INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

NORMAL PREGNANT WOMEN and rats undergo a progressive, massive plasma volume expansion (PVE) (15, 21, 24), which is required for an optimal pregnancy outcome (9). In late pregnancy in the rat, there is clear evidence that the PVE involves net renal sodium and fluid retention. Indeed, sodium restriction inhibits the volume expansion and compromises the pregnancy close to term (30). Metabolic cage studies in the rat have documented positive sodium balance in the second half of pregnancy (4, 10). The mechanisms and nephron sites of the gestational renal sodium retention are unknown. In fact, there are many opposing influences on sodium excretion in normal pregnancy (21). Why the balance is shifted toward sodium retention is not clear. One possibility is that a generalized loss of tubular natriuretic capacity occurs in normal pregnancy and that this plays a facilitatory role in the gestational PVE by permitting antinatriuretic influences to dominate. In support of this possibility, the pregnant rat becomes refractory to the tubular natriuretic (but not hemodynamic) actions of atrial natriuretic peptide (ANP) (11, 24). Also, the pressure-natriuretic response is attenuated in late pregnancy (25). Some studies suggest that the diuretic and natriuretic response to acute volume expansion is also blunted in the pregnant rat (20, 29), although this is controversial (18).

One mechanism that would potentially explain all of these blunted natriuretic responses is the general increase that occurs in sodium reabsorptive capacity. In an earlier study, Lindheimer and Katz (22) suggested that increased renal Na⁺-K⁺-ATPase activity occurs in late-pregnant rats. The present experiments were conducted to further investigate this possibility and, specifically, to measure the Na⁺-K⁺-ATPase activity in kidney cortex and medulla in virgin, mid-, and late-pregnant rats, as well as to determine the presence and relative abundance of the renal Na⁺-K⁺-ATPase -subunits by Western and slot blot analyses.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Studies were performed on 15 female Sprague-Dawley rats, obtained from Harlan Sprague Dawley. Rats designated to become pregnant were placed with a fertile male, and daily vaginal smears were taken to detect the presence of sperm. Three groups of rats were studied: virgins (n = 5), midpregnant rats at 8-11 days of gestation (n = 5), and late-pregnant rats at 18-21 days of gestation (n = 5). (Gestation lasts 22 days in the rat.) On the day of tissue harvesting, rats were anesthetized with pentobarbital sodium (7 mg/100 g body wt). Each renal artery was clamped, and the kidneys were removed, separated into cortex and medulla on dry ice, and then snap-frozen in liquid nitrogen. The rat was then guillotined, and the brain stem was removed and snap-frozen.

Tissue was prepared for measurement of the distribution and abundance of -subunits of the Na⁺-K⁺-ATPase as follows. Tissues were weighed and homogenized in protease inhibitor solution (1 ml/100 mg tissue) containing 0.25 M sucrose, 1.0 mM EDTA, 500 µM phenylmethylsulfonyl fluoride (PMSF), 1.0 mM 4-aminobenzamide, and 1 mg/ml bacitracin (pH 7.0). Tissues (100-300 mg) were then homogenized on ice for three 15-s intervals using a Brinkmann Kinematic Polytron (Speed 5). The homogenate was centrifuged for 15 min at 4°C and 3,333 g in a Jouan CR 3000 SF centrifuge. We have used this method previously (17). The supernatant was centrifuged for 15 min at 4°C and 14,460 g in a Sorvall centrifuge (SS-34 Rotor). This supernatant was then centrifuged for 1 h at 30,590 g in a Sorvall SS-34 centrifuge. The pellet from this final centrifugation was resuspended in 100-300 µl of protease inhibitor using a 1-ml tissue grinder (Potter-Elvehjem). Total protein was then measured, by the method of Bradford (8a), on the microsomes, which were either stored at 70°C or assayed the same day. For Western blot analysis, proteins were separated by SDS-PAGE on 10% polyacrylamide gels. Approximately 30 µg of total protein were loaded per sample, in duplicate, with one lane of Bio-Rad prestained kaleidoscope standards and run on a vertical electrophoresis system (GIBCO Mini V 8.10) at 180 V. Transfer of proteins from the gel to the nitrocellulose membrane (Amersham Hybond ECL) was done with a Bio-Rad Trans-Blot SD electrophoretic transfer cell at 15 V for 15 min. For slot blotting, 300 µl of several different total protein contents (from 0.05 to 10 µg) in Tris-buffered saline (TBS: 137 mM NaCl and 20 mM Tris base, pH 7.6) were loaded (in duplicate) into wells in a Bio-Rad Bio-Dot SF microfiltration apparatus and then vacuum-transferred on to a nitrocellulose membrane. Quantitative examination of the protein transferred to the membrane by Western or slot blot was performed as follows. The membrane was incubated in blocking solution made from a 1:1 mixture of skim milk or Sigma blocking solution (T-8793) and TBS + 0.5% Tween 20 (TBS-T) at room temperature for 60 min with gentle rocking, and then washed three times in TBS-T (1× 15-min and 2× 5-min washes) at room temperature. The membrane was then incubated for 90 min with a monoclonal antibody against the 1- or 2-isoform of the Na⁺-K⁺-ATPase (MCK1, 1:100, and MCB2, 1:30; provided by Dr. K. Sweadner) or the 3-isoform of the Na⁺-K⁺-ATPase (1:500; Affinity Bioreagents), then washed three times with TBS-T (1× 15-min and 2× 5-min washes). The membrane was then incubated for 60 min with the secondary antibody (horseradish peroxidase-conjugated goat anti-mouse IgG; Bio-Rad) diluted in blocking solution (1:1,500), then washed five times with TBS-T (1× 15-min and 4× 5-min washes). The membrane was then incubated for 1 min in Amersham ECL reagents (17) and exposed to Kodak X-Omat X-ray film for 30 s to 10 min. Quantitation for slot blot was done by automated densitometry using the Optimus Densitometry System (Edmonds, WA). The relationship between optical density (OD) and micrograms of protein from kidney cortex was linear (from 1 to 10 µg protein for the 1-subunit of the Na⁺-K⁺-ATPase), and quantitation was performed on 3-µg aliquots. A linear relationship between OD and total protein concentration was also seen for the 1-subunit in renal medulla (from 0.05 to 3 µg protein) and for the 2-subunit (from 0.05 to 3 µg protein), and quantitation was performed on 0.5-µg aliquots. The 3-subunit of the brain stem Na⁺-K⁺-ATPase was quantitated at a total protein concentration of 0.075 µg. Because of differences in background intensity, all values for mid- and late-pregnant rats are expressed as percent change from the virgin (100%) value. Negative controls had been done in our earlier study (17) to demonstrate the specificity of the interactions.

ATPase activity in microsomes was measured at 37°C using a coupled-enzyme assay, as described by Norby (28). Assays were performed in a buffer containing (in mM) 100 NaCl, 10 KCl, 1 MgCl₂, 0.1 EDTA, and 20 imidazole, pH 7.0, with the addition of 0.1 mg/ml microsomes, 0.42 mM phospho(enol)pyruvate, 0.15 mM NADH, 7.5 IU of pyruvate, and 18 U of lactate dehydrogenase. MgATP (5 mM) was added to start the assay, and the time-dependent decrease in NADH absorbance at 340 nm was monitored to determine the rate of ATP hydrolysis. The contribution of Na⁺-K⁺-ATPase activity to the total ATPase activity was determined both by testing the sensitivity of the samples to the Na⁺-K⁺-ATPase-specific inhibitor oligomycin (3) and by measuring the ATPase activity in a K⁺-free medium. The final concentration of oligomycin was 20 µM. Both methods provided equivalent results.

Data are shown as means ± SE, and statistics were determined by one-way ANOVA with a least-squares means test for individual differences and a t-test. P < 0.05 was considered to be statistically significant.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

As shown in Figs. 1 and 2, one band of 1-subunit Na⁺-K⁺-ATPase is seen in homogenates of cortex and medulla from kidneys of virgin, mid-, and late-pregnant rats. This is close to the expected molecular mass of 113 kDa (23), although there is also a fainter band at ~90 kDa in some lanes, which probably represents a truncated form due to NH₂-terminal processing. No 2- or 3-subunits are detectable by this method in kidney cortex (Figs. 3 and 4) or medulla (data not shown). In brain stem, the 2- and 3-subunits of Na⁺-K⁺-ATPase, with expected molecular mass of ~112 kDa (23), are the predominant isoforms detectable by Western blot. The double bands may either reflect the existence of -subunits combined with different -subunits that survived denaturation or be due to some phosphorylation of the isoforms (13). A truncated version of the 3-subunit of Na⁺-K⁺-ATPase is evident in aorta. Data from the slot blot quantitation are shown in Fig. 5, expressed as percent change from the virgin value. Declines of ~50% occur in the quantity of the 1-subunit of Na⁺-K⁺-ATPase in renal cortex of mid- and late-pregnant rats relative to that of virgin rats. In contrast, the abundance of the 1-subunit in medulla and of both 2- and

View larger version (71K): [in this window] [in a new window]   Fig. 1.   Western blot showing presence of 1-subunit of Na+-K+-ATPase in renal cortex from virgin, mid-, and late-pregnant rats. No Na+-K+-ATPase 1-subunit was evident in brain stem.

View larger version (68K): [in this window] [in a new window]   Fig. 2.   Western blot showing presence of 1-subunit of Na+-K+-ATPase in renal medulla from virgin, mid-, and late-pregnant rats. No Na+-K+-ATPase 1-subunit was evident in brain stem.

View larger version (66K): [in this window] [in a new window]   Fig. 3.   Western blot showing presence of 2 isoforms of 2-subunit of Na+-K+-ATPase in brain stem of virgin rat. No 2-subunit was evident in renal cortex from virgin, mid-, and late-pregnant rats.

View larger version (65K): [in this window] [in a new window]   Fig. 4.   Western blot showing presence of 2 isoforms of 3-subunit of Na+-K+-ATPase in brain stem, and of truncated isoform in aorta from virgin rat. These isoforms were not evident in renal cortex from virgin rats.

View larger version (25K): [in this window] [in a new window]   Fig. 5.   Left: summary of relative abundance of 1-subunit of Na+-K+-ATPase expressed as percentage of virgin value (% of V) in cortex and medulla, and of 2- and 3-subunits in brain stem in virgin (V), mid (MP)-, and late-pregnant (LP) rats. Right: comparison with Na+-K+-ATPase activity in cortex, medulla, and brain stem in virgin, mid-, and late-pregnant rats. For all groups, n = 5 rats/group.

  
3-subunits in brain stem are constant in virgin, mid-, and late-pregnant rats.

The Na⁺-K⁺-ATPase activity data are given in Table 1. The total ATPase activity is lower in the renal cortex in midpregnancy and late pregnancy vs. virgins. The oligomycin sensitivity of the ATPase activity and the residual ATPase activity measured in a K⁺-free medium indicates that Na⁺-K⁺-ATPase activity accounts for only ~50% of the total ATPase activity of the microsomes. There are no differences in the Na⁺-K⁺-independent ATPase activity (i.e., the oligomycin or K⁺-insensitive component) in mid- or late-pregnant rats vs. virgins. In contrast, the specific Na⁺-K⁺-ATPase activity in renal cortex is diminished in both midpregnancy and late pregnancy. In renal medulla, there is a small (~20%), transient rise in activity at midterm, which falls in late pregnancy to the virgin value. In brain stem, the total and Na⁺-K⁺-ATPase-specific activities are similar in virgin, mid-, and late-pregnant tissues. This is shown graphically in Fig. 5, where specific Na⁺-K⁺-ATPase activity is shown in mid- and late-pregnant tissues, expressed as percent change from the virgin value.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

There is clear evidence of renal sodium retention in the second part of pregnancy in rats, despite abundant natriuretic stimuli, including the increased glomerular filtration rate (GFR) and renal plasma flow; the antimineralocorticoid action of progesterone; the natriuretic actions of increased dopamine, ANP, prostaglandins, and nitric oxide; and the decrements in plasma oncotic pressure, which decrease proximal tubular sodium reabsorption (7, 10, 11). There are also increases in some antinatriuretic influences, i.e., activation of the renin-ANG II-aldosterone axis, elevations in plasma deoxycorticosterone (21), and a fall in blood pressure (6, 9). Enhanced renal nerve activity may also contribute to the net gestational renal sodium retention, at least in sheep (1). Micropuncture studies have shown that, in rats, sodium reabsorption in the ascending loop of Henle and the collecting duct is increased throughout pregnancy. Proximal sodium reabsorption increases at midterm, although whether this persists late in pregnancy is controversial (5). One possible explanation for the net sodium retention is that either a specific or generalized loss of tubular responsiveness to natriuretic stimuli may occur. This would then allow the increased antinatriuretic stimuli to predominate. Indeed, the pregnant rat becomes refractory to multiple natriuretic stimuli (11, 20, 24, 25, 29).

If an increase occurs in Na⁺-K⁺-ATPase activity during pregnancy, in some or all segments of the nephron, this would act to generally oppose any natriuretic stimulus. This was suggested by Lindheimer and Katz (22), who reported that Na⁺-K⁺-ATPase activity in a homogenate of the entire kidney was higher in late-pregnant rats vs. virgins or midpregnant rats. The present study was directed to further investigate this possibility. Unexpectedly, instead of increased activity, we found a reduction in Na⁺-K⁺-ATPase activity in the renal cortex in mid- and late pregnancy. Although a small rise was seen at midterm, no alteration in Na⁺-transport activity was evident in the renal medulla in late pregnancy. The Na⁺-K⁺-ATPase activity in brain stem was also similar in virgin and pregnant rats. Thus the Na⁺-K⁺-ATPase activity data generally do not track with the micropuncture evidence of increased sodium reabsorption in various nephron sites (5), with the exception of the midterm increase in Na⁺ reabsorption in the distal nephron segments.

These findings of reduced renal cortical Na⁺-K⁺-ATPase activity were contrary to our hypothesis, as well as to earlier work (22). We therefore used a second technique to quantitate the amount of the Na⁺-K⁺-ATPase protein present in the kidney in pregnancy. The Na⁺-K⁺-ATPase is a heterodimer consisting of a  (structural)- and an  (catalytic)-subunit. In the rat kidney, the 1-subunit (complexed with 1) predominates, although there have been reports of 2- and 3-subunits present in some parts of the tubule (2, 12, 26). In the present study, we found evidence of only the 1-subunit of the Na⁺-K⁺-ATPase in the kidney of virgin and pregnant rats. The density of the catalytic 1-subunit of the Na⁺-K⁺-ATPase was reduced in cortex of mid- and late-pregnant rats. In medulla, the abundance of the 1-subunit of the Na⁺-K⁺-ATPase was not significantly affected by pregnancy, although a nonsignificant midterm increment paralleled the small rise in medullary Na⁺-K⁺-ATPase activity. However, by term, when renal sodium retention is most marked, medullary activity and abundance were not different from the virgin value. Thus the enzyme activity and protein quantitation data both indicate that mid- and late pregnancy are associated with a selective reduction in Na⁺-K⁺-ATPase in the renal cortex. The brain stem data are also internally consistent because the Na⁺-K⁺-ATPase activity and the abundance of the 2- and 3-subunits of the Na⁺-K⁺-ATPase are unaffected by pregnancy. Therefore, the present findings suggest that, in late pregnancy, a reduction occurs in Na⁺-K⁺-ATPase activity that is confined to cortical segments of the nephron. The difference between these findings and earlier work by Lindheimer and Katz (22) may have been due to an aberrant reduction in the level of Na⁺-K⁺-ATPase activity in one virgin group. In addition, use of a whole kidney homogenate would obscure differences in Na⁺-K⁺-ATPase activity between cortex and outer medulla. Another problem relates to the use of ouabain to determine the specific Na⁺-K⁺-ATPase activity since it is now recognized that the predominantly 1-subunit found in the rat kidney is relatively insensitive to inhibition by cardiac glycosides (8, 19).

The present evidence suggests that the blunted renal cortical Na⁺-K⁺-ATPase activity of pregnancy is due to a reduction in the density of the pump rather than, for example, a functional inhibition due to some circulating factor. In this regard, there have been reports of increased cardiac glycoside activity in pregnancy (14), although this would have little effect in the rat, even in vivo, in view of the insensitivity of the renal Na⁺-K⁺-ATPase of this species to ouabainlike compounds (8). Progesterone exerts potent antimineralocorticoid actions (21). Pregnancy levels of progesterone decrease the basal activity of the Na⁺-K⁺-ATPase in the proximal convoluted tubule and the cortical collecting duct of rat, as well as in the thick ascending limb of the loop of Henle (27). In addition, progesterone inhibits the stimulatory action of aldosterone on the cortical collecting duct. However, the inhibitory actions of progesterone apparently take place via inhibition of the mineralo-corticoid receptor and short-term reduction in Na⁺-K⁺-ATPase activity. Whether this also causes downregulation of pump density is unclear.

Overall, our findings of reduced renal cortical Na⁺-K⁺-ATPase activity and abundance during pregnancy were unanticipated, particularly since Na⁺-K⁺-ATPase is upregulated in the human erythrocyte during pregnancy (16). There are increases in specific ouabain-binding sites and Na⁺-K⁺-ATPase activity (measured by ⁸⁶Rb uptake) and decreased intracellular Na⁺/increased intracellular K⁺ in red blood cells of normal women in mid- and late pregnancy (16). The main isoforms of the erythrocyte Na⁺-K⁺-ATPase are the 1- and 1-subunits (31), as in the kidney. Why the Na⁺-K⁺-ATPase should be upregulated in erythrocytes, unchanged in brain stem, transiently increased then reduced in renal medulla, and downregulated in kidney cortex during pregnancy is unclear. It is unlikely, however, that the erythrocyte Na⁺-K⁺-ATPase is a good model for the Na⁺-K⁺-ATPase in transporting epithelia since the function and regulation of the enzyme is so different in these cell types.

In summary, we disproved our original hypothesis by showing that renal cortical Na⁺-K⁺-ATPase activity and abundance are downregulated, whereas renal medullary Na⁺-K⁺-ATPase activity and abundance are unchanged in normal mid- and late pregnancy. These data do not help to explain the massive and cumulative sodium and volume retention of normal pregnancy and, in fact, imply that whatever the net antinatriuretic stimulus of pregnancy actually is, it needs to overcome an adverse shift in the Na⁺-K⁺-ATPase activity and abundance, as well as the other natriuretic stimuli that occur in pregnancy (21).