QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.105.090589v1 316/1/106    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Neye, Y. Articles by Krippeit-Drews, P. Search for Related Content PubMed PubMed Citation Articles by Neye, Y. Articles by Krippeit-Drews, P.

CELLULAR AND MOLECULAR

HIV Protease Inhibitors: Suppression of Insulin Secretion by Inhibition of Voltage-Dependent K⁺ Currents and Anion Currents

Pharmazeutisches Institut, Tübingen, Germany

Received June 6, 2005; accepted September 12, 2005.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
We have shown before that the human immunodeficiency virus (HIV) protease inhibitors ritonavir and nelfinavir, but not indinavir, suppress insulin secretion from mouse pancreatic B-cells via reduction of the cytosolic free calcium concentration ([Ca²⁺]_c). This was not because of an effect on ATP-dependent K⁺ channels (K_ATP channels) or L-type Ca²⁺ channels. The study was intended to elucidate the mechanisms by which distinct HIV protease inhibitors decrease [Ca²⁺]_c and thus evoke their adverse side effect on insulin release. Membrane potential and whole-cell currents were measured with the patch-clamp technique, and [Ca²⁺]_c was determined with a fluorescence dye. Ritonavir and nelfinavir both inhibited the same component(s) of voltage-dependent K⁺ currents with a concomitant change in action potential wave form, whereas indinavir was ineffective. Comparison with other blockers of voltage-dependent K⁺ currents revealed that suppression of distinct noninactivating current component(s) altered action potential wave form and decreased [Ca²⁺]_c similar to ritonavir and nelfinavir, whereas blockage of inactivating component(s) was without effect. Complete inhibition of voltage-dependent K⁺ currents by 80 mM TEA⁺ drastically increased [Ca²⁺]_c, demonstrating that voltage-dependent K⁺ channels are not the sole target of ritonavir and nelfinavir. Accordingly, the Ca²⁺-lowering effect of ritonavir was preserved in the presence of 80 mM TEA⁺. This effect was mimicked by the anion channel blocker 4,4'-diisothiocyanatostilbene-2,2'-disulfonic acid (DIDS). Consequentially, ritonavir and nelfinavir inhibited a DIDS-sensitive anion current in B-cells. We suggest that ritonavir and nelfinavir decrease insulin secretion by inhibition of voltage-dependent K⁺ channels and anion channels, which are essential to provide counterion currents for Ca²⁺ influx across the plasma membrane.

ATP-dependent K⁺ channels (K_ATP channels) and voltage-dependent Ca²⁺ channels usually play the key role in stimulus-secretion coupling of pancreatic B-cells. ATP production by glucose metabolism inhibits the K_ATP channels, and the resulting depolarization leads to opening of Ca²⁺ channels, Ca²⁺ influx, and eventually insulin secretion (Ashcroft and Rorsman, 1989; Zhou and Misler, 1996). The Ca²⁺ action potential wave form and thus the Ca²⁺ current during an action potential are dependent on the opening of voltage- and/or Ca²⁺-dependent K⁺ channels (K_v and K_Ca channels) (Smith et al., 1990; Göpel et al., 1999; Su et al., 2001; MacDonald and Wheeler, 2003). Additionally, volume-sensitive anion currents may be involved in the regulation of glucose-induced electrical activity of pancreatic B-cells (Kinard and Satin, 1995; Best et al., 1996; Drews et al., 1998).

We have shown recently that ritonavir and nelfinavir, but not indinavir, diminish glucose-induced insulin secretion of islets in vitro. This effect was because of a reduction in the cytosolic free Ca²⁺ concentration ([Ca²⁺]_c) to basal values by ritonavir and nelfinavir. This decrease in [Ca²⁺]_c could not be counteracted by tolbutamide. Surprisingly, ritonavir had no direct effect on K_ATP or voltage-dependent Ca²⁺ currents (Düfer et al., 2004). Ritonavir affected the electrical activity of B-cells in a characteristic manner that was ascribed to the partial inhibition of voltage-dependent K⁺ currents (Düfer et al., 2004). Inhibition of a K⁺ current can, however, at first glance not explain a reduction in [Ca²⁺]_c and insulin secretion. Therefore, we scrutinized the effects of the HIV protease inhibitors, K⁺ current blockers and inhibitors of the volume-sensitive anion current (VSAC) on B-cell plasma membrane potential (V_m), ion currents, and [Ca²⁺]_c. The results let us conclude that the effect of ritonavir and nelfinavir is due to the inhibition of ion currents providing counterbalancing ions for charge compensation, which is a prerequisite for Ca²⁺ influx.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Cell Preparation. Experiments were performed on islets or dispersed pancreatic B-cells from fed Naval Medical Research Institute mice, killed by cervical dislocation or CO₂. The principles of laboratory animal care were followed (National Institutes of Health publication 85-23, revised 1985). Islet cells were isolated by collagenase digestion of the pancreas. Cells were dispersed in Ca²⁺-free medium and cultured up to 4 days in RPMI 1640 medium supplemented with 10% fetal calf serum, 100 U/ml penicillin, and 100 µg/ml streptomycin (Plant, 1988).

Solutions and Chemicals. Bath solution was composed of 140 mM NaCl, 5 mM KCl, 1.2 mM MgCl₂, 2.5 mM CaCl₂, and 10 mM HEPES, pH 7.4, adjusted with NaOH. Glucose was added in the concentrations indicated. For measurements of anion or voltage-dependent K⁺ currents, 0.1 mM tolbutamide was added to the bath solution. Whole-cell recordings of voltage-dependent K⁺ currents were performed with a pipette solution containing 130 mM KCl, 4 mM MgCl₂, 2 mM CaCl₂, 10 mM EGTA, 0.65 mM Na₂ATP, and 20 mM HEPES, pH 7.15, adjusted with KOH. The same solution was taken for measurements of anion currents, except that 2 mM Na₂ATP was used. Membrane potential recordings were registered with a pipette solution composed of 10 mM KCl, 10 mM NaCl, 70 mM K₂SO₄, 4 mM MgCl₂, 2 mM CaCl₂, 10 mM EGTA, 20 mM HEPES, and 250 µg/ml amphotericin B, pH 7.15, adjusted with KOH.

Fura-2AM was obtained from Molecular Probes (Eugene, OR), and RPMI 1640 medium and penicillin/streptomycin were from Invitrogen (Carlsbad, CA). Ritonavir was a kind gift from Abbott (Wiesbaden, Germany). Nelfinavir and indinavir were kindly provided by Prof. Dr. S. Laufer (Tübingen, Germany). Stock solutions of these substances were prepared with dimethyl sulfoxide. All other chemicals were purchased from Sigma Chemie (Deisenhofen, Germany) and Merck (Darmstadt, Germany) in the purest form available.

Patch-Clamp Recordings. Patch pipettes were pulled from borosilicate glass capillaries (Clarke Electromedical Instruments, Pangbourne, UK). They had resistances between 3 and 5 M when filled with pipette solution. Membrane currents were recorded at 27°C with an EPC-9 patch-clamp amplifier and the software, Pulse (HEKA, Lambrecht/Pfalz, Germany) in the voltage-clamp mode. Membrane potential was measured in current-clamp in the perforated-patch mode. Perforation usually occurred within 10 min after seal formation (series resistance < 30 M). Conventional whole-cell anion currents were measured at a holding potential of –70 mV and during 300-ms pulses to –80 and –60 mV at 15-s intervals in the presence of 100 µM tolbutamide. Currents through voltage-dependent K⁺ channels have been registered in the conventional whole-cell mode by 150-ms pulses from –70 to 0 mV applied every 15 s.

Analysis of Action Potentials. Full width at half-maximum amplitude of action potentials was determined in each experiment by averaging the values of the last five action potentials before changing the solution. The frequency of action potentials was calculated during the last 30 s before bath solution change.

Measurement of [Ca²⁺]_c. [Ca²⁺]_c was measured at 37°C by the fura-2 method according to Grynkiewicz et al. (1985) using equipment and software from TILL Photonics (Gräfelfing, Germany). The cells were loaded with fura-2AM (5 µM) for 30 min. Intracellular fura-2 was excited alternately at 340 or 380 nm by means of an oscillating mirror with a diffraction grating. The excitation light was then directed through the objective (PlanNeofluar 40x objective; Carl Zeiss GmbH, Jena, Germany) by means of a glass fiber light guide and a dichroic mirror. The emitted light was filtered (LP515 nm) and measured by a digital camera. The ratio of the emitted light intensity at 340-/380-nm excitation was used to calculate [Ca²⁺]_c using an in vitro calibration with fura-2 K⁺ salt.

Presentation of Results. Electrophysiological experiments and [Ca²⁺]_c are illustrated by recordings representative of the indicated number of experiments performed with different cells. At least two different cell preparations have been used for each series of experiments. If possible, the means ± S.E.M. are given in the text for the indicated number of experiments. The statistical significance of differences between means was assessed by a one-sample Student's t test or Student's t test for paired values when two samples were compared. Multiple comparisons were made by analysis of variance followed by Student-Newman-Keuls test. P 0.05 was considered significant.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Effects of Ritonavir, Nelfinavir, and Indinavir on Electrical Activity and Voltage-Dependent K⁺ Channels. Comparing the effect of ritonavir, nelfinavir, and indinavir on electrical activity of pancreatic B-cells revealed that ritonavir and nelfinavir clearly changed shape and frequency of glucose-induced action potentials (Fig. 1, A and B), whereas indinavir was without effect (Fig. 1C). The addition of 10 µM ritonavir increased the average length (full width at half-maximum amplitude) of action potentials from 20 ± 3 to 1373 ± 218 ms (n = 5, P 0.005) (Fig. 1A). Concomitantly, the frequency of action potentials decreased from 91 ± 7 to 20 ± 7 AP/min (n = 5, P 0.0001). Similarly, 20 µM nelfinavir caused a broadening of glucose-induced action potentials from 34 ± 13 to 537 ± 64 ms (n = 6, P 0.001) and reduced the frequency from 134 ± 32 to 26 ± 7 AP/min (n = 6, P 0.05) (Fig. 1B). Indinavir up to 70 µM did not change shape and/or frequency of action potentials (Fig. 1C). The length of action potentials was 17 ± 3 ms before versus 17 ± 3 ms after application of indinavir (n = 4). Likewise, the frequency did not change significantly [31 ± 10 AP/min under control conditions versus 59 ± 9 AP/min with indinavir (n = 4)]. Characteristic action potentials in the presence of the respective compound are shown in the insets of Fig. 1, A to C, on an extended time scale.

View larger version (42K): [in this window] [in a new window]   Fig. 1. Effects of ritonavir, nelfinavir and indinavir on electrical activity and voltage-dependent K+ currents. Ritonavir (10 µM; A) and 20 µM nelfinavir (B) changed shape and frequency of glucose-induced action potentials, whereas 70 µM indinavir (C) was without effect. Characteristic action potentials in the presence of the respective compounds are shown in the insets of Fig. 1, A to C, at an extended time scale. The experiments are representative of five with similar results for ritonavir, six for nelfinavir, and four for indinavir. 0.5G and 15G, glucose concentration of 0.5 and 15 mM, respectively. Ritonavir (10 µM; D) and 20 µM nelfinavir (E) partly inhibited the voltage-dependent K+ current, and 70 µM indinavir was without effect (F). The experiments are representative of four with similar results for ritonavir, six for nelfinavir, and seven for indinavir. Voltage-dependent K+ current was monitored every 15 s during 150-ms voltage steps from –70 to 0 mV. Drugs were added as indicated by the horizontal bars.

The changes in electrical activity and action potential shape caused by ritonavir and nelfinavir are unlikely due to changes in K_ATP or Ca²⁺ current (Düfer et al., 2004) but could be associated with an alteration in voltage-dependent K⁺ current. To evaluate this suggestion, we tested the influence of the three substances on the voltage-dependent K⁺ current in the conventional whole-cell mode (Figs. 1, D–F, and 2A). Ritonavir and nelfinavir inhibited noninactivating component(s) of the current, while leaving inactivating component(s) (Fig. 2A; Düfer et al., 2004). In contrast, the voltage-dependent K⁺ current was not influenced by indinavir (Fig. 1F). In the series with ritonavir, the peak current under control conditions in 0.5 mM glucose amounted to 500 ± 57 pA (n = 7). It was reversibly reduced to 234 ± 24 pA by treatment of the cells with 10 µM ritonavir (n = 7, P 0.01) (Fig. 1D). Similarly, 20 µM nelfinavir diminished the amplitude of the K⁺ current from 441 ± 51 to 222 ± 31 pA (n = 6, P 0.001) (Fig. 1E). In the presence of 70 µM indinavir, the amplitude of the K⁺ current remained almost constant with 523 ± 79 pA under control conditions versus 466 ± 87 pA with indinavir (n = 7) (Fig. 1F).

View larger version (30K): [in this window] [in a new window]   Fig. 2. Effects of ritonavir, clotrimazole, and TEA+ (2 mM) on voltage-dependent K+ currents, glucose-induced electrical activity, and [Ca2+]c in the presence of 15 mM glucose. Ritonavir (10 µM; A, n = 4), and 5 µM clotrimazole (B, n = 5) inhibited noninactivating component(s) of the voltage-dependent K+ current, whereas 2 mM TEA+ inhibited inactivating component(s) of the current (C, n = 7). Voltage-dependent K+ currents were monitored every 15 s during 50- to 150-ms voltage steps from –70 to 0 mV. Ritonavir (10 µM; D, n = 5) and 5 µM clotrimazole (E, n = 4) markedly changed the spike pattern of action potentials, which were not influenced by 2 mM TEA+ (F, n = 6). Ritonavir (20 µM; G, n = 8) and 5 µM clotrimazole (H, n = 9) suppressed the glucose-induced increase in [Ca2+]c. TEA+ (2 mM) did not affect [Ca2+]c in the presence of 15 mM glucose (I, n = 15).

Effect of Ritonavir on [Ca²⁺]_c in the Presence of 80 mM TEA⁺. We next tested whether the effect of ritonavir on [Ca²⁺]_c still occurs in the presence of 80 mM TEA⁺, a concentration at which voltage-dependent K⁺ channels are completely blocked. Surprisingly, ritonavir even decreased [Ca²⁺]_c augmented by the high concentration of TEA⁺ (Fig. 3). The addition of 80 mM TEA⁺ in the presence of 15 mM glucose caused a rapid increase in [Ca²⁺]_c consisting of a peak and a sustained plateau without oscillations. Subsequent application of 20 µM ritonavir provoked a clear drop in [Ca²⁺]_c (Fig. 3). In the presence of 15 mM glucose, the peak value of [Ca²⁺]_c amounted to 417 ± 49 nM and increased to 668 ± 58 nM after addition of 80 mM TEA⁺. Directly after application of ritonavir, [Ca²⁺]_c decreased to 90 ± 5 nM (n = 9, P 0.001). The inset demonstrates that voltage-dependent K⁺ currents were abolished by 80 mM TEA⁺ (peak control current was 395 ± 87 pA in this series of experiments versus –5 ± 3 pA with 80 mM TEA⁺, n = 6, P 0.001, Fig. 3, inset).

View larger version (24K): [in this window] [in a new window]   Fig. 3. Effect of ritonavir on [Ca2+]c in the presence of 80 mM TEA+. Despite complete block of voltage-dependent K+ channels, [Ca2+]c was markedly increased. Ritonavir (20 µM) caused an immediate drop in [Ca2+]c. One representative experiment of nine is shown. The inset shows the effect of 80 mM TEA+ on voltage-dependent K+ currents elicited by 150-ms voltage steps from –70 to 0 mV (n = 6).

Effect of DIDS on [Ca²⁺]_c in the Presence of 80 mM TEA⁺. So far, the results suggest that voltage-dependent K⁺ channels are not the only target of ritonavir. Another channel that is assumed to be involved in B-cell function is the volume-sensitive anion channel (Kinard and Satin, 1995; Best et al., 1996; Drews et al., 1998). Thus, we tested the effect of the anion channel blocker DIDS on [Ca²⁺]_c in the presence of 80 mM TEA⁺. Figure 4 shows that we could mimic the effect of ritonavir on [Ca²⁺]_c with the anion channel blocker DIDS during complete blockage of voltage-dependent K⁺ channels by 80 mM TEA⁺. The peak value of [Ca²⁺]_c in the presence of 15 mM glucose was 361 ± 50 nM (data not shown) and increased after addition of 80 mM TEA⁺ to 1017 ± 169 nM (n = 8, P 0.01). The application of 100 µM DIDS caused an immediate initial drop in [Ca²⁺]_c to basal values of 67 ± 9 nM (n = 8, P 0.001). In four of eight cells, this first drop was followed by a small increase or even oscillations in [Ca²⁺]_c (maximal peak value of 352 ± 98 nM) similar to the action of ritonavir observed in some experiments (Düfer et al., 2004). For comparison, in each experiment with DIDS, B-cells were also treated with 20 µM ritonavir.

View larger version (19K): [in this window] [in a new window]   Fig. 4. Effect of DIDS on [Ca2+]c in the presence of 80 mM TEA+. The combined effects of the voltage-dependent K+ channel blocker TEA+ and the anion channel blocker DIDS could mimic the effect of ritonavir on [Ca2+]c. The experiment is representative of eight with similar results. For comparison, 20 µM ritonavir was added after wash-out of DIDS.

Effect of Ritonavir and Nelfinavir on VSAC. To test ritonavir and nelfinavir for an effect on VSACs, K_ATP currents were blocked by 100 µM tolbutamide and DIDS-sensitive VSACs (Kinard and Satin, 1995; Drews et al., 1998) were evoked by lowering the osmolarity of the extracellular solution (Drews et al., 1998) (Fig. 5A). After decreasing the concentration of NaCl from 140 to 100 mM, the anion current developed with an average amplitude of –154 ± 20 pA. The addition of 20 µM ritonavir reversibly reduced the amplitude of the anion current to –60 ± 10 pA (n = 7, P 0.001). After wash-out of ritonavir, the current increased again to values of –145 ± 33 pA (n = 6, P 0.001). The same effect on VSACs was seen with nelfinavir (Fig. 5B). In this series of experiments, the hypoosmotic solution evoked a VSAC of –378 ± 72 pA (n = 5) and amounted to –17 ± 17 pA (n = 3, P 0.01) in the presence of 20 µM nelfinavir. A similar reduction to –20 ± 15 pA was achieved by 100 µM of the VSAC blocker DIDS (n = 3, P 0.01 versus control, not significant versus nelfinavir).

View larger version (26K): [in this window] [in a new window]   Fig. 5. Effect of ritonavir and nelfinavir on VSAC. A, ritonavir partly inhibited the volume-sensitive anion current evoked by lowering the osmolarity of the extracellular solution by 80 mOsM/l. One representative experiment of six is shown. B, nelfinavir and DIDS inhibited the volume-sensitive anion current evoked by hypoosmolarity (–80 mOsM/l). One representative experiment of three is shown for either substance. VSAC was monitored at a holding potential of –70 mV (solid line) by applying 300-ms voltage steps to –60 or –80 mV (upper and lower dashed traces) every 15 s.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
Inhibition of hormone release from B-cells may, besides effects on insulin signaling, contribute to disturbed glucose homeostasis observed in patients treated with HIV protease inhibitors (Behrens et al., 1999; Dubé, 2000; Woerle et al., 2003). We have shown before that ritonavir and nelfinavir, but not indinavir, inhibit insulin secretion of mouse pancreatic B-cells (Düfer et al., 2004). This is in agreement with observations by Schutt et al. (2004), who found insulin secretion of INS-1 tumor cells to be inhibited by ritonavir, nelfinavir, and saquinavir, but not by indinavir or amprenavir. The decrease in insulin secretion of native B-cells by ritonavir and nelfinavir is because of a decrease in [Ca²⁺]_c (Düfer et al., 2004). However, the underlying mechanism remained obscure since neither the K_ATP channel nor the voltage-dependent Ca²⁺ channel were influenced by ritonavir. Moreover, intracellular stores seem not to be involved in the ritonavir-induced lowering of [Ca²⁺]_c (Düfer et al., 2004). Measurements of the V_m revealed that ritonavir changes the wave form of action potentials and broadens the action potentials 75-fold, but they are still sensitive to the Ca²⁺ channel blocker D600. The change in action potential length was ascribed to an inhibition of distinct component(s) of voltage-dependent K⁺ currents by ritonavir and nelfinavir, leaving a fast inactivating component (Düfer et al., 2004). These results left us with the puzzling situation that ritonavir reduced [Ca²⁺]_c, although voltage-dependent Ca²⁺ channels appeared to be open. Additionally, at first glance, it seems not plausible that the closure of a hyperpolarizing (K⁺) current should reduce Ca²⁺ influx through voltage-dependent Ca²⁺ channels.

For pancreatic B-cells, the depolarization-dependent change in [Ca²⁺]_c, and not membrane depolarization, is the vital trigger signal for insulin secretion. Thus, Ca²⁺ current during Ca²⁺ action potentials must flow for a period of time sufficient to augment [Ca²⁺]_c. This is completely different from the signaling mechanism in nerve fibers where membrane depolarization, i.e., the Na⁺ action potential, constitutes the signal that influences cell function. During such an action potential in nerve fibers, the amount of ions crossing the membrane is too low to induce measurable changes in the intracellular ion concentration. Therefore, a change in [Ca²⁺]_c by Ca²⁺ influx, the prerequisite for insulin secretion, requires the concomitant outflux of a cation or influx of an anion. Counterions are needed because charge separation beyond the negligible amount that charges or recharges the membrane capacitor does not occur within solutions or across membranes. Consistently, it has been shown for Ca²⁺ release in stripped skeletal muscles fibers, which also lead to a significant change in [Ca²⁺]_c, that inhibition of K⁺ channels in the membrane of the sarcoplasmic reticulum blocks Ca²⁺ release from these organelles (Abramcheck and Best, 1989). To understand this inhibition of a current by suppression of counterion currents, one has to consider that at any given membrane potential, the sum of all ion currents is zero (if a net current would flow, it would change V_m according to Ohm's law). Consequently, if in a cell all ion channels except one are almost entirely inhibited, the current through the open channel must be marginal.

During glucose-induced electrical activity of mouse pancreatic B-cells, K_ATP channels are almost completely closed (Ashcroft and Rorsman, 1989). Apparently, only L-type Ca²⁺ channels (Plant, 1988; Schulla et al., 2003) and voltage-dependent K⁺ channels (Ashcroft and Rorsman, 1989; MacDonald and Wheeler, 2003) are open during the action potentials. Ritonavir inhibits the entire voltage-dependent K⁺ current except for a fast-inactivating component. This component is completely inactivated after 50 ms at latest and can thus only carry a marginal K⁺ current during the first 30 to 50 ms of the long action potentials lasting about 1 s (Fig. 1; Düfer et al., 2004). Therefore, during these ritonavir-modulated action potentials, the Ca²⁺ influx might be negligible because of the lack of the charge-compensating K⁺ current.

The hypothesis that the suppression of Ca²⁺ influx by ritonavir is because of counterion current inhibition is supported by other results presented here and in a previous publication (Düfer et al., 2004). Nelfinavir had very similar effects on voltage-dependent K⁺ currents, V_m, [Ca²⁺]_c, and insulin secretion, whereas indinavir did not affect these parameters at all (Fig. 1; Düfer et al., 2004). Clotrimazole had effects analogous to ritonavir and nelfinavir on voltage-dependent K⁺ currents and V_m and also diminished [Ca²⁺]_c (Fig. 2). In contrast, TEA⁺ at a low concentration of 2 mM also decreased the voltage-dependent K⁺ current to about 50% as the other substances but left a noninactivating (therefore long-lasting) component and had nearly no effect on V_m or [Ca²⁺]_c (Fig. 2). The latter result is expected because in this case counterions for Ca²⁺ entry are still available. However, even complete suppression of voltage-dependent K⁺ current does not decrease [Ca²⁺]_c. It has been demonstrated that blockage or knockout of voltage-dependent K⁺ channels results in an increase in insulin secretion (e.g., Henquin, 1990; Roe et al., 1996; MacDonald et al., 2001; Su et al., 2001; MacDonald et al., 2002). In addition, the augmenting effect of TEA⁺ on [Ca²⁺]_c at a concentration of 80 mM, which entirely blocked voltage-dependent K⁺ currents (Fig. 3, inset), ruled out that K⁺ currents are the only currents that can provide counterions for Ca²⁺ influx in B-cells. The high concentration of TEA⁺ drastically increased [Ca²⁺]_c, and ritonavir was still effective (Fig. 3). Since no K⁺ current can compensate the charge for Ca²⁺ influx under these conditions, we investigated whether anions can act as counterions, too. This idea is supported by two observations. The unspecific anion channel blocker DIDS had the same effect as ritonavir on [Ca²⁺]_c augmented by 80 mM TEA⁺ (Fig. 4). Ritonavir and nelfinavir both block VSACs (Fig. 5, A and B), which are believed to be involved in glucose-induced electrical activity (Best, 1997). This block was similarly effective as VSAC inhibition by DIDS (Fig. 5B). These results clearly demonstrate that only the combined inhibition of voltage-dependent K⁺ channels and VSACs, either achieved by simultaneous application of TEA⁺ and DIDS or by HIV protease inhibitors that suppress both currents, decreases [Ca²⁺]_c. This observation also explains the similarity in the action of ritonavir or nelfinavir and clotrimazole on [Ca²⁺]_c because the antimycotic also concomitantly blocks voltage-dependent K⁺ channels (Fig. 2) and VSACs (Welker and Drews, 1997).

From the observation that HIV protease inhibitors directly affect [Ca²⁺]_c and inhibit the effect of sulfonylureas on [Ca²⁺]_c (Düfer et al., 2004), one would suggest that insulinotropic oral antidiabetic drugs are less suitable for the treatment of HIV-infected patients than metformin and thiazolidinediones, which increase insulin sensitivity. These drugs are presently under evaluation in diabetic and glucose-intolerant HIV-infected patients treated with highly active antiretroviral therapy (Fantoni et al., 2003; Kamin et al., 2005).

In conclusion, ritonavir and nelfinavir but not indinavir directly interfere with B-cell electrical activity, [Ca²⁺]_c, and insulin secretion by inhibiting currents that are necessary to provide counterions for charge compensation to allow Ca²⁺ influx during action potentials.

	Acknowledgements

 
We acknowledge the purification of nelfinavir and indinavir by S. Laufer and colleagues (Tübingen, Germany). We thank I. Breuning for excellent technical support.

	Footnotes

 
This work was supported by grants from the Deutsche Forschungsgemeinschaft (Dr 225/6-1) and from the Deutsche Diabetes-Stiftung.

Part of this work was published in abstract form: Düfer M, Neye Y, Krippeit-Drews P, and Drews G (2003) The HIV protease inhibitor ritonavir impairs pancreatic B-cell function. Naunyn-Schmiedeberg's Arch Pharmacol 367: R64 and Krippeit-Drews P, Neye Y, Düfer M, and Drews G (2003) Ritonavir impairs pancreatic b-cell function by interference with K_v currents and VSAC. Pflueg Arch Eur J Physiol 445:S77.

Y.N. and M.D. contributed equally to this work.

doi:10.1124/jpet.105.090589.

ABBREVIATIONS: HIV, human immunodeficiency virus; VSAC, volume-sensitive anion current; AP, action potential(s); DIDS, 4,4'-diisothiocyanatostilbene-2,2'-disulfonic acid.

Address correspondence to: Dr. Peter Krippeit-Drews, Pharmazeutisches Institut, Auf der Morgenstelle 8, D-72076 Tübingen, Germany. E-mail: peter.krippeit-drews{at}uni-tuebingen.de

	References

Top Abstract Materials and Methods Results Discussion References

 

Abramcheck CW and Best PM (1989) Physiological role and selectivity of the in situ potassium channel of the sarcoplasmic reticulum in skinned frog skeletal muscle fibers. J Gen Physiol 93: 1–21.[Abstract/Free Full Text]

Ashcroft FM and Rorsman P (1989) Electrophysiology of the pancreatic beta-cell. Prog Biophys Mol Biol 54: 87–143.[CrossRef][Medline]

Behrens G, Dejam A, Schmidt H, Balks HJ, Brabant G, Korner T, Stoll M, and Schmidt RE (1999) Impaired glucose tolerance, beta cell function and lipid metabolism in HIV patients under treatment with protease inhibitors. AIDS 13: F63–F70.[CrossRef][Medline]

Best L (1997) Glucose and alpha-ketoisocaproate induce transient inward currents in rat pancreatic beta cells. Diabetologia 40: 1–6.[CrossRef][Medline]

Best L, Sheader EA, and Brown PD (1996) A volume-activated anion conductance in insulin-secreting cells. Pflügers Arch Eur J Physiol 431: 363–370.[CrossRef][Medline]

Cammalleri C and Germinario RJ (2003) The effects of protease inhibitors on basal and insulin-stimulated lipid metabolism, insulin binding, and signaling. J Lipid Res 44: 103–108.[Abstract/Free Full Text]

Drews G, Zempel G, Krippeit-Drews P, Britsch S, Busch GL, Kaba NK, and Lang F (1998) Ion channels involved in insulin release are activated by osmotic swelling of pancreatic B-cells. Biochim Biophys Acta 1370: 8–16.[Medline]

Dubé MP (2000) Disorders of glucose metabolism in patients infected with human immunodeficiency virus. Clin Infect Dis 31: 1467–1475.[CrossRef][Medline]

Düfer M, Neye Y, Krippeit-Drews P, and Drews G (2004) Direct interference of HIV protease inhibitors with pancreatic beta-cell function. Naunyn-Schmiedeberg's Arch Pharmacol 369: 583–590.[CrossRef][Medline]

Fantoni M, Del Borgo C, and Autore C (2003) Evaluation and management of metabolic and coagulative disorders in HIV-infected patients receiving highly active antiretroviral therapy. AIDS 17: S162–S169.[Medline]

Gatti G, Di Biagio A, Casazza R, De Pascalis C, Bassetti M, Cruciani M, Vella S, and Bassetti D (1999) The relationship between ritonavir plasma levels and side-effects: implications for therapeutic drug monitoring. AIDS 13: 2083–2089.[CrossRef][Medline]

Göpel SO, Kanno T, Barg S, Eliasson L, Galvanovskis J, Renstöm E, and Rorsman P (1999) Activation of Ca⁽²⁺⁾-dependent K⁽⁺⁾ channels contributes to rhythmic firing of action potentials in mouse pancreatic beta cells. J Gen Physiol 114: 759–770.[Abstract/Free Full Text]

Grynkiewicz G, Poenie M, and Tsien RY (1985) A new generation of Ca²⁺ indicators with greatly improved fluorescence properties. J Biol Chem 260: 3440–3450.[Abstract/Free Full Text]

Hardy H, Esch LD, and Morse GD (2001) Glucose disorders associated with HIV and its drug therapy. Ann Pharmacother 35: 343–351.[Abstract]

Henquin JC (1990) Role of voltage- and Ca²⁺-dependent K⁺ channels in the control of glucose-induced electrical activity in pancreatic B-cells. Pflügers Arch Eur J Physiol 416: 568–572.[CrossRef][Medline]

Hertel J, Struthers H, Horj CB, and Hruz PW (2004) A structural basis for the acute effects of HIV protease inhibitors on GLUT4 intrinsic activity. J Biol Chem 279: 55147–55152.[Abstract/Free Full Text]

Justman JE, Benning L, Danoff A, Minkoff H, Levine A, Greenblatt RM, Weber K, Piessens E, Robinson E, and Anastos K (2003) Protease inhibitor use and the incidence of diabetes mellitus in a large cohort of HIV-infected women. J Acquir Immune Defic Syndr 32: 298–302.[Medline]

Kamin D, Hadigan C, Lehrke M, Mazza S, Lazar MA, and Grinspoon S (2005) Resistin levels in human immunodeficiency virus-infected patients with lipoatrophy decrease in response to rosiglitazone. J Clin Endocrinol Metab 90: 3423–3426.[Abstract/Free Full Text]

Kinard TA and Satin LS (1995) An ATP-sensitive Cl^- channel current that is activated by cell swelling, cAMP, and glyburide in insulin-secreting cells. Diabetes 44: 1461–1466.[Abstract]

Koster JC, Remedi MS, Qiu H, Nichols CG, and Hruz PW (2003) HIV protease inhibitors acutely impair glucose-stimulated insulin release. Diabetes 52: 1695–1700.[Abstract/Free Full Text]

Lee GA, Rao MN, and Grunfeld C (2005) The effects of HIV protease inhibitors on carbohydrate and lipid metabolism. Curr HIV/AIDS Rep 2: 39–50.

Lien LF and Feinglos MN (2005) Protease inhibitor-induced diabetic complications: incidence, management and prevention. Drug Saf 28: 209–226.[CrossRef][Medline]

MacDonald PE and Wheeler MB (2003) Voltage-dependent K⁺ channels in pancreatic beta cells: role, regulation and potential as therapeutic targets. Diabetologia 46: 1046–1062.[CrossRef][Medline]

MacDonald PE, Ha XF, Wang J, Smukler SR, Sun AM, Gaisano HY, Salapatek AM, Backx PH, and Wheeler MB (2001) Members of the Kv1 and Kv2 voltage-dependent K⁽⁺⁾ channel families regulate insulin secretion. Mol Endocrinol 15: 1423–1435.[Abstract/Free Full Text]

MacDonald PE, Sewing S, Wang J, Joseph JW, Smukler SR, Sakellaropoulos G, Wang J, Saleh MC, Chan CB, Tsushima RG, et al. (2002) Inhibition of Kv2.1 voltage-dependent K⁺ channels in pancreatic beta-cells enhances glucose-dependent insulin secretion. J Biol Chem 277: 44938–44945.[Abstract/Free Full Text]

Plant TD (1988) Properties and calcium-dependent inactivation of calcium currents in cultured mouse pancreatic B-cells. J Physiol (Lond) 404: 731–747.[Abstract/Free Full Text]

Roe MW, Worley JF, Mittal AA, Kuznetsov A, DasGupta S, Mertz RJ, Witherspoon SM, Blair N, Lancaster ME, McIntyre MS, et al. (1996) Expression and function of pancreatic beta-cell delayed rectifier K⁺ channels: role in stimulus-secretion coupling. J Biol Chem 271: 32241–32246.[Abstract/Free Full Text]

Schulla V, Renström E, Feil R, Feil S, Franklin I, Gjinovci A, Jing XJ, Laux D, Lundquist I, Magnuson MA, et al. (2003) Impaired insulin secretion and glucose tolerance in beta cell-selective Ca(v)1.2 Ca²⁺ channel null mice. EMBO (Eur Mol Biol Organ) J 22: 3844–3854.[CrossRef][Medline]

Schutt M, Zhou J, Meier M, and Klein HH (2004) Long-term effects of HIV-1 protease inhibitors on insulin secretion and insulin signaling in INS-1 beta cells. J Endocrinol 183: 445–454.[Abstract/Free Full Text]

Smith PA, Bokvist K, Arkhammar P, Berggren PO, and Rorsman P (1990) Delayed rectifying and calcium-activated K⁺ channels and their significance for action potential repolarization in mouse pancreatic beta-cells. J Gen Physiol 95: 1041–1059.[Abstract/Free Full Text]

Su J, Yu H, Lenka N, Hescheler J, and Ullrich S (2001) The expression and regulation of depolarization-activated K⁺ channels in the insulin-secreting cell line INS-1. Pflügers Arch Eur J Physiol 442: 49–56.[CrossRef][Medline]

Vernochet C, Azoulay S, Duval D, Guedj R, Cottrez F, Vidal H, Ailhaud G, and Dani C (2005) Human immunodeficiency virus protease inhibitors accumulate into cultured human adipocytes and alter expression of adipocytokines. J Biol Chem 280: 2238–2243.[Abstract/Free Full Text]

Vigouroux C, Maachi M, Nguyen TH, Coussieu C, Gharakhanian S, Funahashi T, Matsuzawa Y, Shimomura I, Rozenbaum W, Capeau J, et al. (2003) Serum adipocytokines are related to lipodystrophy and metabolic disorders in HIV-infected men under antiretroviral therapy. AIDS 17: 1503–1511.[CrossRef][Medline]

Welker S and Drews G (1997) Imidazole antimycotics affect the activity of various ion channels and insulin secretion in mouse pancreatic B-cells. Naunyn-Schmiedeberg's Arch Pharmacol 356: 543–550.[CrossRef][Medline]

Woerle HJ, Mariuz PR, Meyer C, Reichman RC, Popa EM, Dostou JM, Welle SL, and Gerich JE (2003) Mechanisms for the deterioration in glucose tolerance associated with HIV protease inhibitor regimes. Diabetes 52: 918–925.[Abstract/Free Full Text]

Zhou Z and Misler S (1996) Amperometric detection of quantal secretion from patch-clamped rat pancreatic beta-cells. J Biol Chem 271: 270–277.[Abstract/Free Full Text]

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.105.090589v1 316/1/106    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Neye, Y. Articles by Krippeit-Drews, P. Search for Related Content PubMed PubMed Citation Articles by Neye, Y. Articles by Krippeit-Drews, P.