Differential Effects of Diabetes on Rat Choroid Plexus Ion Transporter Expression

All protocols used in this study were approved by the University of Arizona Institutional Animal Care and Use Committee and abide by National Institutes of Health guidelines. Diabetes was induced in male 300- to 325-g Sprague-Dawley rats (Harlan, Indianapolis, IN) via an intraperitoneal injection of 60 mg/kg STZ (Sigma, St. Louis, MO) in sterile phosphate buffered saline. Control animals were injected intraperitoneally with phosphate buffered saline. The animals were housed under standard 12-h light-dark conditions and received food and water ad libitum for 28 days. Induction of diabetes was assessed by weight changes and blood glucose levels.

Heparinized blood samples were collected and analyzed using an ABL 505 blood gas analyzer (Radiometer Copenhagen). CSF samples were taken from the cisterna magna before decapitation and analyzed as for blood.

Rats were anesthetized with sodium pentobarbital (64.8 mg/kg) and decapitated. The brains were rapidly removed, and the choroid plexuses of the lateral ventricles were removed. Protein was isolated from the choroid plexuses by incubating overnight in 6 mol/l urea buffer (6 mol/l urea, 10 mmol/l Tris, 1 mmol/l dithiothreitol, 5 mmol/l MgCl₂, 5 mmol/l EGTA, 150 mmol/l NaCl, pH 8.0, complete mini EDTA free protease inhibitor [one tablet/10 ml of buffer; Roche, Mannheim, Germany]). Protein was quantified using the bicinchoninic acid method (Pierce, Indianapolis, IN).

Protein samples (20 μg) were separated on Novex 4–12% Tris-glycine gels using an electrophoretic field at 125 V for 75–90 min. Proteins were transferred to polyvinylidene fluoride membranes using 240 mA at 4°C for 30 min. The membranes were then blocked using 5% nonfat milk/Tris-buffered saline (20 mmol/l Tris, 137 mmol/l NaCl, pH 7.6) with 0.1% Tween-20. Membranes were incubated overnight at 4°C with primary antibodies (anti-Na⁺-H⁺ exchanger [1:250; Transduction Laborotories]; anti-Na⁺-K⁺-2Cl⁻ cotransporter [1:2,000; The University of Iowa Hybridoma Bank]; anti-rat Na⁺-K⁺-ATPase α1-, β1-, and β2-subunits [1:2,000; Research Diagnostics]). Primary antibodies were chosen based on their ability to bind to rat transporters. Following incubation with primary antibodies, the membranes were washed with 5% nonfat milk/TBS buffer before incubation with respective secondary antibodies (anti-mouse and anti-rabbit [Amersham, Springfield, IL] at 1:2,000 and 1:3,000 dilutions, respectively, in PBS/0.5% BSA) for 30 min at room temperature. Membranes were developed using the enzyme chemiluminescence method (ECLplus; Amersham), and protein bands were visualized on X-ray film (Kodak, Rochester, NY). Semiquantification of the protein was done using Scion Imaging software (National Institutes of Health, Bethesda, MD), and the results were reported as percentage of control.

Efflux of the K⁺ analog ⁸⁶Rb⁺ from choroid plexuses from nondiabetic and diabetic rats was also investigated. Choroid plexuses were removed from the lateral ventricles, and efflux was studied using a method based on Johanson and Preston (18). In brief, the choroid plexuses were equilibrated in artificial CSF (aCSF) (NaCl 127 mmol/l, NaHCO₃ 20 mmol/l, KCl 2.4 mmol/l, KH₂PO₄ 0.5 mmol/l, CaCl₂ 1.1 mmol/l, MgCl₂ 0.85 mmol/l, Na₂SO₄ 0.5 mmol/l, glucose 5 mmol/l, pH 7.4, bubbled with 95% O₂/5% CO₂) at 37°C for 5 min. The osmalality of the aCSF was 275 osm. The plexuses were then incubated in aCSF containing 2 μCi ⁸⁶Rb⁺ (specific activity 5.3 mCi/mg; Amersham, Springfield, IL). Choroid plexuses were rinsed rapidly and placed in the elution chamber containing 2 ml aCSF; 200-μl samples of aCSF were taken at 0, 20, 40, 60, 80, 100, and 120 s. The remaining aCSF and choroid plexuses were taken, and all samples were prepared for counting by adding 1 ml TS2 tissue solubilizer (Research Products International, Mt. Prospect, IL) and incubated overnight. Before counting on a Beckman LS5000 counter (Beckman Instruments Fullerton, CA), 100 μl 30% acetic acid and 4 ml Budget Solve scintillation cocktail (Research Products International, Mt. Prospect, IL) were added to each sample.

Efflux rate constants were calculated for ⁸⁶Rb⁺ from the slope of the logarithmic plot of ⁸⁶Rb⁺ remaining in the choroid plexus against time. Total tissue ⁸⁶Rb⁺ was taken as 100% at time 0. The amount of ⁸⁶Rb⁺ remaining at each time point was based on sequential subtractions of the amount effluxed for each 20-s time point from the total radioactivity.

All data are presented as mean ± SE of measurements from 4–24 separate animals (numbers for individual experiments noted in text and figures). Statistical significance was calculated using Student’s t tests between control and diabetic values, using the Pharmacological Calculation System software package (19).

Diabetes was assessed in this study by monitoring weight changes and blood glucose levels of both PBS- and STZ-injected rats (Table 1). Control rats gained 101 ± 26 g over the 28-day period, while diabetic rats lost 68 ± 9 g. Weight loss was paralleled by a significant increase (P < 0.01) in blood glucose levels, from 181 ± 46 mg/dl in control to 601 ± 22 mg/ml in diabetic rats. Levels of electrolytes within the CSF and plasma were similar, with a small increase in plasma K⁺ levels in diabetic rats compared with controls (Table 2).

Figure 2A–C shows the expression of the α1-, β1-, and β2-subunits of the Na⁺-K⁺-ATPase transporter. A significant increase (P < 0.05) of ∼60% was seen in expression of the α1 catalytic subunit in diabetic rats as compared with controls (Fig. 2A), with no significant change in β1- or β2-subunit expression (Fig. 2B and C).

Figure 2D shows the expression of the Na⁺-K⁺-2Cl⁻ cotransporter, which showed a significant increase (40%, P < 0.05) in expression in diabetic choroid plexuses as compared with controls.

The Na⁺-H⁺ exchanger showed significantly decreased (40%, P < 0.05) expression in diabetic rats compared with controls (Fig. 2E).

Efflux of ⁸⁶Rb⁺ from preloaded choroid plexuses was increased significantly (66%, P < 0.01) in diabetic choroid plexuses as compared with controls (Fig. 3). The efflux constant for ⁸⁶Rb⁺ increased from 0.24 ± 0.2 min⁻¹ in controls to 0.40 ± 0.05 min⁻¹ in diabetic rats.

Ion transport in the choroid plexus plays a critical role in the production of CSF and thus ion homeostasis of the brain. CSF production by the choroid plexus is similar in mechanism to urine production in the kidney and relies on a balance in the transport of Na⁺ and K⁺. Ion imbalance and transporter dysfunction are common events in diabetes. Changes in either the expression or activity of a number of ion transporters have been reported in the diabetic kidney (1), cardiac muscle (20), cardiovasculature (13), and brain (3). There is also evidence of perturbations in the transport of various ions across the BBB in STZ-induced diabetes; Na⁺ and K⁺ uptake in the rat BBB decreases, while Cl⁻ and Ca²⁺ transport are not altered (2,21). There is also evidence that diabetes is a risk factor for normal-pressure hydrocephalus (6). Normal-pressure hydrocephalus is associated with reduced CSF drainage (22) and also, paradoxically, with increases in aqueductal CSF flow velocity (23,24).

In this study, we investigated the effects of STZ-induced diabetes on the expression of choroid plexus ion transporters. We focused on three transport systems that have been shown to play important roles in both the formation of CSF and the maintenance of CSF composition: Na⁺-K⁺-ATPase, Na⁺-K⁺-2Cl⁻ cotransporter, and Na⁺-H⁺ exchanger. Na⁺-K⁺-ATPase is located on the apical membrane of the choroidal epithelial cells. Inhibition of this transporter with cardiac glycosides has been shown to inhibit CSF formation in the rabbit (25,26), the cat (27), and the dog (25). In the rat, the cardiac glycoside ouabain inhibits the uptake of ⁸⁶Rb⁺ into isolated choroid plexus (28,29) and increases its efflux (28). The Na⁺-K⁺-ATPase transporter has also been implicated in maintaining CSF K⁺ levels (30,31). Na⁺-K⁺-ATPase is composed of three subunits (α, β, and γ). The α-subunit is a multipass transmembrane protein that has binding sites for ATP, Na⁺, K⁺, and cardiac glycosides; it is often referred to as the catalytic subunit, while the β-subunit is the regulatory subunit (32). Both α- and β-subunits are required for activity (33,34). In the rat choroid plexus, only the α1-, β1-, and β2-subunits are present (29,35). Diabetic choroid plexus had a significant increase in the expression of the α1-subunit, but no significant change in the β1- or β2-subunit (Fig. 2A–C). A similar response in Na⁺-K⁺-ATPase transporter expression was observed by Klarr et al. (29) while investigating the effects of hyperkalemia on choroid plexus Na⁺-K⁺-ATPase expression. It is interesting to note that in our study, plasma K⁺ levels were also elevated (Table 2).

The Na⁺-K⁺-2Cl⁻ cotransporter is expressed on both the basolateral and apical membranes of the choroid plexus (Fig. 1) (36). This transporter has also been implicated in CSF production. CSF production is decreased by bumetanide, a Na⁺-K⁺-2Cl⁻ cotransporter inhibitor (37). Uptake of Na⁺, K⁺, and Cl⁻ by isolated choroid plexus can be inhibited by bumetanide (18,28). Efflux of ⁸⁶Rb⁺ is also inhibited by bumetanide (18,28,29). In this study, we saw a significant increase in expression of Na⁺-K⁺-2Cl⁻ cotransporter (Fig. 2D).

The Na⁺-H⁺ exchanger is located on the basal membrane of the choroid plexus, and it is involved in transport of Na⁺ from the blood in exchange for H⁺ and important for choroid plexus pH regulation. Inhibition of this transporter reduces both CSF production in rabbit (38) and Na⁺ uptake from blood to choroid plexus in the rat (39). In contrast to both Na⁺-K⁺-ATPase and Na⁺-K⁺-2Cl⁻ cotransporter, the expression of Na⁺-H⁺ exchanger was reduced during diabetes (Fig. 2E). Expression of these three ion transporters during diabetes has been shown to vary in a tissue-specific manner. Na⁺-K⁺-ATPase α1-subunit expression is reduced in the aorta of STZ-induced diabetic rats without changes in Na⁺-K⁺-ATPase α1-subunit in soleus muscle (13). In general, this transporter has reduced activity in diabetic models (13,16,17). In diabetes, the Na⁺-K⁺-2Cl⁻ cotransporter was upregulated in the aorta with no change in expression in the heart (13).

We also investigated the effect of these transporter protein expression changes on the efflux of ⁸⁶Rb⁺ from preloaded choroid plexuses. ⁸⁶Rb⁺ is a marker for K⁺ transport and thus exits the cell via the Na⁺-K⁺-2Cl⁻ cotransporter along the gradient and renters the cell via ATPase transporters against the gradient. From Fig. 3, it is apparent that diabetic rats have an ⁸⁶Rb⁺ efflux rate approximately double that of the nondiabetic rats. This indicates that the transport of K⁺ into the CSF is higher in diabetic rat choroid plexuses than in controls. In our study, the control efflux constant of 0.24 ± 0.02 min⁻¹ is similar to the K⁺ efflux constants obtained previously (18). The efflux constant in diabetic rats was significantly greater (P < 0.05), at 0.40 ± 0.05 min⁻¹. Keep an colleagues observed a similar increase in K⁺ (⁸⁶Rb⁺) efflux either by inhibiting Na⁺-K⁺-ATPase (28) or by increasing aCSF osmalality (40). The increase of ⁸⁶Rb⁺ efflux in this study may be due to an inhibition of Na⁺-K⁺-ATPase, as shown by Keep (28), to an increase in cotransporter activity, or to a combination.

All three of the transporters investigated in this study have previously been shown to be modulated not only in numerous organs during diabetes, but also under various conditions in the choroid plexus. In this study, we saw an increased expression of the Na⁺-K⁺-ATPase α-1 subunit and the Na⁺-K⁺-2Cl⁻ cotransporter and a decreased expression of the Na⁺-H⁺ exchanger. During diabetes, a number of changes occur within the plasma. Perhaps most importantly for ion transport, plasma osmolality and K⁺ levels are elevated (41). Though we did not measure osmolality, we did observe a significant increase in the levels of plasma K⁺. Further, Arieff and Kleeman (41) measured a plasma glucose that is similar to our diabetic value of 600 mg/dl, this group concluded that the increase in osmolality was due to the increased glucose. Thus, we can assume that the osmolality of the plasma in our studies was also elevated. Despite the probable change in osmolality and the observed change in plasma K⁺, we saw no significant difference in CSF K⁺ levels, again similar to the study of Arieff and Kleeman (41). During hyperkalemia, CSF K⁺ levels were maintained despite increased plasma K⁺, and a decreased ⁸⁶Rb⁺ transport across both the BBB and the BCSFB (42). Hyperkalemia also led to an increase in choroidal Na⁺-K⁺-ATPase α1-subunit expression, but no effect was seen in efflux of ⁸⁶Rb⁺ (29). Osmolality can also modulate choroid plexus efflux of ⁸⁶Rb⁺. In isolated choroid plexus studies, Keep et al. (40) raised the osmolality of aCSF to 420 mOsm/kg and observed an increase in ⁸⁶Rb⁺ efflux. It is unlikely that the change in osmolality in our study would reach this level based on the study of Arieff and Kleeman (41). Furthermore, in the current study, the aCSF was of the same osmolality for both control and diabetic rat choroid plexuses. However, it is possible that the increased efflux may be a response of the choroid plexus to being transferred from an environment of high osmolality to one of normal osmolality.

Ion transporter expression can be modulated by a number of post-translational modifications. It is thus possible that the change in expression observed in this study may not be an increase in protein but an elongation of the blots by modifications such as phosphorylation and or glycosylation. All three of the transport mechanisms in this study have previously been shown to be modulated by post-translational mechanisms. The Na⁺-K⁺-ATPase α1-subunit contains a number of serine and threonine amino acids, which can be phosphorylated by protein kinase C (PKC). This phosphorylation can be stimulated by glucose (43), parathyroid hormone (44), and serotonin (45). In diabetes, high levels of glucose are present in the plasma (Table 1). Serine and threonine phosphorylation of Na⁺-K⁺-ATPase α-subunits has been linked with a downregulation of activity either by stimulating Na⁺-K⁺-ATPase endocytosis (46) or by inhibiting enzyme activity itself. Insulin increases the membrane content of Na⁺-K⁺-ATPase (47) via both a PKC-mediated and a tyrosine kinase-mediated mechanism (43), leading to an increase in Na⁺-K⁺-ATPase activity. From Fig. 2A, it is apparent that the increased expression is predominately due to a smearing/elongation of the blot, potentially indicative of post-translational modifications such as phosphorylation. In vitro studies of Na⁺-K⁺-ATPase have shown that in the presence of high glucose, α-subunits can be glycosylated, resulting in reduced activity of ATPase (48,49). Nonenzymatic glycosylation of cell membrane and plasma protein amide groups via the Amadori reaction is a common feature in diabetes (50,51) and has been reported in the CSF of subjects with neurological disorders (52). Either of these mechanisms or a combination of both could explain the increased expression of Na⁺-K⁺-ATPase α1-subunit in our study.

Na⁺-K⁺-2Cl⁻ activity can also be affected by phosphorylation. The nonspecific PKC activator phorbol 12-myristate 13-acetate inhibits the Na⁺-K⁺-2Cl⁻ cotransporter (53), though osmotic shock stimulates Na⁺-K⁺-2Cl⁻ cotransport via a PKC-δ-mediated phosphorylation event, thus indicating that various members of the PKC family have different effects on ion transport. Finally, Na⁺-H⁺ exchanger activity is stimulated by insulin via a PKC-ξ-mediated mechanism. Therefore, lack of insulin in diabetic rats could lead to a decrease in the levels of Na+-H+ exchanger on the basolateral membrane of the choroidal epithelium.

A change in choroid plexus transporter expression may effect either the composition or the production of CSF. From our study, it is apparent that the ionic composition of the CSF is not altered during diabetes. All three transporters are essential for maintaining Cl⁻, K⁺, and Na⁺ concentrations and pH of the CSF, as well as being the major driving forces of CSF production. Previous studies have shown that altering the activity of these transporters can alter the production of CSF. It is thus likely that diabetic animals have a different rate of CSF turnover than do normal animals. A change in CSF flow rate and/or turnover could reduce the ability of the CSF to compensate for alterations in brain extracellular fluid composition and thus contribute to tissue damage during pathological insults such as stroke or hydrocephalus.

The alteration of levels and/or activity of ion transporters at the BCSFB could have an effect on CSF production. There were, however, no major changes in the levels of the major CSF cations (Table 2), indicating that even though changes occurred in transporter expression levels, net CSF ion concentrations were not adversely affected during the early stages (28 days) of untreated type 1 diabetes in the rat. Rb⁺ efflux is increased in isolated choroid plexuses (Fig. 3); we would thus expect some change in ion concentrations in the CSF. This indicates that there may be alterations in other transporters or ion channels that compensate for the change in Na⁺-K⁺-ATPase. This study has investigated the effects of a short-term diabetic insult. In human type 1 diabetes, the disease manifests over a considerably longer time period. It would thus be interesting to investigate the long-term effects of ion transport changes on CSF dynamics. Inhibition of the three main transporters investigated in this study has previously been reported to inhibit CSF production (25–27,37–38). We have shown a decrease in expression of the Na⁺-H⁺ exchanger and a probable increase in phosphorylation of the Na⁺-K⁺-ATPase α1-subunit. This would indicate that both transporters have a decreased activity, potentially resulting in reduced production of balanced CSF by the choroid plexuses. These changes in transporters may have an effect on the long-term health of people with type 1 diabetes and may contribute to the increased probability of stroke and normal-pressure hydrocephalus in diabetes.