Comparison of high salt (HS) intake in wild-type (WT) (n = 12) and CYP4F2 transgenic (TG) (n = 11) mice with normal salt (NS) intake in WT (n = 5) and TG (n = 7) mice in a. urinary 20-HETE excretion, b urine sodium excretion (UNa), c urine volume and d systolic blood pressure (SBP). *p < 0.05 versus values of corresponding NS intake in mice; # p < 0.05 versus values of corresponding HS intake in WT mice;**p < 0.05 versus values of WT mice before HS diet

Full size image

SBP in the wild-type mice was significantly lower than in the transgenic mice under condition of normal salt intake as previously reported, and high salt intake raised the SBP of the wild-type mice (118.03 ± 2.18 vs. 127.55 ± 5.82 mmHg, n = 12), but did not affect significantly the SBP of the transgenic mice (128.96 ± 1.97 vs. 131.83 ± 4.17 mmHg, n = 11) (Fig. 6d). These results indicate that overproduced 20-HETE inhibits Na+ reabsorption and plays a modulatory role in the renal adaptation to high salt intake.

Discussion

We previously demonstrated that overexpressed CYP4F2 and overproduced 20-HETE in the kidneys contribute to the increase of blood pressure in the CYP4F2 transgenic mice with normal salt intake (Liu et al. 2009, 2012). We hypothesized that natriuresis of 20-HETE is exerted in renal adaptation to elevated Na+ intake. Therefore, the CYP4F2 transgenic mice with high 20-HETE level were fed with a diet containing either normal salt (0.4 % NaCl) or high salt (4 % NaCl). We found that NKCC2 protein is dramatically decreased in the transgenic mice relative to the wild-type mice in response to high salt intake. We further elucidated that the reduction of NKCC2 expression is due to enhanced protein degradation via the classic ubiquitin–proteasome pathway, wherein Nedd4-2 mediates the ubiquitination of NKCC2 induced by 20-HETE and high salt intake. To our knowledge, this is the first report that the synergy of 20-HETE with high salt intake causes ubiquitination of NKCC2 and results in a decrease of NKCC2 expression.

In kidneys, 20-HETE is one of the major metabolites of AA produced in the TALH where NKCC2 is located exclusively and provides the major pathway of Na+ transport across the apical membrane (Imig 2004). We compared the NKCC2 expression in the kidneys between the wild type and CYP4F2 transgenic mice by Western blot and found that the abundance of NKCC2 was lower in the transgenic mice than in the wild-type mice, especially with high salt intake. We suppose that, in our CYP4F2 transgenic mice, the downregulation of NKCC2 protein is induced by 20-HETE and this action is triggered by high salt intake. To confirm our conjecture, we administered HET0016 to block the formation of 20-HETE in the transgenic mice; the decrease of NKCC2 protein was reversed pronouncedly in the kidneys of the transgenic mice with high salt intake, but less in the transgenic mice with normal salt intake. These findings were echoed by immunofluorescence of kidney tissue showing that the expression of NKCC2 was less in the transgenic than in the wild-type mice, and HET0016 significantly reversed the reduction of NKCC2 in transgenic mice with high salt intake. For this reason, we come to the conclusion that 20-HETE and high salt intake inhibit synergistically the expression of NKCC2 protein. The unique results in our study could attribute to the high constitutive 20-HETE in the transgenic mice with high salt intake, enhancing effectively the inhibitory action on NKCC2. The present finding of 20-HETE inhibition on NKCC2 protein supports previous studies: Dahl salt-sensitive (DS) rats with a lower renal 20-HETE have elevated activity and expression of NKCC2 in comparison to other strains of rats (Hoagland et al. 2004). Exogenous administration of 20-HETE could normalize the activity of NKCC2 in the TALH of DS rats (Ito and Roman 1999). In addition, our observation on the relationship between 20-HETE and high salt is similar to that chronic blockage of the renal formation of 20-HETE by HET0016, which had no effect on blood pressure in rats fed a low salt diet, and caused a rise in the blood pressure in rats fed a high salt diet (Hoagland et al. 2003).

We observe that the reduced NKCC2 protein is consistent with the low levels of NKCC2 mRNA in the transgenic mice compared with the wild-type mice under conditions of normal salt intake, implying 20-HETE could transcriptionally downregulate NKCC2 expression, which remains to be elucidated in future. However, the fall of NKCC2 protein is inconsistent with moderately increased NKCC2 mRNA, remarkably in the transgenic mice than in the wild-type mice under conditions of high salt intake, indicating that both 20-HETE and high salt intake participate in the modulation of NKCC2 at the protein level. Importantly, we found that ubiquitinated NKCC2 is induced by high salt intake and is significantly higher in the transgenic mice than in the wild-type mice, indicating that both 20-HETE and high salt intake enhance ubiquitination of NKCC2. Furthermore, the evidence that renal proteasome activity is increased in the transgenic mice compared with the wild-type mice, and that the MG132-inhibited proteasome activity can lead to the restoration of NKCC2 protein in the wild-type and transgenic mice with high salt intake, supports that both 20-HETE and high salt intake potentiate the degradation of NKCC2 protein through ubiquitin–proteasome pathway. It has been reported that high salt intake upregulates the renal expression of Nedd4-2, a ubiquitin-protein ligase E3, in Dahl salt-resistant (DR) and DS rat, and Nedd4-2 is involved in the control of ENaC ubiquitination (Umemura et al. 2006). Similar results could be found in another report that low salt intake reduces Nedd4-2 protein expression (Loffing-Cueni et al. 2006). In the present study, we observe that NKCC2 appears to be ubiquitinated and interacts with Nedd4-2 by high salt intake, with a higher level of ubiquitination in the transgenic than in the wild-type mice. Taken together, our data provides a strong evidence that 20-HETE modulates NKCC2 expression via either an ubiquitination–proteasomal degradation under conditions of high salt intake or transcriptional downregulation. However, we have to notice that other renal proteins could be involved in ubiquitin–proteasome degradation besides NKCC2 under conditions of high salt intake and/or 20-HETE, which require further studies. The results presented here extend our understanding of the molecular mechanism by which the synergy of 20-HETE with high salt intake results in the downregulation of NKCC2.

In addition, in the wild-type mice, high salt intake caused the increase of urinary 20-HETE excretion and blood pressure. The findings are in agreement with previous reports that urinary 20-HETE excretion are 66 % higher during salt loading than during salt depletion in both salt-resistant hypertensive patients and salt-sensitive hypertensive patients (Laffer et al. 2003), and that urinary 20-HETE excretion is increased by 20 % in animal models with a high salt diet (Hoagland et al. 2003). The elevated native 20-HETE excretion is likely to be the result of renal regulation adapting to elevated Na+ intake by increasing urinary sodium excretion and may be the reason why NKCC2 protein tended to decrease in the wild-type mice with high salt intake. By contrast, in the transgenic mice with a higher level of 20-HETE, high salt intake did not lead to a marked increase of blood pressure. This is due to the downregulation of NKCC2 protein by high salt intake, which resulted in increase of urinary sodium excretion. Therefore, by comparing the transgenic mice with a high level of 20-HETE and the wild-type mice with a low level of 20-HETE, we found that diverse levels of 20-HETE decided the different type of salt-related hypertension (salt-sensitive or salt-resistant hypertension) under conditions of high salt intake. This is a typical case resulting from the interaction of environmental factors with genetic factors in the development of hypertension. In other words, high level of 20-HETE is related to salt-resistant hypertension, and low level of 20-HETE is related to salt-sensitive hypertension.

In summary, 20-HETE and high salt intake synergistically decrease the abundance of NKCC2 protein via Nedd4-2-mediated ubiquitin–proteasome pathway. These observations support our hypothesis that 20-HETE exerts natriuresis in renal adaptation to elevated Na+ intake. From the present results, we infer that the 20-HETE level of individuals may be one of the important factors responsible for the salt sensitivity of blood pressure. We also propose that individuals with a high level of 20-HETE probably have low NKCC2 activity and high sodium excretion and, thereby, are susceptible to salt-resistant hypertension relative to individuals with a low level of 20-HETE.