Sex-specific adaptations to high-salt diet preserve electrolyte homeostasis with distinct sodium transporter profiles

Abstract

Kidneys continuously filter an enormous amount of sodium and adapt kidney Na⁺ reabsorption to match Na⁺ intake to maintain circulatory volume and electrolyte homeostasis. Males (M) respond to high-salt (HS) diet by translocating proximal tubule Na⁺/H⁺ exchanger isoform 3 (NHE3) to the base of the microvilli, reducing activated forms of the distal NaCl cotransporter (NCC) and epithelial Na⁺ channel (ENaC). Males (M) and females (F) on normal-salt (NS) diet present sex-specific profiles of “transporters” (cotransporters, channels, pumps, and claudins) along the nephron, e.g., F exhibit 40% lower NHE3 and 200% higher NCC abundance than M. We tested the hypothesis that adaptations to HS diet along the nephron will, likewise, exhibit sexual dimorphisms. C57BL/6J mice were fed for 15 days with 4% NaCl diet (HS) versus 0.26% NaCl diet (NS). On HS, M and F exhibited normal plasma [Na⁺] and [K⁺], similar urine volume, Na⁺, K⁺, and osmolal excretion rates normalized to body weight. In F, like M, HS lowered abundance of distal NCC, phosphorylated NCC, and cleaved (activated) forms of ENaC. The adaptations associated with achieving electrolyte homeostasis exhibit sex-dependent and independent mechanisms. Sex differences in baseline “transporters” abundance persist during HS diet, yet the fold changes during HS diet (normalized to NS) are similar along the distal nephron and collecting duct. Sex-dependent differences observed along the proximal tubule during HS show that female kidneys adapt differently from patterns reported in males, yet achieve and maintain fluid and electrolyte homeostasis.

INTRODUCTION

Sodium handling is tightly regulated by the kidneys. High sodium intake increases thirst and circulatory volume, which provide signals to increase renal excretion of sodium and water to restore circulatory volume balance, independent of changes in blood pressure (1, 2). People in modernized cultures around the world chronically consume about twice the daily NaCl intake recommended by the American Heart Association, which can lead to salt-sensitive hypertension driven by elevated inflammation, sympathetic nervous system activation, and intrarenal production of angiotensin II, among other contributors (3–9). These effectors directly or indirectly activate tubular Na⁺ transport and, in salt-sensitive individuals, can increase blood pressure, which, via pressure natriuresis, offsets the tubular dysfunction by reducing Na⁺ reabsorption, matching NaCl excretion to intake (1, 10). Homeostasis is achieved in salt-sensitive individuals at the cost of chronic hypertension.

The natriuretic response of rodents to chronic high-salt (HS) diet involves Na⁺ transporter regulation all along the nephron. Mechanisms include membrane trafficking, covalent modifications, and changes in abundance. In male rats, we reported that 4% NaCl diet triggers a doubling in proximal tubule (PT) Na⁺/H⁺ exchanger isoform 3 (NHE3) phosphorylation, a redistribution of NHE3 to the base of the PT microvilli, 50% reduction in Na⁺-Cl⁻ cotransporter (NCC) pool size, and redistribution of Na⁺-K⁺-2Cl⁻ cotransporter isoform 2 (NKCC2), NCC, and epithelial Na⁺ channel (ENaC) subunits from the plasma membrane to intracellular membrane fractions (11, 12). Subsequently, Vallon et al. (13) demonstrated, in female mice, that HS diet reduced phosphorylation of NCC (NCCp), an indicator of cell surface expression and NCC activation (14, 15); and Frindt and Palmer (16), using in situ biotinylation in male rats, measured reduced cell surface abundance of the cleaved form of ENaC γ-subunit, indicating reduced activation. In male mice, Udwan et al. (17) reported that raising dietary sodium from 0.18% to 1.25% did not alter glomerular filtration, but did reduce PT Na⁺ reabsorption along with cell surface transporters (measured by apical biotinylation). Interestingly, Na⁺ reabsorption was increased along the distal nephron despite the reduced abundance of NCC and NCC phosphorylated, reflecting the predominant impact of greater Na⁺ delivery to the distal nephron during HS diet.

The previous characterization of the mechanisms driving natriuresis during HS diet were almost all conducted in male rodents. Yet, our examination of the abundance, covalent modifications, and regulators of electrolyte transporters, channels, and claudins (cldns) (collectively referred to as Na⁺ “transporters”) revealed striking differences between the females and males, in both rats and mice (18). Pertinent to this study, female mice exhibited smaller pool sizes of PT transporters than males, including NHE3, Na⁺-phosphate cotransporter 2A (NaPi2), claudin-2 (cldn-2), and NHE3 were localized at the base of the microvilli at baseline. Along the distal nephron, females presented larger pool sizes of cortical cotransporters (NCC, NKCC2), their phosphorylation [NCCp, Na⁺-K⁺-2Cl⁻ cotransporter isoform 2-P at S87 (NKCC2p)], and the cotransporter regulatory kinase Ste/SPS-1-related proline-alanine rich kinase (SPAK), cldn-7, and AQP2. Sexually dimorphic patterns have been verified or reported by other laboratories (19–22). Since NHE3 abundance is 40% lower and NCC abundance is 200% higher in F than M mice on normal-salt (NS) diet, we tested the hypothesis that the adaptations to high-salt diet along the nephron also exhibit sexual dimorphisms. In female and male C57BL/6J mice analyzed in parallel on normal-salt (NS) and high-salt (HS) diets, we evaluated abundance and covalent modification of renal transporters, channels, claudins, and regulatory proteins, as well as physiological adaptations.

METHODS

Animal Protocol

All animal procedures were approved by the Institutional Animal Care and Use Committee of the Keck School of Medicine of the University of Southern California (Protocol No. 20181) and were conducted in accordance with the National Institutes of Health “Guide for the Care and Use of Laboratory Animals.” Experiments were performed on male (M) and female (F) C57BL/6J mice at 6–8 wk of age, purchased from Jackson Laboratories (Bar Harbor, ME). Mice were fed a gelled diet (60% water) made from powdered chow (Envigo, AIN-93G, TD.94045) adjusted to either 0.26% of NaCl (normal-salt diet, NS) or 4.0% of NaCl (high-salt diet, HS) for 15 days. Four groups (n = 7) were studied in two sets and data combined: MNS, MHS, FNS, and FHS. After overnight acclimation in single mouse metabolic cages (Techniplast, Italy), at 4:00 PM on day 14, mice were weighed and placed into the metabolic cages for overnight urine collection with free access to water and diet till 8:00 AM (16 h). Overnight food intake was not measured, a recognized limitation of the study. Mice were returned to their standard cages with water but no food for 3 h before termination and tissue and blood collection on day 15. Mice were anesthetized with isoflurane, kidneys were removed and quick-frozen and stored at −80°C, and terminal blood was collected by cardiac puncture. Vaginal smears were analyzed to identify position in estrous cycle of the female mice and no impact was noticed, as previously reported (18). Note: one FNS and one MHS were excluded from the metabolic cage analyses because of insufficient urine collection, and another FNS was excluded because excessive food dropped into the urine collector.

Physiological Measurements

Urine volumes (UVs) were measured with pipettes, and urine and plasma [Na⁺] [K⁺], and [Li⁺] were measured by flame photometry (Cole-Parmer, model No. 02655-10, IL). Urine osmolality was assessed with a µ-Osmette osmometer (Precision Systems, MA). Lithium clearance (C_Li), a marker of volume flow from the PT and medullary thick ascending limb of Henle’s loop (mTAL; 18, 23), was calculated for a subset of each group according to a standard formula: C_Li = urinary [Li⁺] × UV/plasma [Li⁺].

Sample Preparation and Semiquantitative Immunoblotting

As previously described (24), kidneys were thawed in ice-cold PBS, cortex and medulla were dissected on ice and homogenized in 1 mL of isolation buffer [5% sorbitol, 0.5 mM disodium EDTA, and 5 mM histidine-imidazole buffer, pH 7.5, with the addition of 0.2 mM phenylmethylsulfonyl fluoride, 9 µg/mL aprotinin, and 5 µL/mL of a phosphatase inhibitor cocktail (Sigma, P0044)] on ice with an Ultra-Turrax T25 at the lowest speed setting for 5 min. Homogenates were centrifuged at 2,000 g for 10 min and supernatants were retained. Cortex (not medulla) pellets were again homogenized in another 1 mL of isolation buffer, centrifuged, and pooled with supernatant from the first spin. Single-use aliquots were prepared, quick-frozen in liquid N₂, and stored at −80°C.

Samples were defrosted, protein concentration assessed by bicinchoninic acid (BCA) assay (Pierce), and denatured in Laemmli sample buffer for 20 min at 60°C. Optimal protein loading was established empirically, and uniform protein loading was established by quantifying Coomassie-stained gels, as described (25). The semiquantitative nature of the detection system was determined by assessing 1X and 1/2X amounts of each sample on the same blot, with the expectation that the density would be doubled with double the sample amount (full blots in online supplement). Table 1 summarizes the antibodies used, amounts assayed, vendors, dilutions, and references to the production and validation. Blots were never stripped for reprobing. Signals were detected with the Odyssey infrared imaging system (LI-COR, Lincoln, NE) and quantified by accompanying software. Density values were displayed as arbitrary units and/or normalized to the average intensity of MNS, defined as 1.0.

Table 1.

Antibody and immunoblot protocol details

AntibodyTarget	∼kDa	Protein/ Lane Cortex, µg	Protein/ Lane Medulla, µg	Primary Antibody Supplier	AbHost	Dilution	Time	Secondary Antibody Supplier	Host and Target	Dilution(all 1 h)	Ref.
ACE1	190	5, 2.5	NA	Santa Cruz (SC-12187)	Gt	1:5,000	O/N	Invitrogen	DAG 680	1:5,000	(26)
Claudin-2	∼22	5, 2.5	NA	Thermo Fisher (32-5600)	Mu	1:2,000	O/N	Invitrogen	GAM 680	1:5,000	(27)
Claudin-7	∼22	5, 2.5	NA	Thermo Fisher (34-9100)	Rb	1:2,000	O/N	LI-COR	GAR 800	1:5,000	(28)
AQP1	35, 23	2, 1	NA	AlphaDiag. Int’l(AQP11-A)	Rb	1:5,000	O/N	LI-COR	GAR 800	1:5,000	(29)
AQP2	37, 23	2, 1	NA	Santa Cruz (sc-9882)	Gt	1:1,000	O/N	Invitrogen	DAG 680	1:5,000	(30)
AQP2pS256	37, 23	20, 10	NA	Invitrogen (PA5-38407)	Rb	1:1,000	2 h	Invitrogen	GAR 680	1:5,000	(31)
Collectrin	∼40	5, 2.5	NA	Thu Le (U.Rochester)	Rb	1:1,000	O/N	Invitrogen	GAR 680	1:5,000	(32)
ENaC α	100, 30	40, 20	NA	Loffing (Zurich)	Rb	1:5,000	O/N	Invitrogen	GAR 680	1:5,000	(33)
ENaC β	100	40, 20	NA	Loffing (Zurich)	Rb	1:15,000	O/N	Invitrogen	GAR 680	1:5,000	(33)
ENaC γ	80, 60	40, 20	NA	Loffing (Zurich)	Rb	1:15,000	O/N	Invitrogen	GAR 680	1:5,000	(33)
NaPi2	∼85	30, 15	NA	McDonough	Rb	1:1,000	O/N	Invitrogen	GAR 680	1:5,000	(34)
NCC	150	40, 20	NA	McDonough	Rb	1:5,000	O/N	Invitrogen	GAR 680	1:5,000	(35)
NCCpS71	150	40, 20	NA	Loffing (Zurich)	Rb	1:5,000	2 h	Invitrogen	GAR 680	1:5,000	(36)
NCCpT53	150	40, 20	NA	Loffing (Zurich)	Rb	1:5,000	2 h	Invitrogen	GAR 680	1:5,000	(36)
Na+-K+-ATPase α1	100	1, 0.5	1, 0.5	Kashgarian (Yale)	Mu	1:200	2 h	Invitrogen	GAM 680	1:5,000	(37)
Na+-K+-ATPase β1	50	10, 5	10, 5	McDonough	Rb	1:500	O/N	Invitrogen	GAR 680	1:5,000	(25)
NHE3	∼85	40, 20	10, 5	McDonough	Rb	1:2,000	O/N	Invitrogen	GAR 680	1:5,000	(38)
NHE3pS552	∼85	5, 2.5	10, 5	Santa Cruz (sc-53962)	Mu	1:1,000	2 h	Invitrogen	GAM 680	1:5,000	(39)
NHERF1	50	20, 10	NA	Weinman (U. Maryland)	Rb	1:2,000	O/N	Invitrogen	GAR 680	1:5,000	(40)
NKCC2	160	20, 10	10, 5	DSHB (Iowa)	Mu	1:6,000	2 h	LI-COR	GAM 800	1:5,000	(41)
NKCC2pS87	160	40, 20	15, 7.5	DSTT (Dundee)	Sh	1:2,500	2 h	Invitrogen	DAS 680	1:5,000	(42)
SPAK	60–70	20, 10	10, 5	Delpire (Vanderbilt)	Rb	1:3,000	2 h	LI-COR	GAR 800	1:5,000	(43)
SPAKpS373	70	40, 20	10, 5	DSTT (Dundee)	Sh	1:2,500	2 h	Invitrogen	DAS 680	1:5,000	(44)

Immunofluorescence
AntibodyTarget	Primary Antibody Supplier	AbHost	Dilution	Time	Secondary Antibody Supplier	Host and Target	Dilution	Time
NHE3	McDonough	Rb	1:100	2h	Invitrogen	GAR 488	1:500	1h
NHE3pAlexa Fluor 594	Santa Cruz (sc-53962 AF594)	Mu	1:100	2h	NA	NA	NA	NA

Approximate kDa refers to apparent molecular weight determined by BIO-RAD Precision Plus Protein Dual Color Standards. Ab, antibody; AQP1, aquaporin 1; AQP2, aquaporin 2; AQP2pS256, aquaporin 2-P at S256; DAS, donkey anti-sheep; ENaC, epithelial Na+ channel; GAM, goat anti-mouse; GAR, goat anti-rabbit; Mu, mouse; NA, not assayed; NaPi2, Na⁺-phosphate cotransporter 2A; NCC, Na⁺-Cl⁺ cotransporter; NCCpS71, Na⁺-Cl⁺ cotransporter-P at S71; NCCpT53, Na⁺-Cl⁺ cotransporter-P at T53; NHE3, Na⁺/H⁺ exchanger isoform 3; NHE3pS552, Na⁺/H⁺ exchanger isoform 3-P at S552; NHERF1, Na⁺/H⁺ exchanger regulatory factor 1; NKCC2, Na⁺-K⁺-2Cl⁻ cotransporter isoform 2; NKCC2pS87, Na⁺-K⁺-2Cl⁻ cotransporter isoform 2-P at S87; O/N, overnight; Rb, rabbit; Sh, sheep; SPAK, Ste/SPS-1 related proline-alanine rich kinase; SPAKpS373, Ste/SPS-1 relatd proline-alanine rich kinase-P at S373; Ref.: provides relevant citations to antibody production or validation.

Confocal Microscopy

Two mouse kidneys from each of the four groups (MNS, MHS, FNS, and FHS) were fixed in situ for 3 min with ice-cold PLP fixative (2% paraformaldehyde, 75 mM lysine, and 10 mM Na-periodate, pH 7.4), then removed and postfixed in periodate-lysine-paraformaldehyde fixative for 2–4 h, rinsed, cryoprotected overnight in 30% sucrose-PBS, then embedded in Tissue-Tek OCT compound, and frozen at −80°C, all as described (18). Subcellular distribution of proximal tubule NHE3 (green label) and Na⁺/H⁺ exchanger isoform 3-P at S552 (NHE3pS552) (red label) was assessed by immunofluorescence using a ZEISS LSM 510 confocal system with differential interference contrast overlay, as described (18). Tissue slices from all four groups were processed side by side on the same slides (2–3 slides/set) and visualized at the same magnification and settings. Table 1 summarizes antibody and protocol specifics. Specificity of the anti-NHE3 polyclonal (McDonough Lab) was validated against a sample of NHE3-knockout kidney (provided by T. Rieg, Univ. South Florida). Specificity of the anti-NHE3pS552 was established by Kocinsky et al. (39) by ELISA, immunoblot, and microscopy. Twenty micrometer scale bar applies to all panels.

Statistical Analysis

Data are presented as individual values and reported as means ± SE. Statistical analyses were calculated using GraphPad Prism 8.4 (San Diego, CA) and differences below P = 0.05 are provided in figures. Significance of physiological data was analyzed with one-way ANOVA followed by Sidak’s multiple comparison tests of MNS versus MHS, MNS versus FNS, and FNS versus FHS. Lithium clearance of M and F mice in the NS diet groups was compared with parametric unpaired Student’s t test. Significance of transporter abundance was analyzed with two-way ANOVA with Tukey correction for multiple comparisons.

RESULTS

Effect of a High-Salt Diet on Physiological Parameters in F versus M Mice

To test for sex-specific responses to elevating dietary salt, C57BL/6J mice were fed diets with 0.26% of NaCl (NS, normal salt, 0.01% Na⁺) or 4% of NaCl (HS, high salt, 1.6% Na⁺) for 15 days. Physiological responses, collected overnight in metabolic cages at day 15 after acclimation, are summarized in Table 2 in absolute values and in Fig. 1 in values normalized to body weight (bw). Body weight was 30% lower in females than in males on NS diet, as expected from the vendor growth charts. Body weight did not differ between NS and HS diet groups within a sex. Both M and F mice maintained homeostasis of plasma [Na⁺] and [K⁺] during HS diet at ∼150 mM and 4.8 mM, respectively. Overnight urine volume (UV/g bw) tended to be higher by 2.5-fold in M than F on NS diet (P = 0.4); UV, uncorrected for body weight (Table 2), was fourfold higher in M than in F fed NS diet. Other studies we have conducted indicated twofold higher UV in M than F (18, 28). Nonetheless, UV was measured twice, and the values were reproducible. HS diet doubled mean value of UV in M compared with NS (P = 0.2), and UV was fivefold higher in F on HS than in F on NS (P = 0.008). On NS diet, urine Na+ excretion (UNaV)/g bw was not different between M and F and, as expected, UNaV was greater in both sexes on HS diet than on NS diet (P < 0.0001). C_Li, a measure of volume flow from the PT and medullary TAL, was higher by threefold in M (P = 0.001) and sevenfold in F (P = 0.0001) during HS feeding (Fig. 1), consistent with greater changes in UV and UNaV in F than in M (Table 2 and Fig. 1), indicative of increased glomerular filtration rate (GFR) during HS diet and more Na⁺ and volume delivery to the distal nephron during HS diet in both sexes. K⁺ excretion [urine K⁺ excretion (UKV)/g bw] was higher in M than in F (P = 0.03) on NS diet. Interestingly, UKV/g bw was higher during HS diet by threefold in F (P < 0.0001) and 0.26-fold in M (P = 0.08). Urine osmolality [urine osmole excretion (UosmV)/g bw] was similar between sexes under NS diet and higher by 2.5-fold and fivefold, respectively, in M and F after HS diet (P < 0.0001).

Table 2.

Physiological parameters in male and female mice fed normal-salt and high-salt diet for 15 days

Physiological Parameters	Male NS	Male HS	Female NS	Female HS
Body weight, g	27.0 ± 0.6	27.7 ± 0.8	19.3 ± 0.4	20.1 ± 0.5
Two kidney weight, g	0.333 ± 0.009	0.360 ± 0.22	0.249 ± 0.006	0.261 ± 0.01
Plasma Na+, mM	150.6 ± 1.4	148.6 ± 1.2	153.1 ± 2.1	151.1 ± 2.1
Plasma K+, mM	4.7 ± 0.2	4.9 ± 0.2	4.8 ± 0.2	4.8 ± 0.2
UV, mL/16 h	1.8 ± 0.2	3.4 ± 0.6	0.46 ± 0.1	2.4 ± 0.5**
UNaV, mmol/16 h	74 ± 8	1,073 ± 278***	27 ± 4	920 ± 96***
U[Na+] mmol/L	43 ± 7	390 ± 34***	81.1 ± 23	432 ± 5***
UKV, mmol/16 h	114 ± 7	144 ± 33	35 ± 5	124 ± 13***
U[K+], mmol/L	64.8 ± 4	54.2 ± 5	108.9 ± 38	58.5 ± 4.9
UosmV, mosmol/kgH2O/16 h	1.3 ± 0.2	3.25 ± 0.9	0.44 ± 0.1	2.8 ± 0.2
U[osm], mosmol/kgH2O/L	654 ± 54	948 ± 105***	789 ± 62	1,205 ± 165***

Urine was collected for 16 h overnight, n = 5 or 6/group. HS vs. NS within sex: statistical comparisons were performed with GraphPad Prism one-way ANOVA. **P ≤ 0.01, ***P ≤ 0.001. C_Li groups were compared with parametric unpaired Student’s t test. See Fig. 1 for values normalized to for body weight and specific P values. HS, high salt; NS, normal salt; UKV, urinary K⁺ excretion; UNaV, urinary Na⁺ excretion; UosmV, urinary osmole excretion; UV, urine volume.

Figure 1.

Physiological parameters in male vs. female C57BL/6J mice fed normal-salt (NS) and high-salt (HS) diet for 15 days. Displayed are results from Table 2 as individual values normalized to body weight (g) ± SE. Urine was collected for 16 h overnight, n = 5 or 6/group. P values are indicated for comparisons between M vs. F fed NS and between NS and HS within sex by one-way ANOVA with Sidak correction. F, female; M, male; UKV, urinary K⁺ excretion; UNaV, urinary Na⁺ excretion; Uosm, urinary osmolality; UosmV, urinary osmole excretion; UV, urine volume.

Effect of High-Salt Diet on Na⁺ Transporters along the Proximal Tubule and Medullary Thick Ascending Limb in F versus M Mice

The impact of a 2 wk 4% NaCl diet on sodium and water [aquaporins (AQPs)] transporters, channels, and claudins in F versus M was evaluated by semiquantitative immunoblot in homogenates from kidney cortex and medulla. Figure 2 displays results for proteins along the PT and medullary TAL. Figure 2A displays immunoblots from samples of M and F on NS and HS diets, resolved by SDS-PAGE on the same gel, blotted, processed with antibodies, and quantified on the same blot, with results normalized to the mean of the M NS diet densities (defined as 1), which provides impact of sex and HS diet. Assessment of equivalent loading and linearity are described in methods. Figure 2B summarizes the same data displayed as arbitrary density units with P values determined by two-way ANOVA.

Figure 2.

Proximal tubule and medullary thick ascending limb transporter abundance in male and female during high-salt (HS) vs. normal-salt (NS) diet. Abundance of renal transporters were assessed in samples from M and F fed normal-salt (NS) and high-salt (HS) diets, assayed by semiquantitative immunoblot in homogenates from renal cortex and medulla. To account for differences in baseline abundance in F vs. M while also examining impact of dietary salt, data are displayed in two ways. A: samples from all four groups were loaded on the same gels at 1X and ½X loading amounts of protein (only one amount shown), blotted onto same PVDF membrane, and processed with antibodies, as detailed in Table 1. Density values were normalized to mean of MNS = 1.0, displayed under blots as means ± SE (n = 5). MW indicates molecular weight in kDa. Note: MW standard locations are not picked up when imaging with some secondary antibodies, but locations are marked on the blots and indicated in the figure. B: data displayed as individual arbitrary density units (not normalized to MNS), including P values calculated after two-way ANOVA with Tukey correction for multiple comparisons.

On NS diet, NHE3 exhibited 40% and 20% lower abundance in cortex and medulla, respectively, in F versus M, confirming our previous report (18). The PT scaffolding protein NHERF1 was 20% lower in F than in M; angiotensin-converting enzyme 1 (ACE1) was twice as abundant in F than in M; the paracellular junctional protein cldn 2, previously reported less abundant in F than in M mice (18), was variable in F on NS diet but when both diet groups were combined, 50% lower in F than in M. AQP1 35 and 23 kD forms were 1.6- to 1.8-fold more abundant in F than in M, as previously reported (18). Cortical Na,K-ATPase α1 subunit (reflecting abundance along both proximal and distal nephron) was 70% higher in F than in M on NS diet, but no sex difference was evident for medullary Na,K-ATPase. Medullary SPAK, a kinase that activates NKCC2 and NCC through phosphorylation, was 40% more abundant in F than in M, yet Ste/SPS-1-related proline-alanine rich kinase-P at S373 (SPAKp) was not different, both as previously reported in mice on NS (18); in addition, dietary salt did not affect medullary SPAK or SPAKp in M or F.

On HS diet, M, but not F, mice tended to have higher NHE3p (P = 0.1). Since NHE3p in male rats is associated with translocation of NHE3 to the base of the PT microvilli (11, 45), where it is less active (45–47), we examined NHE3 distribution. Renal cortical samples from MNS, MHS, FNS, and FHS were dual labeled green for NHE3 and red for NHE3p and processed with identical settings. Five tubules from each group, sampled from two to three tissues slices from two mice per group, are displayed in Fig. 3. A sex effect is evident: NHE3 and NHE3p are more intense in M than in F, as evident in the immunoblot results (Fig. 2). In M with NS diet, the signals colocalize as a broad band in the body of the microvilli in males and as a very narrow band at the base of the microvilli in F. With HS diet, there is a heterogeneous redistribution in M toward the base of the microvilli that results in a distribution distinct from that in NS group. The variability precluded the determination of whether there was any significant redistribution to the base, as we previously reported in male rats on HS diet (11). In F on HS diet, NHE3 remained at the base of the microvilli. Abundance of the transporter chaperone protein collectrin was greater by 30% in M on HS (P = 0.01), raising it to the abundance measured in F with both NS and HS diets. Similarly, cortical Na,K-ATPase was greater by 50% in M on HS diet, raising it to the abundance in F with both NS and HS diets. NHE3 and AQP1 were not altered by HS diet, and both maintained the sex differences measured with NS. Along the mTAL, total medullary Na⁺-K⁺-2Cl⁻ cotransporter isoform 2 (mNKCC2) was similar in M and F on NS and HS diet, whereas NKCC2pS87 was 50% lower in F than in M on either diet (P = 0.016), which suggests less NKCC2 transport activity in F versus M.

Figure 3.

NHE3 distribution in proximal tubule microvilli of male (M) and female (F) mice fed normal-salt (NS) or high-salt (HS) diet. Tissue slices from kidneys from all four groups were processed side by side on the same slides (n = 2 kidneys from each group and 2–3 slides/set). Table 1 summarizes antibody and protocol specifics. All slides were visualized by confocal immunofluorescence at the same magnification and settings. Twenty micrometer scale bar applies to all panels. Images show tubules from tissue slices dual-labeled to detect total NHE3 (green label) and NHE3pS552 (red label). Specificity of the anti-NHE3 polyclonal (McDonough lab) was validated against a sample of NHE3-knockout kidney (provided by T. Rieg, Univ. South Florida). Specificity of the anti-NHE3pS552 was established by Kocinsky et al. (39) by ELISA, immunoblot, and microscopy. NHE3, Na⁺/H⁺ exchanger isoform 3; NHE3pS552, Na⁺/H⁺ exchanger isoform 3-P at S552.

Effect of High-Salt Diet on Na⁺ Transporters along the Cortical Thick Ascending Limb, Distal Tubule, and Cortical Collecting Duct in F versus M Mice

We previously reported that along the distal nephron, F mice at baseline consuming NS diet exhibit greater abundance of cotransporters, their phosphorylated forms, and the cotransporter kinase SPAK compared with M mice (18). These results are confirmed in this study (Fig. 4). The following are greater in F versus M with NS diet by the indicated percentages: NKCC2 and NCC by 20%; NCCpS71, NCCpT53, claudin-7 [paracellular Cl⁻ permeability regulator (48)], and SPAK by 50%–70%; and SPAKp by 30%. In contrast, NKCC2pS87 is lower by 40% in F in cortex, as in medulla (Fig. 2).

Figure. 4.

Distal nephron Na⁺ transporter abundance during high-salt (HS) vs. normal-salt (NS) diets in males and females. Abundance of renal transporters, in samples from M and F fed normal-salt (NS) and high-salt (HS) diets, assayed by semiquantitative immunoblot in homogenates from renal cortex and medulla. To account for differences in baseline abundance in F vs. M while also examining impact of dietary salt, data are displayed in two ways. A: samples from all four groups were loaded on the same gels at 1X and ½X loading amounts of protein (only one amount shown), blotted onto same PVDF membrane and processed with antibodies, as detailed in Table 1. Density values were normalized to mean of MNS = 1.0, displayed under blots as means ± SE (n = 5). MW indicates molecular weight in kDa. Note: MW standard locations are not picked up when imaging with some secondary antibodies, but locations are marked on the blots and indicated in the figure. B: data displayed as individual arbitrary density units (not normalized to MNS) including P values calculated after two-way ANOVA with Tukey correction for multiple comparisons. n = 5/group.

In response to 15 days HS diet, the abundance of distal NCC is lower in both sexes. Specifically, the following are lower by the indicated percentages in both sexes: NCC by 20%–30%, NCCpS71 by 40%, and NCCpT53 by 50%. NKCC2, SPAK, and cldn-7 are not suppressed by HS diet (Fig. 4). Full-length α-ENaC, cleaved (activated) α-ENaC, and cleaved γ-ENaC subunits are lower by 20%, 30%–50%, and 20%, respectively. Counter to expectations, abundance of both β-ENaC and full-length γ-ENaC increase during HS feeding in both sexes. AQP2, more abundant in females than males on NS diet (49, 50), responded to HS diet with increased pool sizes of both the 37-kD form (P < 0.002) and the 23-kD form (P < 0.04), as well as increased phosphorylation of the 23-kD form in both M (1.6-fold) and F (1.4-fold) (P ≤ 0.012). Figure 5 summarizes the relative changes in transporters along the nephron during HS diet, normalized to M or F abundance on NS diet, defined as 1.0. These normalized values of HSD/NSD in M and F were compared to identify sexual dimorphic responses to HS diet along the nephron. Increases with HS diet were significantly greater in M than in F (by unpaired t test) for: NHE3p (P = 0.003), collectrin (P = 0.02), cldn-2 (P = 0.006), Na,K-ATPase α1 (P = 0.0005), and β-ENaC (P = 0.006), and less for SPAKp (P = 0.001) and cleaved α-ENaC (P = 0.015).

Figure. 5.

Comparison of renal transporter profiles of male and female mice during high-salt diet. To compare the impact of high-salt (HS) vs. normal-salt (NS) diet on transporter abundance in M and F, independent of differences in their NS abundance, relative abundance of HSD values for F and M were normalized to mean relative abundance of NS values, defined as 1.0. Data are from Figs. 2 and 4. n = 5/group. *P < 0.05 (exact P values provided in Figs. 2 and 4). Responses to HS diet in F vs. M, determined by unpaired Student’s t test, were greater in M vs. F for: NHE3p (P = 0.003), collectrin (P = 0.02), cldn-2 (P = 0.006), and Na,K-ATPase α1 (P = 0.0005), whereas changes with HS diet were greater in F vs. M for: SPAKp (P = 0.001), cleaved α ENaC (P = 0.015), and β ENaC (P = 0.006), indicated by white stars at the base of the columns. HSD, high-salt diet.

DISCUSSION

Overall, transporter adaptations to HS diet were similar between sexes along the distal nephron and collecting duct when abundance was normalized to that measured in NS groups. However, on NS diet, there are many differences between F and M along the nephron: female mice exhibit lower NHE3, NHERF1, cldn-2, AQP1, and NKCC2p along the PT and thick ascending limb of Henle’s loop (TAL), and higher levels of NKCC2, NCC, NCCp, cldn-7, SPAK, SPAKp, AQP2, and AQP2p along the distal nephron compared with levels in M, indicating the regulatory path to adaptation is sex-specific. Table 2 and Fig. 1 show that homeostasis of body weight (bw) and plasma electrolytes was maintained in M and F on 4% of NaCl (∼1.6% Na) diet. This level of dietary salt can raise glomerular filtration rate, an appropriate adaptation to excrete elevated Na⁺ intake. We analyzed adaptations only at steady state and recognized that females may respond differently than males in the acute phase or to the lower dietary Na⁺ content. Since both the NS and HS diets were prepared from the same powdered chow with addition of NaCl, they have the same potassium content, thus, we propose that urinary K⁺ output can be a surrogate measure of dietary intake (assuming steady state by 15 days). Males on NS diet excrete twice the UKV/g bw as females, suggesting they eat twice as much NS diet/g bw in metabolic cages, whereas M and F on HS diet excrete about the same UKV/g bw, suggesting they eat about the same amount of HS diet normalized to their bw. This pattern of dietary intake gives rise to larger fold changes in UNaV, UKV, and UosmV in F than in M on HS versus NS diets (Table 2). However, when normalized to bw, the absolute values of these parameters during HS are indistinguishable (Fig. 1). Accompanying these differences were increases in lithium clearance, indicative of increased volume flow from the PT and medullary TAL, and values of C_Li in mL/16 h/bw were indistinguishable between F and M on HS diet. These results indicate that the physiological responses to HS are ultimately similar in F and M. The changes in C_Li are large (fivefold to sevenfold), suggesting a role for increased GFR during HS diet. Although not measured in this study, GFR was not changed in the study of Udwan et al. (17), in which C57BL/6 male mice were fed 1.25% Na⁺ for 7 days (vs. the 1.6% Na⁺ over 15 days herein).

Previous studies have shown that M and F mice have discrete patterns of distribution and abundance of the renal transporters along the nephron (18, 19, 21, 22, 28, 51). In this study, we investigated how a high-salt diet impacted the sex-specific patterns with semiquantitative immunoblots of cortical and medullary homogenates, normalizing to a constant amount of protein assayed. Although the male body weight and kidney weight are ∼1.4-fold greater than female body weight and kidney weight (Table 2), we are likely sampling equivalent cellular content. However, since there are sexual dimorphisms in specific tubular lengths (20, 52, 53), differences detected in relative abundance normalized to protein may be attributed to either a difference per cell or difference in relative tubular length. On NS diet, we observed less NHE3 and NHERF1 in F than in M, as previously reported (18). Renal NHERF1 is reported to confer salt sensitivity in male mice (54), so less NHERF1 in females may be protective. PT cldn-2 is, overall, 50% lower in F than M, as previously reported (18), which can facilitate natriuresis by reducing paracellular sodium reabsorption (27). TAL NKCC2pS87 was also 50% lower in F than M on either diet (P = 0.016 on NS, P = 0.002 on HS), which suggests less NKCC2 transport activity along the TAL in F. We did not previously detect this sex difference with an antibody directed to NKCC2pT96T101 (18). This discrepancy is now explained by a recent report of a five amino acid deletion (DeltaF97-T101) in NKCC2 in the C57BL/6 strain that prevents detection of NKCC2 phosphorylation at this site; the site at NKCC2pS87 is not mutated (55). The nonspecific signals previously obtained with antibodies to NKCC2pT96T101 in the C57BL/6 strain can be attributed to cross-reaction with a homologous phosphorylation site on the Na⁺-Cl⁻ cotransporter (55). Lower NHE3, cldn-2, and NKCC2p in females would likely facilitate more rapid adaptation to a dietary Na⁺ challenge by expediting natriuresis and diuresis along the proximal nephron.

Higher C_Li on HS diet predicts less sodium reabsorption along the PT and medullary TAL; however, differences in transporter abundance between NS and HS diets were not evident in either sex. Collectrin was 30% greater in M on HS diet (P = 0.01), potentially significant because this transmembrane glycoprotein, highly expressed in the kidney proximal tubule and collecting duct, as well as in vascular endothelium (56), facilitates l-arginine uptake and nitric oxide (NO) production, which could expedite natriuresis (57). What accounts for the proximal nephron natriuresis during HS diet in this study? In a 2005 study in male rats on HS diet, we reported increased abundance of NHE3p, an indicator of NHE3 localized to the base of the microvilli [where it is less active (46)], and confocal microscopy evidence for this redistribution (11). In that same study, we reported, by membrane fractionation, redistribution of transporters to higher density membranes enriched in endosomal markers during HS diet (11). Two recent studies in male mice are informative: Udwan et al. (17) implementing surface biotinylation of transporters, measured a 50% fall in surface expression of NHE3 and other transporters along the nephron in male mice fed HS (1.25% Na⁺) and depressed protein pool sizes; Ralph et al. (8) measured downregulation of NHE3 mRNA in mice fed 3% Na⁺ diet (near double the Na⁺ content in this study). Taken together, the proximal natriuresis during HS diet likely results from smaller pools of active transporters within apical villar domains in both sexes. However, questions remain, since Fig. 3 shows NHE3 resides at the base of the microvilli in F on both NS and HS diets.

On the receiving end of the increased volume flow from the proximal nephron during HS diet is the cortical TAL and distal tubule, where cotransporters NKCC2 and NCC, cldn-7, the cotransporter kinase SPAK, and AQP2 are more abundant in F than in M mice (Fig. 3), as reported before (18). Cortical Na,K-ATPase α1 subunit (Fig. 2B) was also 0.7-fold more abundant in F than M, consistent with higher cotransporter activity along the distal nephron. In response to elevated dietary salt and volume flow, abundance of NCC, NCCpS71, and NCCpT53 is lower in males by 0.7-, 0.6-, and 0.5-fold, respectively, versus levels in NS. Similar decreases in NCC and NCCp are evident in females; however, overall abundance of NCC and NCCp per protein remains greater in F than in M on HS diet (P < 0.02 for NCC, NCCpS71, and NCCpT53; Fig. 4). The decreases in NCCp were not associated with the changes in plasma [K⁺] in either M or F, indicating [K⁺]-independent regulation, in agreement with results of related studies in males on HS diets (11, 12, 17). NKCC2p is unchanged by HS diet in cortex and medulla in M and F, remaining more abundant in M than F on HS diet (P = 0.002). SPAK and SPAKp were unchanged by HS feeding in either sex. Two studies are relevant to dissociation of SPAKp abundance from abundance of its target NCC/NCCp during HS feeding: Feng et al. (58) reported that when SPAK knockout mice were fed high-salt diet, total NCC abundance decreased almost 50%, similar to the response in this study (wild-type mice), indicating that the NCC abundance adaptation in HS diet is independent of SPAK. The Loffing Lab (59, 60) has elucidated a pathway regulating NCC dephosphorylation mediated through activation of protein phosphatase-1 via inhibition of the distal tubule enriched protein phosphatase 1 inhibitor-1 (I-1), suggesting the hypothesis that HS diet may inhibit I-1.

Further along the nephron, cleaved (activated) forms of α-ENaC and γ-ENaC subunits were lower during HS diet in both sexes, consistent with natriuresis. A recent study in diabetic (db/db) mice reports decreases in ENaC cleavage in F but not M during HS diet, correlating with resistance to salt-sensitive hypertension in F but not M mice in this strain (61). In the C57BL/6, we see similar changes in ENaC subunits in F versus M on HS versus NS diet (Fig. 4), including counterintuitive elevations in pools of full-length β- and γ-ENaC that prompt comparison with other related studies. Full ENaC activity requires assembly of three subunits in the apical membrane as well as proteolytic cleavage of the α- and γ-subunits. Using their novel in situ kidney biotinylation approach in HS versus NS diet-fed male rats, Frindt and Palmer (16) also detected significantly increased β-ENaC, marginally increased full-length γ-ENaC, and significantly suppressed cleaved (activated) γ-ENaC in surface (apical) membranes (α not detected). The Feraille group, also implementing whole kidney surface biotinylation but in male mice fed HS versus NS diets, reported reduced surface full-length α-, full-length γ- and cleaved γ-ENaC (17). Taken together, HS diet provokes substantial reductions in the cleaved subunits across studies and in both sexes in this study, consistent with physiological responses to HS diet. The accompanying accumulation of full-length β- and γ-subunits during HS diet may, hypothetically, reflect unassembled subunits.

Udwan et al. (17) modeled the HS diet-dependent changes in sodium transport along the nephron in M mice that predicted a strong impact of volume flow on Na⁺ reabsorption. HS diet decreased proximal nephron Na⁺ reabsorption and increased Na⁺ reabsorption in distal segments with reduced energy efficiency despite reduced pool sizes of Na⁺ transporters. We measured a 50% increase in cortical (not medullary) Na,K-ATPase α1 abundance in males on HS diet (Fig. 2B). This α1 emanates from both proximal and distal cortical elements. Along the nephron, Na,K-ATPase activity is highest in the loop of Henle and distal convolute tubule, where, due to water impermeability, tubular fluid Na⁺ must be pumped up a steep gradient back into the circulation. Higher cortical Na,K-ATPase activity during HS diet is consistent with lower proximal transporters and reabsorption, shifting a greater load of Na⁺ and volume from the proximal to the distal nephron, which reduces the overall energy efficiency of sodium reabsorption, that is, requiring higher distal Na,K-ATPase. This difference was not observed in our previous analysis of adaptation to HS diet in male rats (11). An explanation for this differential response in mice versus rats would require an examination of Na,K-ATPase subunits in individual tubule regions with a semiquantitative approach (62).

In rodent models, sexual dimorphisms in M versus F often favor higher blood pressure and pathophysiology that are attributed to sex-related responses in sympathetic nervous system, renin-angiotensin system (RAS), and immune system, which are all impacted by dietary salt intake and sex hormones (6, 63). In contrast, we found similar impact of angiotensin II infusion hypertension on F versus M C57BL/6 mice for both the hypertension and tubular transporter responses (28). Sex differences were detected in the FVB/N mouse strain subjected to 6 days 5% NaCl diet. Although both M and F exhibited similar natriuretic responses, as in this current study, female mice exhibited the natriuresis at lower blood pressures (measured with radiotelemetry), evident as a leftward shift in the pressure-natriuresis relationship (64). This dimorphism was attributed, at least in part, to estrogen-dependent regulation of RAS, which raises levels of the angiotensin type 2 receptor (AT₂R) more in F than in M (64). Lower NHE3, claudin-2, and mNKCC2p on both NS and HS diets in F versus M proximal nephron, (all P < 0.002 on HS) likely contribute to the more robust pressure natriuresis in female mice; it remains to be determined whether these differences are due to hormones or genes (65).

In summary, urine and blood electrolytes are similar in M and F on NS diet and are maintained during 15 days HS diet, indicating successful homeostatic adaptations (Table 2 and Fig. 1). The adaptations associated with maintaining the homeostasis (Figs. 2, 3, 4, and 5) exhibit sex-dependent and -independent components. Differences in abundance of transporters, claudins, and channels measured at baseline on NS diet largely confirm previous reports (18, 19, 22, 28), and most of these differences persist during HS diet. Despite these significant differences at baseline, the fold changes during HS diet (normalized to the NS diet pool sizes) are similar from the medullary TAL to the collecting duct, as summarized in Fig. 5. This figure also highlights sexual dimorphisms in the changes in HS diet relative to NS diet (highlighted in Fig. 5 with ⋆ at base of columns). Increases with HS diet were greater in M versus F for: NHE3p, collectrin, cldn-2, Na,K-ATPase α1, and β-ENaC, and less for SPAKp (P = 0.001) and cleaved α-ENaC (P = 0.015). The sex-dependent differences in regulation indicate that during HS diet, like NS diet, female kidneys work and adapt differently from patterns reported in males. The disparate patterns, however, achieve and maintain fluid and electrolyte homeostasis.

SUPPLEMENTAL DATA

Data containing full immunoblots at: https://doi.org/10.6084/m9.figshare.16735060.v1.

GRANTS

This study was supported by NIH National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) 2R01DK083785 and R56 DK123780 (to A.A.M.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

D.L.T-P., D.L.R., L.C.V., and A.A.M. conceived and designed research; D.L.T-P., D.L.R., L.C.V., and A.A.M. performed experiments; D.L.T-P., D.L.R., L.C.V., and A.A.M. analyzed data; D.L.T-P., L.C.V., and A.A.M. interpreted results of experiments; D.L.T-P. and A.A.M. prepared figures; D.L.T-P. and A.A.M. drafted manuscript; D.L.T-P., L.C.V., and A.A.M. edited and revised manuscript; D.L.T-P., D.L.R., L.C.V., and A.A.M. approved final version of manuscript.