RSS Feeds
A Mechanism for Pentamidine-Induced Hyperkalemia: Inhibition of Distal Nephron Sodium Transport
- Thomas R. Kleyman, MD;
- Camille Roberts; and
- Brian N. Ling, MD
- From the University of Pennsylvania School of Medicine and Veterans Affairs Medical Center, Philadelphia, Pennsylvania, and Emory University School of Medicine and Veterans Affairs Medical Center, Atlanta, Georgia. Requests for Reprints: Brian N. Ling, MD, Emory University School of Medicine, Renal Division, 1364 Clifton Road Northeast, Atlanta, GA 30322. Acknowledgments: The authors thank Allyson Morrison and Elisabeth E. Seal for their technical assistance in preparing and maintaining the A6 cell cultures and rabbit cortical collecting tubule primary cultures. Grant Support: By a Veterans Affairs Merit Review Award, an American Heart Association Established Investigator Award, a Grant-in-Aid Award, a grant from the Center for Excellence Program, a National Institutes of Health Clinical Investigator Award (K08-DK02111), and an Emory University Research Committee Award.
Abstract
Objectives: To determine whether pentamidine directly affects the transport of renal ions and thus provides a mechanism for hyperkalemia, which develops in as many as 100% of patients with the acquired immunodeficiency syndrome (AIDS) who receive pentamidine for more than 6 days.
Design: Transepithelial and single-channel electrical measurements were made on two models of distal-nephron ion transport: an amphibian distal-nephron cell line (A6) and primary cultures of rabbit cortical collecting tubules.
Results: Luminal bath application of pentamidine to A6 monolayers inhibited the amiloride-sensitive, short-circuit current with a 50% inhibitory concentration of 700 µM (five experiments). In the principal cell apical membranes of cortical collecting tubule primary cultures, amiloride-sensitive, 4-picosiemen Na+ channels in cell-attached patches were also identified. When the luminal membrane was directly exposed to 1.0 µM of pentamidine in the patch pipette solution, channel activity decreased by 40% (11 experiments). Channel inhibition rapidly reversed with washout of intrapipette pentamidine (four experiments). In contrast, replacement of either the luminal bath outside the patch pipette (four experiments) or the serosal bath (five experiments) with pentamidine did not significantly affect Na+ channel activity in the patches.
Conclusions: Luminal or “urinary” pentamidine inhibits distal nephron reabsorption of Na+ by blocking apical Na+ channels in a manner similar to “potassium-sparing” diuretics (for example, amiloride and triamterene). This results in a decrease in the electrochemical gradients that drive secretion of distal nephron K+. Because pentamidine is eliminated through urinary excretion, this renal tubular effect provides a mechanism for pentamidine-induced hyperkalemia.
Pentamidine is an antiparasitic agent used to treat the opportunistic infection Pneumocystis carinii pneumonia [1-6]. Unfortunately, hyperkalemia is an important complication of therapy, observed in as many as 100% of patients with the acquired immunodeficiency syndrome (AIDS) receiving pentamidine for more than 6 days [1-4]. This elevation in the serum potassium level can be seen in the absence of adrenal insufficiency, hyporeninemic hypoaldosteronism, interstitial nephritis, or hyperglycemia (pentamidine-induced pancreatic islet cell dysfunction). Investigators have also reported azotemia in 25% to 95% of patients infected with the human immunodeficiency virus (HIV) who receive pentamidine [6-8]. However, hyperkalemia is usually out of proportion to the degree of coexisting renal insufficiency and is frequently associated with hyperchloremic metabolic acidosis [6, 7, 9]. These findings have led several groups to postulate that pentamidine might directly effect renal tubular K+ secretion [1, 4, 6].
Most renal K+ is excreted through secretory K+ channels located in the apical membrane of principal cells in the cortical collecting tubule [10-12]. The electrochemical driving force for distal nephron K+ secretion (that is, increased urinary lumen negativity and high intracellular K+ levels) is maintained by luminal Na+ entry through apical Na+ channels and serosal K+ uptake through the basolateral Na+/K +-adenosine triphosphatase pump [10-13].
Pentamidine is an aromatic diamidine structurally similar to “potassium-sparing” diuretics such as amiloride, triamterene, and trimethoprim [5, 13, 14] (Figure 1). We therefore applied transepithelial and single-channel measurement techniques to two well-established models of cortical collecting tubule ion transport (A6 amphibian cell line and primary cultured rabbit cortical collecting tubules) to investigate the effects of pentamidine on renal tubular Na+ reabsorption and, therefore, K+ secretion.
Methods
Transepithelial Measurements on A6 Distal-Nephron Cell Line Cultures
The A6 cell (American Type Culture Collection, Rockville, Maryland) subpassages 91 to 96 were grown to confluency on collagen-coated polycarbonate filters (Costar; Cambridge, Massachusetts) in the presence of 1.5 µM of aldosterone. Filters were mounted in a modified Ussing chamber, and macroscopic current was measured at room temperature with a DVC-1000 voltage clamp (World Precision Instruments, Sarasota, Florida) [14-16]. We measured short-circuit current under voltage-clamp conditions and added 10 µM of amiloride to the luminal bath at the end of each experiment to determine the amiloride-sensitive component of the short-circuit current. Bath solutions contained the following: 100 mM of NaCl, 4 mM of KCl, 2.5 mM of NaHCO3, 1 mM of KPO4, 1 mM of CaCl2, 11 mM of glucose, and 10 mM of n-2-hydroxyethylpiperazine-n-2-ethanesulfonic (HEPES) (pH, 7.4).
Patch Clamp Measurements on Primary Cultures of Rabbit Cortical Collecting Tubules
As previously described [17], renal cortices from New Zealand white rabbits (body weight, 1 to 2 kg) were collagenase digested and subjected to Percoll density centrifugation. Rabbit cortical collecting tubule fragments were separated and grown to confluency on permeable, collagen-coated Millipore-CM filters (Millipore Corp., Bedford, Massachusetts) in the presence of 1.5 µM of aldosterone. Unitary channel events were measured at 37 °C with a List EPC-7 Patch Clamp (Medical Systems Corp., Greenvale, New York). Data were digitized, recorded, and analyzed as previously described [10, 14, 17]. Patch pipette and extracellular bath solutions consisted of a physiologic saline solution containing the following: 140 mM of NaCl (final NaCl concentration after titration to a pH of 7.4 with NaOH), 5 mM of KCl, 1 mM of CaCl2, 1 mM of MgCl2, and 10 mM of HEPES (pH, 7.4).
Chemicals
Pentamidine (1,5-bis[p-Amidinophenoxyl]-pentane bis[2-hydoxyethane-sulfonate salt]) was of the highest commercial grade available (Sigma Chemical, St. Louis, Missouri). We added appropriate solvent vehicles to control baths that by themselves did not change the Na (+) channel activity.
Results
The amiloride-sensitive component of the short-circuit current is a measure of the net Na+ transport across the renal epithelium [14, 16, 18]. When we applied various concentrations of pentamidine to the solution, bathing the luminal or “urinary” surface of A6 distal nephron cell monolayers, we observed a dose-dependent inhibition of the amiloride-sensitive, short-circuit current (five experiments) (Figure 2). This inhibition developed rapidly, and 50% blockage of the amiloride-sensitive, short-circuit current (50% inhibitory concentration) occurred with 700 µM of pentamidine at a pH of 7.4.
In previous studies, we have characterized the amiloride-sensitive, 4-picosiemen Na+ channel that is responsible for physiologic, mineralocorticoid-dependent Na+ reabsorption in the mammalian distal nephron [17]. We therefore examined the effect of pentamidine on 4-picosiemen Na+ channel activity in apical, cell-attached patches on principal cells of primary cultured rabbit cortical collecting tubules. The results, summarized in Table 1, show that addition of 1.0 µM of pentamidine to the serosal bath or luminal bath (outside the cell-attached pipette) did not significantly affect Na+ channel activity within the unexposed patch membrane (no pentamidine was added to the pipette solution). In contrast, when pentamidine was placed in direct contact with the luminal surface of the patch membrane (1.0 µM of pentamidine in the pipette solution), Na+ channel activity (measured as the number of channels times the open probability) decreased to 40% of control values.
The pentamidine-induced inhibition of Na+ channel activity appeared to be primarily caused by a decrease in open probability (percentage of time an individual channel is open) rather than the number of channels per patch (membrane channel density) (Table 1). We confirmed this impression in four cell-attached patches that contained only one Na+ channel and therefore permitted us to directly calculate open probability (Figure 3). In these latter experiments, we could also show reversibility of the inhibitory effects of luminal pentamidine on Na+ channel kinetics.
Discussion
The electrochemical driving force for K+ secretion in the cortical collecting tubule is maintained by luminal Na+ entry through apical Na+ channels and by serosal K+ uptake through the basolateral Na+/K+-adenosine triphosphatase pump [10-12]. We have previously shown that the mechanism of action for “potassium-sparing” diuretics, such as amiloride, trimethoprim, and triamterene, is direct blockade of these apical Na+ channels [13, 14, 17, 18]. Therefore, unlike most diuretics, which promote both natriuresis and kaliuresis, these latter drugs actually inhibit kaliuresis [13, 16].
In this study, we showed that luminal exposure to pentamidine inhibits Na+ reabsorption in both mammalian and amphibian distal nephron cells. Luminal pentamidine inhibited both amiloride-sensitive, macroscopic short-circuit currents and individual Na+ channels at concentrations greater than 50 µM and 1.0 µM, respectively. Differences in the apparent inhibitory constant of pentamidine for the Na+ channel in rabbit principal cells and in amphibian A6 cells may reflect either subtle variabilities within the structure of Na+ channels expressed in mammalian and amphibian renal cells or the different techniques used to examine Na+ transport [14, 19].
Little information is available on the pharmacokinetics of pentamidine in humans [5]. Pentamidine has a serum half-life of 5 to 6 hours after one parenteral dose, but the half-life increases to a mean of 52 ± 89 hours after multiple doses [20]. Only 15% to 20% of a single intramuscular dose is excreted daily in the urine, yielding urinary concentrations of 20 to 25 µg/mL (1 µg/mL ≈ µM) [5]. In patients with impaired renal function, parenteral pentamidine is excreted in the urine at a rate of 1.85 to 7.13 mg/d [20]. Urinary concentrations after daily administration of aerosolized pentamidine range from 1.3 to 778 ng/mg of creatinine per milliliter [21]. Therefore, most of the administered pentamidine becomes protein- and tissue-bound, with an estimated volume of distribution of 3 L/kg body weight and the greatest accumulation occurring in the kidney [5]. Urinary levels are detectable for months after therapy is discontinued. In patients with AIDS, autopsy studies have found detectable levels of pentamidine in renal tissue for as long as 1 year after the last dose [1]. Tissue depot release probably accounts for the wide range of reported urinary concentrations and the observation that pentamidine-induced hyperkalemia can persist for days after the drug is withdrawn [2].
In conclusion, pentamidine therapy for treating HIV-infected patients with P. carinii pneumonia can be associated with life-threatening hyperkalemia. We have shown that pentamidine, at concentrations found clinically in the urine, directly and reversibly blocks apical Na+ channels in a manner similar to potassium-sparing diuretics. The result is a decrease in the electrochemical driving force for both K+ and H+ secretion in the cortical collecting tubule. It is therefore not surprising that hyperkalemia and hyperchloremic metabolic acidosis are observed in patients treated with pentamidine, amiloride, triamterene, or trimethoprim [6, 7, 9, 14, 22]. These renal tubular effects provide a mechanism for pentamidine-induced hyperkalemia in patients without severe renal failure, tubulointerstitial damage, adrenocortical insufficiency, or hyporeninemic hypoaldosteronism.
Portions of this work were presented at the American Society of Nephrology Annual Meeting in November 1993.
- Copyright ©2004 by the American College of Physicians
Related Clinical Content
