Stimulatory effects on Na⁺ transport in renal epithelia induced by extracts of Nigella arvensis are caused by adenosine

The plant nigella is a genre of three species: Nigella sativa L., Nigella damascena L. and Nigella arvensis L., commonly known as black seeds and belonging to the botanical family of Ranunculaceae. The three species have been in use in many Middle Eastern and Far Eastern countries as a natural remedy for over 2000 years. Nigella seeds are ascribed to have many medicinal properties in traditional medicine. In Arabic countries, these seeds are considered as a real panacea. So, they are commonly taken alone, in combination with honey or added to many food preparations. Consequently, many researchers have studied the antibacterial, antifungal and antihelmintic effects of nigella seeds(Agarwal et al., 1949; Akhtar and Riffat, 1991; Rathee et al., 1982). Previous work on nigella seed extracts has shown that it inhibits the growth of the bacteria Escherichia coli, Bacillus subtilis and Streptococcus faecalis (Saxena and Vyas,1986). The antimicrobial activity of N. sativa against several other species of pathogenic bacteria (Staphylococcus aureus,Pseudomonas aeruginosa) and pathogenic yeast (Candida albicans)has also been established (Hanafy and Hatem, 1991).

For a long time, plant remedies, including nigella, have been used to treat diabetes. It has been proposed that the anti-diabetic action of the nigella extracts may, at least partly, be mediated through decreased hepatic gluconeogenesis (al-Awadi et al.,1991). Traditionally, these seeds are well known for their action on stone dissolution in the kidney and bladder. Therefore, we wished to investigate the effects of N. arvensis at the molecular level in renal A6 cells.

We have demonstrated the effects of extracts of N. arvensis(NA) seeds on transepithelial Na⁺ transport in a distal tubule cell line, A6, isolated from the kidney of the toad Xenopus laevis, by recording short-circuit current (I_sc),transepithelial conductance (G_T) and transepithelial capacitance (C_T) and by analyzing the fluctuations induced by a reversible blocker of the apical Na⁺ channel[6-chloro-3,5-diaminopyrazine-2-carboxamide (CDPC)]. Analysis of the purified active fraction by mass spectrometry demonstrated the presence of adenosine as the single organic compound in the NA extract that had a stimulatory effect on Na⁺ transport.

A6 cells (obtained from Dr J. P. Johnson, University of Pittsburgh, PA,USA) were cultured as described previously(De Smet et al., 1995; Jans et al., 2000), grown on permeable culture support (Anopore, Nunc Intermed, Roskilde, Denmark) and used after 15-30 days of growth (passages 103-109). The epithelial monolayers were mounted with minimal edge damage in Ussing-type chambers for electrophysiological measurements. These chambers are suited for continuous perfusion of both compartments and rapid exchange of the solutions. All experiments were carried out under short-circuit conditions by continuously clamping transepithelial voltage to zero using a low-noise or high-speed voltage-clamp for noise analysis or capacitance measurements,respectively.

In a previous report from our laboratory(Van Driessche et al., 1999),we described in detail the equipment used and extensively discussed the theoretical background of the measurement of transepithelial capacitance(C_T). Briefly, in our study, we used sine-wave analysis at 2 kHz, 2.7 kHz, 4.1 kHz, 5.4 kHz and 8.2 kHz. The data in the present paper illustrate records at 4.1 kHz. Phase shift and amplitude ratio between the voltage and current signal was calculated using regression analysis. With these data, we calculated the parameters of the equivalent circuit of the epithelium represented by a simple RC network that consists of a series resistance, transepithelial capacitance and its equivalent parallel resistance. The graphical interface (Labview, National Instruments, Austin,TX, USA) enabled real-time display of transepithelial conductance(G_T), short-circuit current (I_sc) and C_T.

Model calculations based on a lumped two-membrane model demonstrated that,in the high-frequency range, C_T equals the equivalent capacitance of the series arrangement of the apical (C_ap)and basolateral (C_b1) capacitance(Van Driessche et al., 1999):1/C_T=1/C_ap+1/C_b1. As C_b1 is approximately 12 times larger than C_ap (Erlij et al.,1994), changes in C_T will mainly reflect alterations at the apical membrane, i.e. the result of endo- and exocytotic processes at that border.

We used a pulse protocol similar to that described by Blazer-Yost et al.(1998). The apical surface was alternately exposed to 10 μmol l^-1 and 40 μmol l^-1CDPC for 5 min. Current noise at both blocker concentrations was amplified,digitized and Fourier transformed to yield PDS during each 5 min period. The amiloride-insensitive current (I_ami) was measured by blocking the channels at the apical side with 50 μmol l^-1amiloride. The blocker-sensitive macroscopic current

The decoction was prepared by boiling 5 g of dried and pulverized Nigella arvensis seeds in 100 ml distilled water for 10 min. Using filter paper, the plant extract was then filtered (filtrate I). Subsequently,a sample (2 ml) of NA extract was filtered through a polyvinylidene difluoride (PVDF) syringe filter with a pore size of 0.45 μm (Alltech Europe, Laarne, Belgium) (filtrate II). In addition, the PVDF filter was washed with 1 ml 5% isopropanol in high-pressure liquid chromatography (HPLC)water (filtrate III). Filtrates II and III were evaporated in a Savant Speed-Vac concentrator (Savant Instruments, Hicksville, NY, USA) and then dissolved in 2.5 ml and 100 μl of distilled water, respectively. These fractions were used at a concentration of 250 μll^-1 to analyze activity on the Na⁺ channel. Filtrate II showed the maximum activity on Na⁺ transport.

A sample (2.5 ml) of filtrate II was lyophilized in a Savant Speed-Vac concentrator. The sample was dissolved in 300 μl of FPLC column buffer (100 mmoll^-1 ammonium bicarbonate in HPLC water, pH 7.5), and 250 μl was loaded onto a Superdex 200 gel-filtration column (Marsha Pharmacia Biotech AB, SE751-84, Upscale, Sweden). Flow rate was 0.5 ml min^-1, and fractions of 250 μl were collected. Elution of the various compounds was established by monitoring the absorbance at 280 nm. Each fraction was lyophilized and analyzed for its activity on the Na⁺ channel. The highest stimulating activity on Na⁺ transport was detected in fraction 108. It should be noted that the volatile column buffer allowed for a complete lyophilization without subsequent generation of high salt concentrations.

Fraction 108 was further purified by performing RP-HPLC using a C2/C18 column (μRPC C2/C18 SC2.1/10 column, Amersham Pharmacia Biotech,Buckinghamshire, UK) connected to a SMART system (Amersham Pharmacia Biotech). Operating conditions were as follows: solvent A, 0.1% trifluoracetic acid(TFA) in HPLC water, solvent B, 95% acetonitrile in 0.1% TFA. Column conditions were as follows: 0% solvent B for 7 min followed by a linear gradient to 70% solvent B in 83 min at a flow rate of 80 μl min^-1.

The sample was injected onto the reverse-phase column, which had previously been equilibrated in solvent A. The column was washed with 420 μl of solvent A. Subsequently, a linear gradient from 0% to 70% acetonitrile in solvent A was performed. Elution was monitored at three wavelengths (215 nm,254 nm and 280 nm), and peak fractions of fraction 108 were collected manually in 500 μl Eppendorf tubes based on the absorbance at 215 nm. Each fraction was lyophilized, dissolved in analysis buffer (Ringer solution; see composition below) and analyzed for its activity on Na⁺ transport in the renal epithelial cells. Activity was demonstrated in a 215 nm peak eluting at 20% acetonitrile, which corresponded to fraction 12. To identify the molecular identity of the active principle, this fraction was further analyzed by mass spectrometry.

Mass spectrometry analysis was performed on a Perkin Elmer API 3000 LC/MS/MS system (PE Biosystems, Foster City, CA, USA) equipped with a nanospray (Protana Engineering, Odense M, Denmark) at a flow rate of 1 μl h^-1. Calibration was performed externally with a polypropylene glycol test solution (Perkin Elmer test-kit) and horse heart myoglobin (16.95 kDa). 2 μl of the prepared fraction was introduced into the nanospray. Scans were made between m/z 5 and m/z 2000. Data from 50 shots to 100 shots were averaged to obtain the final spectrum.

In all experiments, the apical and basolateral Ringer solutions contained 102 mmoll^-1 Na⁺, 2.5 mmoll^-1 K⁺,2.5 mmoll^-1 HCO₃^-, 1 mmoll^-1Ca²⁺ and 104 mmoll^-1 Cl^- (pH 8; osmolality,200 mosmol kg^-1 H₂O). For the experiments in which we investigated Cl^- secretion, the apical solution was NaCl-free and,instead, contained 69 mmoll^-1N-methyl-D-glucamine sulphate [(NMDG)₂SO₄]; osmolality was 180 mosmol kg^-1 H₂O. At the end of this type of experiment, we removed Cl^- from the basolateral solution, which contained 102 mmoll^-1 Na⁺, 2.5 mmoll^-1 K⁺, 2.5 mmoll^-1 HCO₃^-, 1 mmoll^-1Ca²⁺ and 52 mmoll^-1 SO₄^2- (the osmolality was adjusted to 200 mosmol kg^-1 H₂O with sucrose).

Amiloride (50 μmoll^-1; Sigma, St Louis, MO, USA) was used to determine the amiloride-sensitive component of the I_sc. CDPC (Aldrich Chemical, Milwaukee, WI, USA; stock solution in dimethyl sulfoxide) was used in concentrations of up to 100 μmoll^-1. Nigella arvensis was brought from Fès, Morocco. The lyophilized NA extract was used in most experiments at a concentration of 250 μll^-1. We chose to use this concentration of NA extract because an absolute maximal stimulation was recorded with 500 μll^-1 (see Results) and because of the limited access to the extract. Adenosine (1 μmoll^-1,9-β-D-ribofuranosyladenine) was also purchased from Sigma.

For pooled data, means ± S.E.M. were calculated. Statistical significance was evaluated using a Student's t-test. P<0.05 was accepted as significant.

In the presence of NaCl-Ringer on both sides of the A6 epithelia, the addition of NA extract to the apical side evoked a small transient increase in I_sc and G_T (dashed lines, Fig. 1). The augmentation of I_sc and G_T following apical treatment was not significant and therefore is not further discussed in this paper. By contrast, addition of NA extract to the basolateral compartment gave rise to a pronounced increase in I_sc, G_T and C_T (solid lines, Fig. 1). Within 20 min, this increase reached its maximum effect of twice the initial value of I_sc and was followed by a steady state. The response was abolished when amiloride (50 μmoll^-1) was added to the apical bath. Mean values of I_sc, G_T and C_T are listed in Table 1.

Although amiloride is able to almost completely block I_sc (Fig. 1; Table 1), we verified the effect of NA extract on Cl^- secretion through the epithelium. We therefore abolished Na⁺ transport by incubating the epithelium with Na⁺-free apical solutions(Fig. 2A). The addition of basolateral NA extract in the absence of apical Na⁺(replaced by NMDG⁺) did not lead to an increase in I_sc and G_T. Subsequently, when apical NMDGCl-Ringer was replaced by NaCl-Ringer, I_sc increased rapidly, reached a peak and declined to a plateau. The peak in I_sc and decline to a steady plateau value are caused by Na⁺ self-inhibition, which has been described for several Na⁺-transporting epithelia. However, it should be noted that the plateau values of I_sc and G_T are markedly higher than the control values recorded at the beginning of the experiment. Moreover, amiloride was able to almost completely block I_sc and to strongly depress G_T. These observations confirm the effect of NA extract on Na⁺transport.

To further verify a possible effect of NA extract on Cl^- secretion, we performed experiments with a Cl^-gradient directed from the basolateral to the apical side(Fig. 2B). Apical NaCl-Ringer was replaced by (NMDG)₂SO₄-Ringer. No effect on I_sc and G_T was observed after addition of NA extract to the basolateral side. However, the increase in C_T was comparable with the effect observed in Fig. 1, demonstrating that the NA extract exerted its effect under these conditions. At the end of the experiment, basolateral NaCl-Ringer was replaced by Na₂SO₄-Ringer to check the Cl^- current, but practically no effect was observed. These data show that, with both protocols(Fig. 2A,B), the NAextract does not activate a Cl^- pathway.

Fig. 3 depicts the different steps that lead to the determination of the ON and OFF rates(k_ob and k_bo kinetics) of the interaction of CDPC with the Na⁺ channel. Fig. 3A shows the inhibition of I_sc caused by apical application of increasing concentrations of CDPC, ranging from 10 μmol l^-1 to 100 μmol l^-1, before and after basolateral stimulation with NAextract. The observed relative instability of I_sc after the application of different CDPC concentrations has been reported and discussed in the literature (Baxendale-Cox et al., 1997). It was attributed to the feedback regulation of Na⁺ transport and therefore becomes more pronounced at higher transport rates, as observed after stimulation of I_sc by the NA extract. Similarly, after washout of CDPC from the apical solution, just before NA extract application, relatively high transient overshoots in I_sc were observed. Typically,amiloride inhibited Na⁺ transport rapidly and completely when applied at the end of the experiment following the highest dose of CDPC.

Fig. 3B illustrates that 2πf_c data correlate linearly with the CDPC concentrations. Therefore, k_ob and k_bocan be determined by linear regression analysis using the following equation:2πf_c=k_ob[CDPC]+k_bo(see Materials and methods). The ON and OFF rates for CDPC during the control period were consistent with the rates previously reported(Jans et al., 2000). Treatment with NA extract did not change k_ob and slightly increased k_bo. Mean values of the kinetic parameters (k_ob, k_bo and K_B) are presented in Table 2.

As Na⁺ channels in the apical membrane of A6 cells are rate limiting for transepithelial Na⁺ transport(Granitzer et al., 1991), the increase in I_sc observed during application of NAextract could result from a rise in i_Na and/or N_T and/or P_o. To resolve this question, we performed noise analysis experiments. Fig. 4A illustrates typical I_sc responses to basolateral NA extract in such an experiment where apical [CDPC] was switched alternately between 10 μmol l^-1 and 40 μmol l^-1 every 5 min. It should be noted that basolateral NA was added in the presence of 10 μmol l^-1 apical CDPC and that the first 40 μmol l^-1 CDPC pulse was executed approximately 10 min after addition of the NAextract. I_sc increased from 8.62 μA cm^-2 to 22.12 μA cm^-2, and G_T increased from 0.20 mS cm^-2 to 0.36 mS cm^-2. In addition, indicated on these tracings is the consistent finding that the I_sc in these studies is amiloride sensitive, as shown by the depression of I_sc and G_T by 50 μmol l^-1 amiloride at the end of the experiment.

We used noise analysis to determine the contributions of i_Na, P_o and N_T to the I_sc. Current noise PDS were alternately measured during exposure to 10 μmol l^-1 and 40 μmol l^-1CDPC, thus providing the S_o and f_cvalues of the blocker-induced Lorentzians. Fig. 4B summarizes the influence of NA extract on I_sc, i_Na, N_T and P_o. NA extract increased I_sc from 8.3±0.44μA cm^-2 to 21.36±0.71 μA cm^-2. The most important factor involved in the activation of I_sc after application of NA extract was N_T, which increased from 0.75±0.06 μm^-2 to 3.54±0.14μm^-2. The relatively small decrease in i_Nafrom 0.42±0.01 pA to 0.36±0.01 pA during exposure to NAextract probably represents the immediate response to depolarization of the apical membrane by activation of the Na⁺ permeability. P_o in the presence of basolateral NA decreased from 0.29±0.02 to 0.18±0.01. Results are presented as means± S.E.M. (N=6).

Fig. 5A shows the elution position (fraction 108) of the activity during the separation of the NA extract by FPLC on a Superdex 200 column. This corresponds to compounds with a molecular mass of <2000 Da. Fraction 108, in turn, was separated using RP-HPLC (Fig. 5B). Fraction 12 of the separation of fraction 108, corresponding to a 215 nm peak eluting at 20% acetonitrile, showed maximum stimulatory effect on I_sc (data not shown). Therefore, it was taken for further identification. Functional tests with other fractions of FPLC and RP-HPLC did not demonstrate a stimulatory activity.

Mass spectrometry analysis (Fig. 6) allowed the identification of a 268-mass compound as adenosine(M_r 267.24), from which adenine (M_r135.15) is further derived. The compounds with other masses could be identified as contaminants generated during the purification procedure.

The possible role of adenosine was functionally tested by verifying its effect on Na⁺ transport and by subsequently investigating the effect of NA extract on Na⁺ transport(Fig. 7). The addition of NA extract 30 min after an adenosine response(Fig. 7) did not show an additive effect on apical membrane Na⁺ influx. We obtained similar effects on Na⁺ transport with adenosine. Moreover, adenosine blunted the effect of NA extract. The NA extract concentrations were chosen to elicit maximal responses. These data suggest that the stimulation by both agents occurs through a common signaling pathway and confirm that adenosine is the single organic compound in the NAextract that increased Na⁺ transport. Fig. 7 represents the mean of six experiments.

The present study demonstrates that the NA extract activates transepithelial Na⁺ transport in A6 cells. The receptors and/or mechanisms are located at the basolateral side of the cells, as apical administration of NA extract had only a small stimulatory effect on I_sc and G_T. The activation of transepithelial Na⁺ transport in A6 cells by basolateral NA extract evokes a parallel increase in C_T, I_sc and G_T(Fig. 1). As demonstrated previously (Van Driessche et al.,1999), the value of C_T is dominated by the value of C_ap, which, in turn, is proportional to the apical membrane area. Hence, our findings indicate that NA extract causes an increase of the apical membrane area of the epithelium. This increase in membrane area could be due to membrane trafficking and Na⁺- channel insertion in the apical membrane. Based on this observation, we performed noise analysis to further explore this hypothesis. As shown by the analysis of the CDPC-induced noise, the NA extract increases the number of open amiloride-sensitive Na⁺ channels in the apical membrane of A6 cells. This finding, together with the increase in apical membrane area, suggests a stimulation of epithelial transport by promoting the exocytotic transfer of membranes from the cytoplasmic compartment into the membrane surface. However, the possibility that pre-existing Na⁺ channels are activated by NA extract,independently of the transfer of new membranes, cannot be ruled out. The number of open Na⁺ channels increased threefold, while C_T increased by only 8%. These numbers could be interpreted as indicating that the additional membranes contain a considerably higher density of Na⁺ channels than the resting apical membrane. Similar results were obtained with insulin(Erlij et al., 1994) and aldosterone (Blazer-Yost et al.,1998). Previous data(Coupaye-Gerard et al., 1994)showed that aldosterone-induced Na⁺ transport in A6 cells was sensitive to brefeldin A (BFA), an inhibitor of exocytotic events. In addition, the decrease of I_sc after BFA was primarily the result of a decrease in open-channel density in the apical membrane of A6 cells (Fisher et al., 1996). Full elucidation of these phenomena will, however, require further investigation.

It is remarkable that the activation of Na⁺ transport by NA extract is exclusively due to an increase of apical Na⁺channels, whereas the i_Na and P_odecreased. The decrease in i_Na probably results from cell depolarization, as previously reported(Van Driessche and Zeiske,1985). In the present study, the depolarization was caused by an increase in apical membrane Na⁺ conductance(Fig. 4A). Possible mechanisms that could explain the decrease of P_o by 38% are, as yet,unknown. Despite appreciable NA-extract-related decreases of P_o, the net stimulation of transport occurs as a result of a considerable increase in N_o. Studies with forskolin,which is known to increase cytosolic cyclic adenosine monophosphate (cAMP),have shown similar effects (Els et al.,1991), suggesting a possible role for cAMP in the regulation of P_o. It is also possible that cytosolic Ca²⁺causes this decrease in P_o. In recent experiments with ionomycin, elevation of cytosolic Ca²⁺ induced by this ionophore led to a decrease of P_o in A6 epithelia(Helman et al., 1998).

To elucidate the mechanisms of action of the NA extract on transepithelial Na⁺ transport in A6 cells, we attempted to identify the principle active compound(s) by using biochemical approaches: FPLC,RP-HPLC and mass spectrometry. The molecule was identified as adenosine(Fig. 6). The function tests with adenosine confirmed that it stimulates active Na⁺ transport in A6 epithelia and that its action blunted the effect of the NA extract(Fig. 7).

Adenosine regulates both Na⁺ uptake and Cl^- secretion in A6 cells, as in other epithelia. It has been reported that the regulation of Cl^- secretion is mediated by A₁ adenosine receptors located at the apical cell surface and transduced by Ca²⁺ release from intracellular stores (Banderali et al., 1999; Schwiebert et al.,1992). Similar results were shown by a metabolically stable analogue of adenosine N⁶- cyclopentyladenosine (CPA),which binds to both the A₁ and A₂ adenosine receptors in A6 cells (Casavola et al.,1996) but has a higher affinity for the A₁ receptors. When CPA was added to the apical side, it induced an increase in Cl^- conductance by acting on the A₁ receptors. Recently,Di Sole et al. (1999) reported results indicating that stimulation of A₃ receptors induced an elevation of cytosolic Ca²⁺ in A6/C₁ cells. These findings were confirmed by Reshkin et al.(2000), who demonstrated the presence of A₃ receptors in the apical membrane. When added to the basolateral side, adenosine increases Na⁺ transport by interacting with basolaterally located A₂ receptors. The activation of Na⁺ conductance occurs through an adenylate cyclase-dependent mechanism (Lang et al., 1985). Moreover, Dobbins et al.(1984) have demonstrated that adenosine and some of its analogues increase cAMP levels, which results in secondary Cl^- secretion, indicating the presence of an A₂ adenosine receptor on rabbit ileum mucosal cells that activates adenylate cyclase. On the other hand, adenosine stimulation of electrogenic Na⁺ transport in renal cells has been demonstrated to occur, at least in part, through Ca²⁺-dependent signal transduction events and not through regulation of adenylate cyclase(Hayslett et al., 1995). The observation that increased levels of intracellular Ca²⁺ correlated with a two- to threefold increase in inositol (1,4,5)-trisphosphate suggests that Ca²⁺ was released from intracellular stores. Furthermore,Kurtz (1988) has demonstrated that, in isolated juxtaglomerular cells,activation of A₁ receptors is associated with an elevation of cyclic guanosine-3′,5′-monophosphate (cGMP) but not with changes in either cytosolic Ca²⁺ or cAMP, suggesting the involvement of yet another second messenger system for adenosine. An effect of adenosine on intracellular Ca²⁺ was also found in experiments where A₃ receptors were stimulated by 2-chloro-N⁶-(3-iodobenzyl)-adenosine-5′-methyluronamide,which activates Cl^- secretion by Ca²⁺ and cAMP-regulated channels.

In the present study, we did not find Cl^- secretion induced by adenosine or NA extract in A6 cells(Fig. 2). On the other hand, we have evidence that cAMP and/or Ca²⁺ activate Cl^-channels in the same clone of A6 cells(Atia et al., 1999; Zeiske et al., 1998). It is puzzling that, in this study, we found a marked activation of Na⁺absorption without affecting Cl^- secretion. The magnitude of the effect on Na⁺ transport suggests that, if the stimulation occurs through cAMP, its rise in cytosolic concentration should be significant and probably sufficient to activate Cl^- secretion. However, in experiments where the additive effect of forskolin and adenosine was tested,we found that forskolin could still markedly activate Na⁺ transport in tissues pretreated with adenosine, whereas the effect of adenosine after forskolin addition was rather small (F. Atia, unpublished observation). So, it remains possible that the cAMP levels reached by adenosine or NAextract treatment are still too small to activate Cl^- conductance but are sufficiently large enough to activate Na⁺ transport. This issue requires further investigation and leaves the possibility open of the involvement of a cAMP-independent pathway.

The present study is of importance because of the widespread use of spices and other plant products by humans. Plants have always been used in traditional medicine. Their effects may provide a source of inspiration for the development of new drugs based on careful scientific studies that are required to avoid adverse effects that may occur. In fact, there are numerous plants, including nigella, awaiting further investigation of their therapeutic potential.

This project was supported by research grants from the `Fonds voor wetenschappelijk onderzoek — Vlaanderen' (G.0179.99), the Interuniversity Poles of Attraction Program — Belgian State, Prime Minister's Office — Federal Office for Scientific, Technical, and Cultural Affairs IUAP P4/23. F. Atia was supported by a grant from the Belgian Ministry of Foreign Affairs (ABOS).