Carlo Brugnara, M.D.
Departments of Pathology and Laboratory Medicine
The Children's Hospital
Harvard Medical School, Boston, Massachusetts, 02115, USA

Address correspondence to:
Carlo Brugnara, M.D.,
Department of Laboratory Medicine
The Children's Hospital
300 Longwood Avenue, Bader 760
Boston MA 02115
Phone: (617) 355-6610
FAX : (617) 355-6081

revised December 23, 2000



Cells with a markedly increased Hb S concentration are a prominent feature of sickle cell disease, as a consequence of the loss of K, Cl and water from the erythrocyte. The extreme dependence of polymerization kinetics on Hb S concentration means that these dehydrated erythrocytes rapidly sickle when deoxygenated. Blockade of K loss from the erythrocyte should, therefore, prevent the increase in Hb S concentration and reduce erythrocyte sickling. Detailed knowledge of the mechanisms leading to cell dehydration makes this a viable therapeutic option. Two ion transport pathways, the K-Cl cotransport and the Ca2+-activated K+ channel play prominent roles in the dehydration of sickle erythrocytes. Possible therapeutic strategies include inhibition of K-Cl cotransport by increasing red cell Mg2+content and inhibition of the Ca2+-activated K channel by oral administration of clotrimazole.
AA - normal subjects with red cells containing Hb A; ChTX - Charybdotoxin; DIDS - di-isothiocyano-disulfonyl stilbene; DIOA- [(dihydroindenyl)oxy]alkanoic acid; DTT- dithiothreitol; ISC- irreversibly sickled cells; KTX - kaliotoxin; L1- 1,2-dimethyl-3- hydroxypyrid-4-one; MCHC- mean corpuscular hemoglobin concentration; MCV- mean corpuscular volume; SS - homozygous sickle cell anemia.
Note: Parts of this review were taken from: Brugnara, C. Red cell dehydration in pathophysiology and treatment of sickle cell disease. Curr. Op. Hematol. 1995. 2:132-138.

Deoxygenation leads to formation of Hb S polymers and cell sickling. Detailed studies of the kinetics of Hb S polymerization have shown a latency phase preceding the formation of the polymer strands and their explosive growth inside the erythrocyte . This latent phase (delay time) is inversely proportional to the 15-35th power of Hb S concentration. Thus, small changes in Hb S concentration markedly affect Hb polymerization and cell sickling. The delay time plays a crucial role in the pathophysiology of sickle cell disease. If the delay time for Hb S polymerization is shorter than the capillary transit time, sickling will occur in the capillaries with likely vaso-occlusion. If the delay time is prolonged by decreasing the cell Hb S concentration so that it exceeds the capillary transit time, polymerization and sickling would occur in the venulae, with no vaso-occlusion.

Hyponatremia have been shown to lead to decreased red cell mean corpuscular hemoglobin concentration (MCHC) and sickling but a subsequent study failed to replicate these findings , suggesting that the levels of clinical hyponatremia required to induce significant cell swelling are difficult to maintain. Additional strategies aimed at inhibiting Hb S polymerization by increasing Hb F concentration have included treatment with hydroxyurea (alone or with erythropoietin) and with butyrate derivatives.

Ion transport and dehydration in sickle erythrocytes:
One of the distinguishing characteristics of sickle erythrocytes is the presence of cell dehydration. The fraction of red cells containing dense dehydrated cells with markedly increased cell Hb S concentration (to values of 40-50 g/dL compared with the normal 33 g/dL) has the highest percentage of irreversibly sickle cells (ISC), which maintain their deformed shape even in the presence of normal O2 tension. This dense fraction contains mostly cells with low Hb F content and dehydrated reticulocytes. There are probably different processes leading to formation of dehydrated cells, with a "fast track" process leading to formation of dehydrated reticulocytes and a slow process yielding dehydrated red cells. In either case, dehydration is due to loss of cell potassium. The loss of K+ is partially offset by an increase in cell Na+ content, probably resulting from cell membrane damage, increased Na+ leak and relative Na-K pump inhibition. There are four major mechanisms for K+ loss and sickle cell dehydration:

A) Ca2+-activated K+ channel (Gardos pathway):
When the intracellular free Ca2+ concentration is increased in human red cells, a large K+ loss with accompanying movement of Cl- and water is observed. This effect is due to activation of a Ca2+activated K+ channel, first described by Gardos . The Gardos channel of human red cells belongs to a family of Ca2+activated K+channels, present in several cell types . Ligand binding studies with 125I- charybdotoxin (ChTX), a peptide toxin, specific inhibitor of the channel, have indicated the presence of 100-150 binding sites per erythrocyte. The red cell Gardos channel is inhibited by clotrimazole and other imidazole antimycotics.
Schematic representation of the Gardos channel in erythrocytes
Figure 1. Schematic Representation of Gardos Channel Activity in Sickle Erythrocytes. All red cells contain Gardos channel proteins in their membranes. Sickle cells accumulate excessive quantities of calcium, that activate the Gardos channel (illustrated as the blue circle on the red cell membrane.) The Gardos channel is a protein "pump" that expels intracellular potassium when activated by calcium. The red cells loose water along with calcium. The result is a higher intracellular hemoglobin concentration. This promotes polymerization of deoxygenated HbS. 

The cDNA of a high conductance Ca2+-activated K+ channel (maxi-K+ channel) of mouse brain and skeletal muscle has recently been cloned. The molecular identity of the intermediate conductance Gardos channel of red cells remains unknown.

 In normal red cells, loss of cell K+ via the Gardos channel plays a role in preventing colloido-osmotic lysis during transient complement activation on the red cell surface . The Gardos channel of sickle cells, either alone or in conjunction with K-Cl cotransport, plays a major role in cell dehydration. In vitro dehydration of sickle erythrocytes depends on external Ca2+ and can be prevented by inhibitors of the Gardos channel such as charybdotoxin, nifedipine or clotrimazole.

B) K-Cl cotransport:
In human red cells, this system was first identified in erythrocytes of patients homozygous for Hb C disease and subsequently in the least dense, reticulocyte-rich fraction of normal erythrocytes (AA cells) and in sickle cells (SS cells). Cell age appears to be one determinant of K-Cl cotransport activity in human red cells, since the system is active almost exclusively in normal reticulocytes and not in normal mature red cells. Another important determinant of this system's activity is the presence of positively charged hemoglobins (C and S). Studies in different Hb variants have indicated that activation of K-Cl cotransport is not a common feature of all relatively positively charged Hb, but is limited to positively charged variants at ß-6 and ß-7. Activation of K-Cl cotransport is also observed in ß-thalassemia intermedia erythrocytes. A large portion of this activation is most likely a consequence of the oxidative damage of the red cell membrane, since it can be reduced in vitro by exposure to dithiothreitol (DTT). Studies have shown that SS cells can become denser via the loss of K+ mediated by the K-Cl cotransport system. Cytoplasmic acidification of SS erythrocytes leads to activation of the K-Cl cotransport and dehydration. Cells with a low content of Hb F display elevated K-Cl cotransport activity, which can account for the rapid dehydration of reticulocytes and their increase in cell density. Stimulation of K-Cl cotransport by cell swelling in rabbit and human erythrocytes is inhibited by protein phosphatases inhibitors.

There are no known inhibitors of sufficient specificity to be used in patients to prevent dehydration via the K-Cl cotransport system. [(Dihydroindenyl)oxy]alkanoic acid, (DIOA), partially inhibits K+ movement via K-Cl cotransport. Red cell Mg2+ content is an important modulator of the activity of this system. Increasing cell Mg2+above the physiologic levels markedly decreases K-Cl cotransport activity. The increase in free cell magnesium levels during deoxygenation is responsible for reducing the volume sensitive flux via K-Cl cotransport in both SS and AA cells. Thus, any maneuver leading to an increase cell Mg2+ content should inhibit dehydration via the K-Cl cotransport system.

 Two additional pathways are potentially involved in sickle cell dehydration but their contribution accounts for only a small fraction of the total K+ loss compared with the effect of the two pathways described above. They are:

 C) Deoxygenation induced Na and K fluxes and Na-K pump
Original studies by Tosteson and subsequent studies by others have characterized the increased Na+ and K+ permeabilities associated with erythrocyte sickling. These deoxygenation induced Na+ and K+ fluxes are distinguished by their insensitivity to inhibitors of other red cell transport systems, by their chloride- and membrane potential-independence and by the inhibitory effect of di-isothiocyano-disulfonyl stilbene (DIDS). The increased Na+ and K+ permeability of sickling is associated with increased entry of Ca2+, which is also sensitive to DIDS and inhibitors of Ca2+ channels such as nifedipine. Activation of the Na-K pump, which has a 3 Na+out/2 K+ in stoichiometry, may lead to cell dehydration. Until recently, no inhibitors of possible clinical used were available for the deoxygenation fluxes. However, there is evidence for a specific inhibition of the deoxygenation-induced fluxes by dipyridamole (Persantine¨), with an effective half-maximal inhibition around 0.5-1 µM. Based on the known effects of dipyridamole on platelet and endothelial cell functions, there are now multiple rationales for testing the effect of this drug in sickle cell disease.

 D) Oxidative damage of the cell membrane and K loss
Oxidative damage to the sickle cell membrane is the result of :

  1. accelerated (auto)oxidation of Hb S
  2. Decompartimentalization of iron, with membrane deposition of denatured hemoglobin, ferritin, free heme and iron
  3. Abnormal membrane deposition of iron induces oxidative damage of protein thiols and lipids
Oxidative damage is responsible for activation of K-Cl cotransport in an in vitro model of human thalassemia and for a portion of the abnormal activity of K-Cl cotransport in sickle erythrocytes (based on the inhibitory effect of dithiothreitol, DTT). A possible therapeutic approach based on reducing membrane-associated iron with the use of the orally -absorbable iron chelator L1 (1,2-dimethyl-3- hydroxypyrid-4-one) is currently being considered for thalassemia and sickle cell disease.

Clotrimazole and blockade of Gardos channel
Figure 2. Structure of clotrimazole
Schematic representation of clotrimazole
Clotrimazole and other imidazole antimycotics are potent and specific inhibitors of the Ca2+-activated K+ channel pathway of normal and sickle erythrocytes. The original report of Alvarez et al. described the inhibition of the normal human red cell Gardos channel by clotrimazole (CLT) and other imidazole antimycotics. We have shown in sickle erythrocytes that CLT blocks K+ transport via the Gardos channel, prevents the change in membrane potential observed when the Gardos channel is activated by internal Ca2+, and inhibits dehydration induced by either the Ca2+ ionophore A 23187 or cyclic oxygenation-deoxygenation. We have also shown in normal erythrocytes that CLT displaces bound 125I-ChTX, indicating a specific interaction with the outside portion of the Ca2+activated K channel. Electrophysiological studies have shown that CLT blocks K+ -currents in several cell types (murine erythroleukemia, PC 12, and ferret portal vein smooth muscle cells), with a mode of action compatible with that of a pore blocker.

Stuart et al. combined CLT with a compound which stabilizes the oxyconformation of Hb S: CLT induced an additive reduction in the rate at which sickle cells clog micropore filters, and may therefore attenuate formation of irreversibly sickled cells.

Studies in a transgenic mouse model for sickle cell disease (SAD mouse) have indicated that CLT inhibits transport and dehydration via the Gardos channel in vitro. Oral administration of CLT (one week at 160 mg/Kg body weight/day or 4 weeks at 80 mg/Kg body weight/day) is associated with marked inhibition of K+ transport via the red cell Gardos channel, increased red cell K+ content, decreased MCHC and density of the red cells. These results in an animal model of sickle cell disease confirm the feasibility of this therapeutic approach of preventing sickle cell dehydration by specifically blocking the Gardos channel.

Studies with normal volunteers taking CLT orally have identified a dosage range (10-20 mg/Kg body weight) which leads to marked inhibition of transport via the red cell Gardos channel . This dosage is substantially lower than that used in the past to treat systemic mycotic infections (60-160 mg/Kg body weight/day). No significant side effects were observed in normal volunteers taking oral CLT for up to 6 days.

We treated 5 subjects who have sickle cell anemia with oral clotrimazole, a specific Gardos channel inhibitor. Patients were started on a dose of 10 mg clotrimazole/kg/day for one week. Protocol design allowed the daily dose to be escalated by 10mg/kg each week until significant changes in erythrocyte density and K+ transport were achieved. Blood was sampled three times a week for hematological and chemical assays, erythrocyte density, cation content and K+ transport. At dosages of 20 mg clotrimazole/kg/day, all subjects showed Gardos channel inhibition, reduced erythrocyte dehydration, increased cell K+ content, and somewhat increased hemoglobin levels. Adverse effects were limited to mild/moderate dysuria in all subjects, and a reversible increase in plasma alanine transaminase and aspartic transaminase levels in two subjects treated with 30 mg clotrimazole/kg/day. This is the first in vivo evidence that the Gardos channel causes dehydration of sickle erythrocytes, and that its pharmacological inhibition provides a realistic anti-sickling strategy. (abstract from Brugnara C, B Gee, C Armsby, S Kurth, M Sakamoto, N Rifai, SL Alper, O. Platt. Therapy with oral clotrimazole induces inhibition of the Gardos channel and reduction of erythrocyte dehydration in patients with sickle cell disease. J. Clin. Invest. 1996; 97: 1227-1234.)

Beneficial effects of increasing red cell Mg2+ in sickle erythrocytes
There are at least two theoretical justification for attempting to increase red cell Mg2+ content in sickle cell disease: a) Direct effect of increasing cell Mg2+ on cell water content: when the cell Mg2+ content is increased, negatively charged Cl- ions enter the cell to maintain electroneutrality. Water also moves into the cell, causing a decrease in cell Hb concentration. b) Modulation by Mg2+ of K-Cl cotransport: increasing internal Mg2+ markedly inhibits K-Cl cotransport. Small changes in cell Mg2+ content from the normal range induce significant changes in K+ flux via K-Cl cotransport.

There is a substantial amount of information available on the mechanisms controlling red cell Mg 2+ content. Cell Mg2+ can be either free in the cytoplasm or bound to ATP and 2,3-DPG . The free concentration of Mg2+ is determined not only by the ratio free/bound, but also by the balance between the inward movement of Mg2+ down an electrochemical gradient and the outward movement of Mg2+ via the Mg/Na exchange system. This Mg2+ transport system has been described in detail in chicken and human erythrocytes.

 Epidemiological studies have indicated that red cell magnesium content is genetically controlled in normal males, with lower levels in HLA-B35 carriers . It is intriguing to postulate a connection to the observation that certain complications of sickle cell disease are more frequent among patients with HLA-B35. One study reported lower levels of red blood cell magnesium in SS individuals compared with AA individuals. A detailed study on Mg2+ content and transport in sickle cells has indicated that dense SS cells have an abnormally low total Mg2+ content, with a reduced buffering capacity for Mg2+ which leads to an increase of cell free Mg2+. Deoxygenation of SS cells induces a net loss of Mg2+ , because of the increased membrane permeability to Mg2+ and the increased free cell Mg2+ of the dense cells.

 Several studies have shown that oral magnesium supplementation can successfully increase erythrocyte magnesium levels. There have been some uncontrolled reports of a beneficial effect of Mg2+ in patients with SS disease. A 7 day course of Mg2+ supplementation did not show any change in red cell survival in 3 patients with SS disease. In transgenic SAD-1 (EMBO J. 1991; 10:3157) and control C576BL/6 mice we investigated the effect of two weeks of diet with either low Mg (6±2 mg/kg body weight/day; n=6) or high Mg (1,000 ± 20 mg/kg body weight/day; n=4), in comparison with a standard diet (Mg: 400 ± 20 mg/kg body weight/day; n=6). High Mg diet increased SAD 1 erythrocyte Mg and K contents, and reduced K-Cl cotransport activity, MCHC, cell density and reticulocyte count. SAD 1 mice treated with low Mg diet showed a significant reduction in erythrocyte Mg and K contents, and increases in K-Cl cotransport, MCHC, cell density, and reticulocyte counts. In SAD mice, Hct and Hb decreased significantly with low Mg diet and increased significantly with high Mg diet. C576BL/6 controls showed significant changes only in erythrocyte Mg and K content, and K-Cl cotransport activities, similar to those observed in SAD mice. Thus, in the SAD mouse, changes in dietary Mg modulate K-Cl cotransport, modify erythrocyte dehydration and ultimately affect Hb levels (from De Franceschi, L., C. Brugnara, and Y. Beuzard. Magnesium intake affects K-Cl cotransport and cell dehydration in transgenic SAD mice: a model for sickle cell disease therapy. Blood 86:299a, 1995).

 Due to the low toxicity and side effects of Mg2+ supplementation, this therapeutic approach should be examined carefully for SS disease. Several promising therapeutic strategies based on inhibition of sickle cell dehydration are now at the clinical evaluation stage. If these studies are successful, it seems likely that future therapies of sickle cell disease will be based on combinations of drugs promoting Hb F synthesis, and inhibiting sickle cell dehydration.

Supported by grants from the National Institutes of Health Heart, Lung and Blood Institute (2-P60-HL15157), Diabetes, and Digestive, and Kidney Diseases Institute (R01-DK50422). by a General Clinical Research Center grant to Children's Hospital (M01 RR 02172) and by a FDA Orphan Products Development grant (FD-R-001022-01). 

Suggested readings:

  1. Alvarez J , Montero M, Garcia-Sancho J: High affinity inhibition of Ca2+- dependent K+ channels by cytochrome P-450 inhibitors. J Biol Chem 1992, 267:11789-11793.
  2. Bookchin RM, Ortiz OE, Lew VL: Evidence for a direct reticulocyte origin of dense red cells in sickle cell anemia. J Clin Invest 1991, 87 :113-124.
  3. Bookchin R , Tiffert JT, Davies SC, Vichinsky E,Lew VL: Magnesium therapy for sickle cell anemia: a new rationale. Proceeeding of "New trends in therapy for hemoglobinopathies and thalassemias", September 19-22, 1994, Paris. IX-6
  4. Brugnara C, De Franceschi L, Alper SL: Inhibition of Ca2+-dependent K+ transport and cell dehydration in sickle erythrocytes by CLT and other imidazole derivatives. J Clin Invest 1993, 92: 520-526.
  5. Brugnara C, Tosteson DC: Inhibition of K transport by divalent cations in sickle erythrocytes. Blood 1987, 70 :1810-1815.
  6. Brugnara C , Armsby CA, Sakamoto M, Rifai N, Alper SL, Platt O: Oral administration of clotrimazole and blockade of human erythrocyte Ca2+ -activated K+ channel: the imidazole ring is not required for inhibitory activity. J. Pharmacol. Exper. Therap. 273: 266-272, 1995.
  7. Brugnara C, B Gee, C Armsby, S Kurth, M Sakamoto, N Rifai, SL Alper, O. Platt: Therapy with oral clotrimazole induces inhibition of the Gardos channel and reduction of erythrocyte dehydration in patients with sickle cell disease. J. Clin. Invest. 1996; 97: 1227-1234.
  8. De Franceschi L, Saadane N, Trudel M, Alper SL, Brugnara C, Beuzard Y: Treatment with oral clotrimazole blocks Ca2+-activated K+ transport and reverses erythrocyte dehydration in transgenic SAD mice. A model for therapy of sickle cell disease. J Clin Invest 1994, 93: 1670-1676.
  9. Eaton WA, Hofrichter, J: Hemoglobin S gelation and sickle cell disease. Blood 1987, 70: 1245-1266.
  10. Hebbel RP, Shalev O, Rachmilewitz EA, Repka T: Oxidative processes in thalassemic and sickle disease pathobiology: potential opportunities for therapeutic intervention. Proceeeding of "New trends in therapy for hemoglobinopathies and thalassemias", September 19-22, 1994, Paris. IX-6
  11. Joiner CH, Jiang M: Dipyridamole inhibits deoxygenation-induced cation fluxes in sickle cells. Proceeeding of "New trends in therapy for hemoglobinopathies and thalassemias", September 19-22, 1994, Paris. IX-10.
  12. Rifai, N., M. Sakamoto, T. Law, O. Platt, M. Mikati, C.C. Armsby, and C. Brugnara. Measurement by HPLC, blood distribution and pharmacokinetics of oral clotrimazole, a new potential anti-sickling agent. Clin. Chem. 41:387-391, 1995.
  13. Stuart J, Mojiminiyi BO, Stone PC, Culliford SJ, Ellory JC: Additive in vitro effects of anti-sickling drugs. Br J Haematol 1994, 86: 820-823.