Potassium (K + ) deficiency is a common and eventually life-treating condition. Hypokalemia is defined as serum K+ level less than 3.5 mM. This chapter, together with cornerstone mechanisms, reviews the newest molecular regulators of the K + homeostasis, including the feedforward signals recently hypothesized in the gut. The renal and extra-renal diseases leading to hypokalemia are reported here. Newest findings in the mechanism underlying the renal inherited K + -losing syndrome, like Gitelman, Bartter and Liddle syndromes are described in details.
Chronic K + deficiency has been described to induce a kaliopenic nephropathy . This term includes all the morphological and functional changes induced by hypokalemia in the kidneys. The hypokalemia-induced renal morphological alterations in both humans and experimental models are depicted. A segment-detailed analysis of the hypokalemia-induced functional and molecular changes on sodium, water and urea handling and on the acid-base homeostasis is reported. These mechanisms provide the molecular basis underlying the development of metabolic alkalosis, increase urinary ammonium excretion and the urine concentrating defect secondary to K + deficiency. Studies based on a system biology approach to the K + deficiency have been included. These could be suggestive of new experimental hypothesis on the K + deficient states. Finally, we reviewed the new findings on the systemic influence of a K + deficiency pointing on the blood pressure regulation and glucose intolerance.
Hypokalemia, Potassium deficiency, glucagon, TTKG, hypokalemic periodic paralysis, Bartter Syndrome, SLC12A1, NKCC2, ROMK, CLCNKA, CLCNKB, BSND, Barttin subunit, CaSR, NCC, Gitelman Syndrome, Liddle Syndrome, ENaC, Kaliopenic Nephropathy, metabolic alkalosis, polyuria, ammonium, hypertension, aldosterone, WNK, glucose intolerance.
Hypokalemia is a common clinical disorder that can be the end-result of: (1) potassium (K + ) redistribution between plasma and intracellular fluid (ICF); (2) insufficient K + intake; (3) disproportionate K + excretion. It is commonly defined as a plasma K + concentration less than 3.5 mmol/L, but this level infrequently causes trouble unless it has fallen rapidly: patients are usually symptomatic when plasma K + is lower than 2.5 mmol/L. Major muscle weakness have a tendency to occur at plasma K + less than 2 mmol/L.
The average K + intake in a typical western diet is roughly 70 mmol. The intestine absorbs almost all of the ingested K + ; only negligible quantities of K + are excreted in the feces. The kidney plays an important role in K + balance, which is the result of glomerular filtration, extensive proximal tubule reabsorption, and highly regulated secretory/reabsorbtive processes located along the distal tubule and the collecting duct (CD). Total body K + is roughly 55 mmol/kg of body weight, with 98% distributed to the intracellular fluid (primarily in muscle, liver, and erythrocytes) and 2% in the extracellular fluid. Na/K-ATPase actively pumps K + into the cell and preserves the electrochemical gradient between the intra- and extracellular pool.
New Concepts on the Integrative Control of K + Homeostasis
Feedback Control of K + Homeostasis
A large increase in plasma K + concentration triggers aldosterone release from the adrenal glands. Aldosterone, in turn, stimulates the activity and synthesis of both Na/K-ATPase and luminal K + channels in CD principal cells, thus promoting K + excretion. In addition, aldosterone enhances K + secretion in the distal colon, which can exert an essential role when renal function is reduced.
On the other hand, if plasma K + concentration decreases as a consequence of reduced K + intake or increased K + excretion, then feedback regulation redistributes K + from ICF to plasma. At the same extent, skeletal muscle becomes insulin-resistant to K + (but not glucose) uptake, blocks the entry of K + from plasma into the cell. Hypokalemia also causes a decreased expression of skeletal muscle Na/K-ATPase 2 isoform, thus allowing a leak of K + from ICF to the plasma. The low plasma K + concentration suppresses adrenal aldosterone release so that the kidney reduces urinary K + excretion.
Feedforward Control of K + Homeostasis
However, besides the classic feedback control, some findings suggest a feedforward control. It is clear that plasma K + stimulates aldosterone secretion only at supra-physiological levels, with little effect within the physiological range. Indeed, it has been shown that, in sheep, a meal intake produced a substantial kaliuresis in the absence of changes in plasma aldosterone concentration. From these experiments it was concluded that the increased renal K + excretion following a meal cannot be explained by changes in aldosterone concentration, but it may be dependent on the existence of a kaliuretic reflex arising from sensors in the splanchnic bed (i.e., gut, portal circulation, and/or liver) ( Figure 50.1 ).
One of the potential effectors of the feedforward control of serum K + is glucagon. Glucagon secretion is definitely stimulated after a protein-rich meal, and intraportal glucagon infusion produces significant increases in renal blood flow and glomerular filtration rate (GFR), suggesting the existence of a hepatorenal axis.
The feedforward regulation may act through three different mechanisms: (1) insulin release rapidly stimulates cellular K + uptake into insulin-responsive tissues; (2) glucagon, through cAMP released from the liver, quickly increases renal K + excretion after a protein-rich meal; (3) a yet-unidentified gut factor senses K + ingestion and enhances renal K + excretion. When plasma K + level increases despite these layers of control, feedback regulation is activated. Aldosterone acts only after a certain time, it is not involved in rapid control of K + homeostasis, but it can chronically increase K + secretion until plasma K + is normalized.
Assessment of Urinary K + Excretion
Several urine parameters are used to identify whether hypokalemia is dependent on renal loss. Renal K + excretion can be assessed with a 24 hour urine collection or a spot urine test determining the K + : creatinine ratio. A 24-hour urinary K + excretion lower than 15 mEq or a K + (mmol)/creatinine (mmol) ratio <1 suggests an extrarenal cause of hypokalemia.
In the clinical practise, as an initial test to address the origin of K + losses, a random urine K + is used. However, this approach is hampered by the effect of renal water handling on urine K + concentration. Determining the transtubular K + gradient (TTKG) is still an accepted way to assess renal K + handling:
TTKG = [ uk / ( u Osmolality / s Osmolality ) ] / s k
Tubular fluid K + concentration in the last part of the CD is mainly dependent on aldosterone, because most K + secretion takes place in the CD. Thereafter, urinary K + concentration increases as a consequence of water reabsorption. The TTKG reflects the tubular fluid K + concentration at the end of the cortical CD, by accounting for water reabsorption that takes place distal of where K + secretion has ended.
However, there are few limitations to the clinical use of this formula. First, the calculation assumes that there is no significant solute transport and only water reabsorption along the medullary CD. Any Na + or urea reabsorption in this segment would tend to lower urine osmolality and cause the TTKG to overestimate the gradient for K + secretion in the upstream CD. Second, there must be optimal conditions for K + secretion at the time that the TTKG is measured. In this regard, urinary Na + should be no less than 25 mEq/L, to guarantee that Na + delivery to the CD is not rate-limiting in K + secretion. In addition, urine osmolality should be equal to or higher than the plasma. A higher urine osmolality indicates increased vasopressin release, which is known to stimulate K + secretion in the CD.
Hypokalemia Associated With Intracellular Shift
The regulation of K + distribution between the intracellular and extracellular space is known as internal K + balance. Even though the kidney is in charge of the preservation of total body K + , factors that adjust internal balance are central to the removal of acute K + -loads. A large K + intake could potentially double extracellular K + concentration in the absence of a rapid shift into the cells. This process is mainly regulated by insulin and catecholamines, with a minor role of metabolic and respiratory alkalosis.
Hypokalemic periodic paralysis is a rare disorder characterized by muscle weakness or paralysis as a result of the sudden movement of K + into cells. Measurement of the TTKG at the time of the attacks typically shows values of <1. The attacks may be triggered by exercise, stress, intake of large quantities of carbohydrates, and increased release of catecholamines or insulin.
This disorder is classified as primary, due to a genetic defect or acquired, due to drugs or glandular diseases. The genetic forms are associated with mutations in genes encoding for subunits of muscular sodium, calcium, and potassium channels. Mutations in the α-subunit of the calcium channel [dihydropyridine (DHP)-receptor] (CACNA1S) gene and the α-subunit of the sodium channel (SCN4A) have been described. Loss of function mutations of CACNA1S reduce current density. A mutation in the KCNJ2 gene encoding for the inward-rectifying potassium channel Kir2.1 causes Andersen-Tawil syndrome, characterized by the triad of periodic muscle weakness, cardiac arrhythmias, and multiple dysmorphic features (short stature, hypertelorism, micrognathia). Whether hypokalemia determines the attack is not well-established. The onset of attacks occurs generally between 15 and 35 years of age; the severity of the clinical manifestations range from rare episodes in a lifetime to daily and severe attacks. The attacks can be triggered by all conditions which favor hypokalemia, such as physical exercise, a carbohydrate-rich meal, alcohol, and cold. Myalgia after the attack is a frequent complaint. The acquired form is mainly associated with thyrotoxicosis. Excess thyroid hormone may predispose to paralytic episodes by increasing Na/K-ATPase activity. The activity of this pump is further induced by catecholamines, which are typically increased in this setting. The underlying cause of thyrotoxicosis is most commonly Graves disease, but it can also be a solitary thyroid adenoma (Plummer disease). The acute attacks of hypokalemic periodic paralysis are best treated with intravenous KCl and propranolol.
It is important to administer KCl in non-dextrose-containing solutions, because glucose will stimulate insulin release, potentially exacerbating K + shift into the cells. Propranolol (a nonspecific adrenergic β-blocker) blocks the effects of catecholamines, and inhibits the peripheral conversion of T4 to T3.
Extrarenal K + Loss from the Body
Diarrhea is a common cause of hypokalemia due to gastrointestinal loss. Secretory diarrhea may be the consequence of two processes that can occur either alone or together. First, it may be related to inhibition of active intestinal NaCl and NaHCO 3 reabsorption and, second, it may be dependent on increased active secretion of Cl − coupled to passive secretion of an identical quantity of Na + in order to maintain the electrochemical balance. Under both circumstances, the stool electrolyte composition is analogous to plasma with a high concentration of NaCl and a much lower K + concentration. Despite the low K + concentration in the stool, large K + losses can take place in the setting of large fecal fluid volume.
Hypokalemia may also be associated with infectious diarrhea. In particular, malaria and leptospirosis may cause alterations in fluid and electrolyte balance. Hypokalemia is particularly frequent in children with severe malaria, and may arise within several hours of initiation of therapy. Hypokalemia develops in about one-third of patients with leptospirosis. Such patients are at risk of both gastrointestinal and renal losses. In the outer membrane of the organism there is a substance that has an inhibitory effect on the Na/K-ATPase within the nephron. It has been hypothesized that this inhibitory effect reduces Na + reabsorption along the proximal tubule, thus increasing distal Na + delivery, resulting in kaliuresis. Hypokalemia may also be associated with watery diarrhea and achlorhydria, a condition secondary to hypersecretion of vasoactive intestinal polypeptide (VIP). In adults, this syndrome is most commonly a complication of pancreatic islet cell tumors, and sometimes of bronchogenic carcinoma, medullary thyroid carcinoma or retroperitoneal histiocytoma. There are now few reports describing chronic watery diarrhea and hypokalemia due to adrenal pheochromocytoma containing immunoreactive VIP.
A recent report describes five consecutive patients with acute or subacute colonic pseudo-obstruction suffering a typical secretory diarrhea characterized by very high fecal K + concentrations (over 100 mEq/kg) and low Na + concentration (Ogilvie syndrome). These elevated fecal concentrations of K + in large volume diarrhea induced important outputs of K + salts responsible for profound hypokalemia and decreased urinary excretion of K + .
Renal K + -Losing Syndromes
Bartter Syndrome (BS)
BS results from a defect in any of the major components of NaCl reabsorbtive machinery along the TAL. So far, mutations of five genes have been described. The defect determines renal loss of water and electrolytes resulting in hypovolemia, with a compensatory increase in renin and aldosterone levels.
Genetic and Molecular Biology
BS type I is sustained by mutations of the SLC12A1 gene, encoding the kidney-specific furosemide-sensitive NKCC2. A number of point mutations have been described in homozygosis or compound heterozygosis, mostly frameshift and non-conservative missense mutations. To date, over 30 mutations in the SLC12A1 gene have been reported ; phenotypic variability among patients with SLC12A1 mutations may be due to the effect of genetic mutations on protein function, and milder phenotypes may correlate with residual NKCC2 function.
BS type II depends on inactivating mutations of the KCNJ1 gene, encoding the K + channel ROMK. These channels are the main renal K + secretory channels. Along the TAL, ROMK mediates K + efflux to the lumen, which is critical for supporting Na/K-2Cl absorption via NKCC2. At this level of the nephron, ROMK channels contribute to the generation of the lumen-positive transepithelial voltage which allows paracellular calcium and magnesium absorption. An inactivating mutation of ROMK is thought to inhibit salt reabsorption along the TAL. Over 35 genetic mutations have been described, such as missense mutations, frameshift mutations, and stop codons which result in a truncated protein. The majority of these mutations reduces or eliminates ROMK surface expression, as a consequence of misfolding and/or mistrafficking; others compromise K + permeation and channel regulation. Besides the TAL, ROMK channels are also expressed along the apical membrane of principal cells in the cortical CD, where they mediate K + secretion into the lumen. A defect in ROMK results in the classical BS phenotype, including the presence of hypokalemia. This finding brought attention to its role in K + secretion. Subsequent studies have demonstrated that, in the absence of functional ROMK channels, K + secretion is guaranteed by the upregulation of flux-sensitive Maxi-K channels along the CD in mice.
BS type III depends on the mutation of the kidney-specific Cl − channel, CLC-K. Two genes belonging to the CLC family are involved in Cl − efflux across the basolateral membrane, CLCNKA and CLCNKB. Their products are nearly identical at protein level, and both channels are associated with the Barttin subunit, essential for their insertion on plasma membrane and their activity. These channels differ only in their distribution along the nephron, with CLC-Ka expressed predominantly along the TAL, while CLC-Kb is expressed along the DT. CLCKNB defects are associated with a Bartter phenotype in humans; a high rate of deletions encompassing the entire gene has been described, together with frameshift and splice-site mutations. These mutations are supposed to disrupt the protein, altering its function. The predominant location of CLCKb along the DT explains why this variant of BS is less commonly associated with a defect in concentrating mechanism, and with hypercalciuria. There is no evidence CLCKNA mutations may generate a Bartter like syndrome. ClCk1 (the ortholog of CLC-Ka)-deficient mice show a phenotype of nephrogenic diabetes insipidus. However a combination of defects in both CLCNKA and CLCNKB genes result in a phenotype of antenatal BS.
BS type IV refers to the mutations in the BSND gene product. In contrast to other BS variants, the gene does not encode for an ion channel or transporter, but for an accessory subunit of CLC-Ka and CLC-Kb, defined as Barttin . CLC-K/Barttin Cl − channels also localize in the cochlea, along the basolateral membrane of marginal cells of the stria. Barttin has been found mutated in patients suffering from BS; different mutations generate phenotypes of varying severity. In heterologous expression, CLC-K channels do not yield currents in the absence of a functional Barttin subunit, suggesting that Barttin is essential for their function. As in the TAL, CLC-K channels participate in Cl − reabsorption in the inner ear. Recent studies have shown that the Barttin subunit is essential for the generation of endocochlear potential; in the absence of Barttin, the degeneration of cochlear outer cells and the collapse of endolymphatic space may contribute to the pathogenesis of deafness in this BS subtype.
BS type V depends on the activating mutation of the calcium-sensing receptor (CaSR). The protein is expressed in the parathyroid and in the kidney, and it is mainly involved in calcium and magnesium homeostasis. Along the TAL the CaSR is expressed on the basolateral membrane and it can inhibit salt absorption. A case report showed that activating mutations of the CaSR gene associated with a BS phenotype inhibit ROMK, explaining the inhibition of salt absorption at this site. A number of gain-of-function mutations of the CaSR gene have been identified as causing an inherited form of hypocalcemia/hypoparathyroidism (autosomal-dominant hypoparathyroidism). Whether those disarrangements lead to a different phenotype is still unknown.
Mutations in these five genes do not explain all cases of BS, and many patients do not get a genetic diagnosis. It is presumable that other genes are involved in the pathogenesis of this syndrome. Recently a role has been proposed for claudins, a family of transmembrane proteins expressed within the tight junction. Mutation in claudin-16 is associated with familiar hypomagnesemia with hypercalciuria and nephrocalcinosis. A single nucleotide substitution has been found in the extracellular domain of claudin-8 in four African-American BS patients. The effect of the mutation on protein function has not yet been addressed.
The defective NaCl absorption along the TAL caused by mutations in any of these five genes leads to an increased salt delivery to the distal nephron. The subsequent volume-depletion leads to a compensatory hyper-reninemia. Mice models of BS type I and II show early onset of polyuria, metabolic alkalosis, increased calcium and magnesium urinary excretion, and hyper-prostaglandinemia, a phenotype which resembles the abuse of loop diuretics. The activation of RAAS leads to a compensatory increase in Na + absorption along the PT and the aldosterone-sensitive segments of the nephron. The latter favors K + secretion along the CD, enhancing the kaliuretic effect of the increased Na + delivery. Clinical differences among BS subtypes depend on the specific physiological role of the causative gene in the kidney and in other organs.
Type II BS is characterized by relatively mild hypokalemia compared with type I, and by the dual role of ROMK in the kidney in controlling NaCl absorption along the TAL (through K + recycling) and K + secretion along the CD. The presence of hypokalemia is ensured by the activation of the flux-sensitive Maxi-K channels which mediate urinary K + secretion along the distal nephron. However, newborn infants suffering from type II BS show transient hyperkalemia before developing normohypokalemia later in the infancy. This effect may be due to the delay in BK-dependent K + secretion, which later is responsible for urinary K + excretion.
The widespread distribution of CLC-K channels along the distal nephron and the compensatory activation of Cl − absorption through other channels explains why CLCKNB mutations may result in a pure BS phenotype, GS phenotype or a combination of these. Hypercalciuria and nephrocalcinosis are typical signs of type I and II BS, but are rare in type III ; however, a broad spectrum of phenotypes has been associated with mutations of the CLCNKB gene, ranging from antenatal BS to classic BS and Gitelman-like syndrome, without any correlation with the type of genetic mutation. Additional studies are needed for a better understanding of the phenotypic variability.
The presence of deafness is a hallmark of type IV BS. Barttin, as pointed out above, is necessary for CLC-K channels trafficking to the membrane. ClCK-barttin complex is expressed in the kidney and in the inner ear. Mice lacking a functional CLC-Ka have a phenotype resembling nephrogenic diabetes insipidus, with high vasopressin plasma levels, and low osmolality of renal papilla even after water restriction, suggesting a role in the urinary concentrating mechanism. CLCKB-null mice show the classic form of BS, whereas only a defect of Barttin determines deafness. In the inner ear CLC-K/Barttin channels participate in Cl − transcellular extrusion across the basolateral membrane. It is possible that the absence of CLC-Kb could be compensated by the CLC-Ka-Barttin in the inner ear, while the absence of Barttin equals a double defect in CLC-k a and b, leading to deafness. Hypercalciuria and nephrocalcinosis are the main features of type V BS. Activating mutation of the CaSR leads to autosomal dominant hypoparathyroidism, characterized by hypocalcaemia and hyperphosphoremia, with low-normal PTH levels.
Clinically, BS is divided into antenatal and classic BS with or without deafness.
Antenatal BS , or hyper-prostaglandin E BS , is the most severe form, characterized by polyhydramnios for excessive urinary output and premature birth. It is sustained by type I and II, and sometimes type III. After birth, patients have a life-threatening clinical course, with fever, vomiting, and lethargy. Biochemical analysis shows the presence of metabolic alkalosis, hypokalemia, isohypostenuria, and hypercalciuria. Nephrocalcinosis is frequent. High urinary prostaglandin excretion of E2 or its metabolites is typical of the antenatal form, and high levels of renin and aldosterone are secondary to volume-depletion. The reason for the high urinary and plasma prostaglandin levels is still unknown, but it seems to be secondary to the defect of NaCl absorption along the TAL.
Classic BS is sustained more often by type III BS. Clinical appearance occurs during infancy or childhood, in the absence of polyhydramnios and prematurity. The clinical course is milder than the antenatal subtype; patients manifest polyuria, polydipsia, vomiting, and dehydration. Nephrocalcinosis is an infrequent sign, and a less severe defect in urinary concentrating mechanism is present.
BS with sensorineural deafness. The presentation of patients with type IV BS shows remarkable variation, ranging from prenatal diagnosis with severe polyhydramnios and prematurity to late diagnosis.
Gitelman syndrome (GS) differs from BS because of the presence of hypocalciuria and hypomagnesaemia. It is often diagnosed in adulthood.
Genetics and Molecular Biology
The syndrome correlates with mutations of the SLC12A3 gene located on chromosome 16q, encoding the thiazide-sensitive sodium-chloride co-transporter (NCC). The transporter is expressed on the apical membrane of distal tubule, and represents the major NaCl transport pathway in this segment. More than 140 mutations have been described; the majority of mutations are missense substitutions, but frameshift and splice-site mutations have also been described. Heterozygous subjects show a tendency for low blood pressure, while the complete GS phenotype occurs only in homozygosis. De Jong et al. have performed, in xenopus laevis oocytes, functional and immunohistochemical analysis of mutant human NCC of GS subjects. This study has found class I mutants, characterized by the absence of significant metazolone-sensitive Na + uptake with undetectable protein distribution on the membrane, and class II mutants, which exhibited significant, albeit low, metazolone-sensitive Na + uptake, while NCC staining was equally present in plasma membrane and cytoplasm. These findings suggest that some mutations compromise NCC abundance in plasma membrane (class I), leading to a defect in protein activity; other mutations only partly impair NCC routing to the membrane, as suggested by the presence of mutant NCC both on plasma membrane and cytoplasm. However, different mechanisms are involved in the impaired trafficking for the two classes of mutations, and the precise mechanism has still to be established. Previous studies suggest the role of defective post-translational changes, such as protein glycosylation, which seems to be required for proper folding and trafficking to plasma membrane. A minority of patients with GS phenotype show mutations in the CLCNKB gene, which is also responsible for BS type III.
Both NCC and CLC-Kb dysfunction result in decreased Na + and Cl − absorption along the DT. The volume-contraction resulting from defective NaCl absorption determines a compensatory activation of RAAS, which promotes electrogenic Na + absorption along the CD through ENaC. The latter enhances K + and H + secretion along the CD, favoring hypokalemia and metabolic alkalosis. The pathogenesis of hypocalciuria and hypomagnesemia refers to a not yet completely-elucidated mechanism. Micropuncture experiments have demonstrated an increased Ca 2+ absorption along the proximal tubule (PT) after chronic administration of thiazides, whereas DT calcium absorption was unaffected. This hypothesis is supported by the presence of thiazide-induced hypocalciuria in a mouse model lacking the calcium channel (TRPV5) along the DT. These findings demonstrate that increased calcium absorption parallels a compensatory increased Na + absorption along the PT secondary to volume-contraction. Other studies suggest that enhanced calcium absorption along the DT participates in the generation of hypocalciuria. In a mouse model of GS, the expression of TRPV5 and TRPV6 were increased, and TRPV5 is also overexpressed in renal tissue from patients with GS.
Hypomagnesemia, another hallmark which distinguishes GS from BS, has a controversial origin. Magnesium is freely filtered by the glomerulus, and it is reabsorbed in a small fraction along the PT. The majority of Mg 2+ is reabsorbed along the TAL, via paracellular pathway, and DT, via transcellular pathway. In the latter, Mg 2+ reabsorption is mediated by the transient receptor potential cation channel, TRPM6. In NCC knockout mice a downregulation of TRPM6 in DT has been shown. This effect could explain the defective Mg 2+ absorption, and the subsequent hypomagnesemia.
The clinical phenotype in CLCNKB mutations is extremely variable; several reports have described subjects with phenotypic features of GS without any defect in SLC12A3 gene, carrying homozygous mutations of CLCNKB gene or mixed BS-GS phenotype.
GS is characterized by an extreme inter- and intra-familial phenotype variability, varying from mild or undiagnosed forms to severe conditions complicated by growth retardation, ventricular arrhythmias, and neuromuscular symptoms. In most cases the diagnosis occurs in adulthood. The patients suffer from tetany, especially during conditions which determine further Mg 2+ loss, like vomiting or diarrhea. Some patients experience fatigue which compromises daily activities, in relation to the degree of hypokalemia. In contrast with BS, those patients do not manifest polyuria and growth retardation. Chronic K + and Mg 2+ deficiency may predispose to a higher risk for ventricular arrhythmias. However, lethal arrhythmias have been reported rarely in GS patients, and may be related to underlying triggering mechanisms besides hypokalemia. Riviera et al. have described a subgroup of GS patients with severe phenotype, characterized by early onset, and severe neuromuscular and cardiac symptoms. Almost all patients of the subgroup were male, and showed a higher incidence of splicing mutation leading to a truncated transcript compared with mild and classic GS. This study suggests that male gender and splicing mutations, resulting in a severe protein dysfunction, may account for the clinical severity of GS. Biochemical analysis shows hypocalciuria, hypokalemia, and hypomagnesemia. Although hypocalciuria and hypomagnesemia have been considered necessary for the diagnosis of GS, recently a report of a GS patient with a proven mutation in the GS gene did not manifest those signs. Plasma renin and aldosterone levels are only slightly increased compared with BS.
Liddle syndrome (LS) is an autosomal dominant disease leading to hypertension. Besides the early onset of hypertension, it is characterized by hypokalemic metabolic alkalosis, with downregulation of RAAS. In the last 10 years the Lifton group has identified several mutations mapping in β- and γ-ENaC subunits, resulting in the expression of a higher number of channels in the plasma membrane.
Genetic and Molecular Biology
ENaC is comprised of three subunits, α, β, and γ, which are assembled at stoichiometry α2βγ. Several studies have suggested that also αβ and αγ are functional channels. The first description of Liddle syndrome (LS) in literature was correlated with the mutation in the gene encoding the β-ENaC subunit, resulting in a premature stop codon which leads to the loss of the last amino acids of the C-terminus. Genetic screening of subjects suffering from LS has showed that the syndrome also segregates with mutations of the gene encoding for the γ-subunit. The expression of either truncated γ- or β-ENaC subunit increases amiloride-sensitive Na + uptake in oocytes. These findings suggest that the C-terminus of β- and γ-subunits is a crucial region for the activity of the channel. Different investigations have found that most mutations resulting in LS phenotype are mapped on a highly-conserved PPPXY sequence (PPY motif) of the C-terminus in β and γ; disruption of the motif preserves the channel from inhibitory stimuli.
The amiloride-sensitive ENaC is an ion channel which mediates Na + absorption along the distal segment of CNT and CD. ENaC channels are positively regulated by aldosterone and vasopressin. In mice, aldosterone has been shown to activate ENaC via the ser/thr kinase Sgk1. This kinase induces ENaC activation partly through the inhibition of ENaC downregulation by the E3 ubiquitine ligase, Nedd4-2, which targets ENaC to degradation in cultured cells. Sgk1 phosphorylates Nedd4-2 to an inhibitory site, thus preventing ENaC removal from the plasma membrane. Mutations in β- and γ-subunits of ENaC in LS result in a constitutive activation of ENaC activity. These mutations impair the PY motif in the C-terminus, which is required for Nedd4-2-mediated ENaC ubiquitination. In the absence of the binding site, Nedd4-2 fails to target the channel for degradation. However, these mutations increase the overall ENaC expression at the apical membrane, contributing to increased Na + absorption. The resulting volume-expansion explains the suppression of RAAS. Interestingly, mice bearing a LS deletion revealed normal aldosterone response. Consistently, in cultured cell, aldosterone, vasopressin, and Sgk1 are still able to increase ENaC surface abundance in the presence of LS causative mutations, suggesting that Nedd4-2 is not necessary for aldosterone-dependent increased ENaC activity. However, in LS patients, constitutive increased ENaC activity leads to volume-expansion, high blood pressure, and suppressed RAAS. As a consequence of increased channel activity along the CD, patients do manifest hypokalemia and metabolic alkalosis. Mice bearing LS become hypertensive only when fed a high-salt diet.
LS was first described in 1963 as a condition resembling hyperaldosteronism, because of the coexistence of hypertension, hypokalemia, and metabolic alkalosis. However, due to the low/normal levels of plasma aldosterone this condition is described as pseudohyperaldosteronism of the second type. Hypertension usually develops at an early age in affected individuals, and worsens throughout the lifetime. Recently, Tapolyia et al. have proposed the presence of serum bicarbonate over 25 mmol/L, K + levels lower than 4 mEq/L, plasma renin activity lower than 0.35 μU/ml/h, and plasma aldosterone levels lower than 15 ng/dl as screening criteria for LS. Based on these criteria, this group found a prevalence of 6.7% patients that satisfied the criteria for likely LS in a cross-sectional investigation of 149 hypertensive patients with hypokalemia and metabolic alkalosis. Hypertension is usually refractory to common antihypertensive drugs, while it is responsive to the use of the ENaC inhibitor, amiloride.
Renal Changes in K + Deficiency
Hypokalemic states are associated with several renal morphological alterations. The severity and exposure time to hypokalemia are fundamental to the development of renal injuries. Mild hypokalemia ranging among 3.6±0.2 is not associated with significant glomerular or tubular abnormalities in Sprague-Dawley rats. Renal hypertrophy is a universal finding in studies of K + -depletion (KD) since the early report by Schraeder et al. in 1937. This renal growth has distinct morphologic characteristics. The increase in kidney mass is not uniform, being more prominent in the outer medulla, as reflected in the relative contribution of the different zones to total kidney weight in K + deficient rats.
The growth process consists of both hyperplasic and hypertrophic components. Despite the fact that hypokalemic mice present a different metabolic behavior (metabolic acidosis), they develop morphological abnormalities of CD similar to rats and humans. K + deficient mice show morphological alterations in outer medulla CD (OMCD) after three days of a low-K + diet. CD epithelial cells from K + deficient mice have a higher proliferation rate than control mice. Intercalated cells show PCNA positive staining (proliferation marker) more than principal cells. No signs of cellular hypertrophy are detected at this time point in CD epithelial cells, despite IC already display an enlargement of the apical membrane domain. After 14 days of a low-K + diet both principal and IC cells appear hypertrophic as evaluated at EM level. In hypokalemia-induced morphologic alteration of OMCD, hyperplasia precedes the development of cellular hypertrophy. Transcriptome analysis of OMCD suggests Gdf15 (growth differentiation factor 15) increases significantly from day 3 to day 14 of the low-K + diet. It could be a potential candidate in driving the switch from hyperplasia to hypertrophy. Gdf15 belongs to the TGFβ superfamily, and it could be a growth-stop signal in OMCD.
The growth-promoting effect of K + deficiency is not limited by the state of the organ. It occurs in intact, as well as in previously damaged, kidneys. Peterson et al . have shown that institution of K + deficiency in a model of remnant kidney (5/6 nephrectomy) led to an increase in renal mass and RNA content beyond that expected of the surgical ablation alone.
CD is the main nephron site where morphologic alterations occur in K + deficiency. Glomerular and vascular lesions have rarely been observed in pure K + deficiency, and changes in proximal convoluted tubules have been limited to vacuolar degeneration. While several studies demonstrate the occurrence of morphologic changes in individual CD cells, conflicting statements have been made regarding the proportion of cell types and number. Hansen et al . observed extensive swelling of the epithelium of the CD involving both principal and intercalated cells along the OMCD ( Figure 50.2 ). No changes were observed in either principal or intercalated cells of the initial CD. Intercalated cells in the MCD segment develop extensive microplicae over the entire luminal surface with increased luminal surface boundary length, whereas no change was observed in basolateral membrane length or in the luminal or basolateral aspects of principal cells ( Figure 50.3 ). This increased luminal surface could be due to the fusion of the cytoplasmic vesicles with the apical membrane domain. A large number of rod-shaped structures are opened in the lumen in this way. This event is thought to be useful for H + secretion and K + reabsorption through H + -ATPase or H + -K + -ATPase activity, respectively. There are conflicting findings about the proportion of intercalated and principal cells in any part of the CD in K + deficiency. Evidences indicate the development of morphologic changes in individual cells as well as an increase in cell number.
Recently, progressive capillary loss has been identified in hypokalemic nephropathy. This injury was first observed in the ISOM after 2 weeks of a K-restriction diet, expanded to the OSOM at 4 weeks, and then to the cortex by week 12. Capillary loss significantly correlated with local macrophage infiltration and low endothelial cell proliferation rate, an effect probably secondary to a decrease of VEGF and eNOS expression.
The most remarkable ultrastructural change is the accumulation of cytoplasmic droplets in tubular cells of the medulla. The appearance of droplets starts from the CD at the tip of the papilla, and then extends upward into the outer medulla until the cortico-medullary junction. The extension depends on the duration of K + deficiency. Besides epithelial cells, interstitial and other cells in the medulla also show cytoplasmic droplets, with considerable enlargement of cell volume. With K + repletion, the droplets reversed progressively. The droplets are believed to be the consequence of phospholipids dysregulation, and their lysosomal origin is suggested by the presence of hydrolytic enzymes. It has been suggested that increased ammonia production secondary to hypokalemia may contribute to vacuolation of nucleated cells in KD. In fact, ammonia has been shown to induce vacuolization of lysosomes, inhibition of endocytosis, and lysosomial protein degradation. Another hypothesis is that the cytoplasmic droplets could be related to a cellular autophagy phenomenon. Lipid droplets share common protein with the autophagosome in hepatocytes and cardiac myocytes, and those organelles may be an expression of autophagy. Ureteral ligation in K + deficient rats is followed by resolution of droplets, presumably because of increased renal medullary K + content. The severity of droplet formation depends on the method of induction of K + deficiency, with minimal droplet formation developing with DOCA-induced compared with dietary deficiency.
Tubular and interstitial apoptosis is observed during K + deficiency. Apoptotic cells are located mainly in the outer medulla. Bcl-2 protein distributed in the tubules of the outer medulla is significantly decreased in KD rats, while immunoreactivity for Bax protein tends to increase above control levels. These results suggest that apoptosis is associated with progression of cellular proliferation in hypokalemic nephropathy, and a decrease in bcl2 may be involved in promoting this apoptotic process.
Unlike experimental studies where K + deficiency can be induced selectively, clinical observations are frequently based on complex conditions where K + deficiency is complicated by numerous other alterations, including Na + and acid–base homeostasis. The issue is even more complex if we consider additional factors like duration of disease, and therapeutic interventions that may substantially alter histopathologic manifestations in humans. Finally, clinical data are usually limited to cortical biopsies, and sampling of medullary and papillary structures may be lacking. An important observation comes from autopsy studies in healthy Thai adults who died of vehicular accidents. K + deficiency prevails among the healthy population of northeast Thailand. In this study, none of the patients had renal histopathological change compatible with a diagnosis of focal or diffuse interstitial nephritis, and there were minimal renal tubulo-interstitial changes.
Glomerular changes consisted mostly of a reduction in capillary surface and Bowman capsule areas, and an increase in the mesangial space. The glomerular shrinkage was not associated with any alterations in cellular constitution of the tuft. Of note is the increase in juxtaglomerular complex size in many of the patients to levels comparable to, or even exceeding, those observed in BS. Such findings underscore the non-specificity of this change, which is likely secondary to defective Cl − absorption in the thick ascending limb, rather than a characteristic feature of BS. Tubular vacuolization was not a frequent or pathognomic observation, although it was considered characteristic by Conn and Johnson, who coined the term kaliopenic nephropathy in 1956. More important for the prognosis of these patients is the occurrence of degenerative changes such as tubular atrophy, dilatation, epithelial flattening, and thickening of the basement membrane. Again, these alterations are non-specific, and reflect evidence of chronic injury. Increased interstitial surface area and lymphocytic cellular infiltration were frequent. PAS-positive granules have been observed in some of these patients in all medullary cells when available in the specimen.
Taken together, these observations suggest that many of the renal changes seen in K + deficient animals have their counterparts in clinical cases. Some of these changes (glomerular shrinkage, tubular atrophy, interstitial fibrosis) indicate irreversible renal damage, and the occurrence of renal insufficiency in this group of patients has been convincingly demonstrated. Indeed, in some young women the condition has led to end-stage renal disease. Elements of the kaliopenic nephropathy mixed to morphological changes secondary to hypertension are also found in patients with primary aldosteronism. The apparently additive insults of hypertension and K + deficiency are illustrated in the study of 18 patients with primary aldosteronism by Danforth et al . Moderate to severe hypertensive changes (fibrous thickening of small vessels and glomerular hyalinization) and kaliopenic lesions (vacuolization and degeneration of tubular epithelia) were observed in 14 (78%) and 16 (89%) patients, respectively. The hypertensive changes appeared to be more severe in this group than in subjects with primary hypertension of similar severity, suggesting a possible synergistic effect of K + deficiency.
A heterogeneous pattern of structural and functional changes occurs along the nephron, with a progressive increase from proximal to distal segments.
Proximal Convoluted Tubule
Transport along the proximal tubule has been extensively studied in both acute and chronic K + deficiency models with micropuncture and isolated tubule techniques. An acute change in K + concentration in the perfusion bath below 2.5 mM inhibits NaCl transport in isolated and perfused PT from rabbit. Decrements in NaCl reabsorption, as evidenced by decreased net fluid absorption in this segment, have also been observed during capillary microperfusion with hypokalemic solutions in rats. Studies in K + deficient animals show a different pattern of alterations in transport. Walter et al . performed micropuncture studies on anesthetized rats which had been kept on a K deficient diet for 2 weeks. In these animals total glomerular filtration rate (GFR) and single-nephron filtration rate were significantly lower than controls. Fractional reabsorption by the proximal convoluted tubule was enhanced, and end proximal fluid delivery was markedly reduced. These observations are in line with other studies showing reduced fractional excretion of lithium in K + deficiency, presumably a surrogate measurement for enhanced proximal Na + reabsorption.
Chronic K + deficiency, but not an acute luminal exposure, leads to increased NaCl reabsorption. K + deficiency increased the expression of both the adrenergic receptor alfa 2B , and AT1 (angiotensin II receptors) in rat PT. These changes are coupled with an increase in renin and angiotensin II level, as demonstrated in many species. The higher Na + reabsorption could be in part transcellular, mainly driven by several Na + transporters upregulated in hypokalemia ( vide infra ).
Hypokalemic rats have an increased expression of Na-H exchanger 3 (NHE3) in membrane fractions of renal cortex and outer stripe of the outer medulla (OSOM). Immunohistochemistry confirms that NHE3 labeling is increased in the luminal membrane domain of the PT of hypokalemic rats. However, Wang et al . found no change in the expression of NHE3 mRNA and its cognate protein after 6 and 14 days in rats on a low-K + diet. The mRNA levels for NHE1, NHE2, and NHE4 also remained unchanged at 6 and 14 days of the low-K + diet. These apparent inconsistencies between protein expression and mRNA levels could be due to a posttranslational level of regulation of NHE3, as recently found in hypertensive models. In parallel with Na + reabsorption, NHE3 promotes proton extrusion. This mechanism is functional for net bicarbonate reabsorption through the PT.
Chronic K + deficiency is associated with metabolic alkalosis and an increase in bicarbonate reabsorption, as demonstrated by free-flow micropuncture and microperfusion studies in rats. In contrast, acute exposure of proximal tubules to ow (K + 2 mM) in capillary microperfusion experiments had no effect on bicarbonate reabsorption. The mechanism of enhanced bicarbonate reabsorption is mediated by stimulation of NHE3 and the basolateral Na/3HCO 3 -co-transporter. In OKP cells, low-K medium causes a decrease in intracellular pH, which leads to increased NHE3 activity. Accordingly, NHE3 total protein abundance and immunostaining along the apical membrane of PT and TAL were dramatically increased in hypokalemic rats.
The bicarbonate basolateral exit pathway from the PT cells is mediated by the sodium bicarbonate co-transporter (NBC1). During K + deficiency states both the activity and mRNA expression of NBC1 are increased. It is not clear if the upregulation of NBC1 causes the lowering intracellular pH or whether it is a consequence of the higher bicarbonate reabsorption.
Angiotensin II could promote bicarbonate reabsorption in the PT. In fact, angiotensin II is found to increase NBC1 expression in rat PT, while it is debated whether it also regulates NHE3 function.
Metabolic alkalosis developing during K + deficiency is also sustained by increased bicarbonate generation from the ammoniagenesis pathway in the PT. Chronic hypokalemia is associated with increased ammonium excretion in humans as well. K + deficiency leads to a three-fold increase in ammonia production in the S1, and to a two-fold increase in S2 segments of PT. No changes in ammonia production were observed in the S3 segment (pars recta) or in the thick ascending limb or distal convoluted tubule. The increased ammonia production in S1 occurred in both superficial and juxtamedullary nephrons, with a greater extent in the former. The primary mechanism by which K + deficiency stimulates renal ammonia production is not precisely known. Increased renal ammonia production and excretion, despite the simultaneous presence of metabolic alkalosis, suggest that intracellular factors, rather than extracellular pH, modulate renal ammonia metabolism. Intracellular acidosis that occurs in K + deficiency may initiate the adaptive response in ammoniagenesis. Both mitochondrial (glutamine transaminase activity) and cytosolic (phosphoenolypyruvate carboxykinase, PEPCK) enzymes involved in glutamine catabolism and gluconeogenesis are increased in the hypokalemic state. This pathway induces a net gain of new bicarbonate production. K + repletion leads to prompt decreases in ammonia production. Although mice present a species-different acid–base adaptation to hypokalemia, namely metabolic acidosis, they share a high urinary ammonium excretion with rats. Hypokalemic mice show an increase of glutamine transporter SNAT3. SNAT3, normally located in the S3 segment of the PT, spreads to S1 and S2 segments during K + deficiency, supporting the increase ammoniagenesis in the PT ( Figure 50.4 ).
Chronic K + deficiency is associated with low serum phosphate (Pi) and an increase urinary Pi excretion in rats and mice. Key regulators of renal Pi excretion are the Na dependent co-transporter NaPi IIa and NaPi IIc, expressed almost exclusively in the kidney. Another group of Pi transporters, Pit-1 and Pit-2, have been described. In hypokalemia, a phosphaturic phenotype is associated with a differential regulation of the different NaPi transporters. NaPi IIa protein abundance in brush border membranes (BBM) and protein targeting to the apical membrane are increased, while NaPi IIc and PiT2 are decreased in animals fed with a low-K diet. In addition, NaPi IIc relocates from the apical membrane domain to cytoplasmic vesicles. The downregulation of NaPi IIc seems to sustain the hypokalemia-induced urinary phosphate loss.
Other Proximal Tubule Dysfunction
Chronic K + deficiency causes hypocitraturia. Urinary citrate excretion is mainly a function of citrate absorption along the proximal tubule, a process mediated by the Na-dicarboxilate co-transporter NaDC1, and citrate metabolism. Levi et al . have shown an increased NaDC1 activity on renal cortical BBM from K + deficient rats. Hypokalemia may also participate in proximal tubular dysfunctions observed in children with primary distal renal tubular acidosis. Indeed, in these patients, correction of hypokalemia ameliorates low molecular weight proteinuria, phosphaturia, and generalized aminoaciduria.
Loop of Henle
Eknoyan et al . were the first to suggest that a defect in Na + transport by the thick ascending limb (TAL) could explain the concentrating defect of K + deficiency. Micropuncture and microperfusion studies showed diminished net Cl − reabsorption between the latest proximal and earliest distal segment, as well as increased luminal Cl − concentration at the latter site. These findings were consistent with impaired TAL absorption, and the defect was only partially corrected with indomethacin. Gutsche et al ., using the micro stop-flow technique, have provided evidence for defective Na + transport in the TAL of K + deficient rats. The severity of the defect correlated with the decrease in plasma K + concentration, and was rapidly reversed with acute K + administration. In addition, net NaCl transport is inhibited by reduction in bath K + concentration in isolated perfused TAL.
A more recent study has uncovered the molecular basis of the observed defects in electrolyte reabsorption in the TAL. In rats, NKCC2 mRNA expression in medulla is decreased about 56 and 51% after 6 and 14 days of K + restriction diet respectively. Functional studies in tubular suspensions of medullary TAL from K + deficient rats demonstrated a 45 and 37% decreased NKCC2 activity at 6 and 14 days, respectively. NKCC2 protein abundance of membrane fraction from renal ISOM is downregulated in rats fed for 4 days on a K + restricted diet. Immunohistochemical localization confirms a lower expression of NKCC2 in mTAL. Na/K-ATPase plays a key role in the Na + -dependent transport in this nephron segment. Despite the increase in the number of Na/K-ATPase units, the transport capacity of the Na/K pump, determined by ouabain-sensitive Rb uptake, was reduced in mTAL from K + deficient rats. Inhibition of the Na/K pump was the consequence of a reduced affinity for Na.
Bicarbonate and Ammonium
Basolateral and apical Na-H exchangers (NHEs) in TAL are involved in <SPAN role=presentation tabIndex=0 id=MathJax-Element-2-Frame class=MathJax style="POSITION: relative" data-mathml='NH4+’>NH+4NH+4
NH 4 +
and <SPAN role=presentation tabIndex=0 id=MathJax-Element-3-Frame class=MathJax style="POSITION: relative" data-mathml='HCO3−’>HCO−3HCO−3
HCO 3 −
absorption, respectively. The <SPAN role=presentation tabIndex=0 id=MathJax-Element-4-Frame class=MathJax style="POSITION: relative" data-mathml='NH4+’>NH+4NH+4
NH 4 +
absorption rate in Henle’s loop is increased in K + deficiency, which may be secondary to the increased <SPAN role=presentation tabIndex=0 id=MathJax-Element-5-Frame class=MathJax style="POSITION: relative" data-mathml='NH4+’>NH+4NH+4
NH 4 +
concentration in luminal fluid and/or to an increased <SPAN role=presentation tabIndex=0 id=MathJax-Element-6-Frame class=MathJax style="POSITION: relative" data-mathml='NH4+’>NH+4NH+4
NH 4 +
absorptive capacity of TAL. <SPAN role=presentation tabIndex=0 id=MathJax-Element-7-Frame class=MathJax style="POSITION: relative" data-mathml='HCO3−’>HCO−3HCO−3
HCO 3 −
absorptive capacity in Henle’s loop is unchanged in K + deficiency, despite the presence of metabolic alkalosis. The effects of K + deficiency on the expression of basolateral NHE-1 and the expression of apical NHE3 in TAL have been examined by Laghmani et al . NHE1 protein abundance was similarly increased (approximately 90%) at 2 and 5 weeks of K + deficiency, while NHE1 mRNA amount in TAL cells was increased at 2 weeks, and returned to normal values by 5 weeks. NHE3 protein abundance and mRNA remained unchanged after 2 weeks of K + deficiency. NHE3 mRNA was reduced by approximately 50% at 5 weeks. In K + deficiency, the increased NHE1 expression may support an increased TAL <SPAN role=presentation tabIndex=0 id=MathJax-Element-8-Frame class=MathJax style="POSITION: relative" data-mathml='NH4+’>NH+4NH+4
NH 4 +
absorptive capacity. The lack of change in NHE3 expression, despite the presence of metabolic alkalosis, is in agreement with the unchanged <SPAN role=presentation tabIndex=0 id=MathJax-Element-9-Frame class=MathJax style="POSITION: relative" data-mathml='HCO3−’>HCO−3HCO−3
HCO 3 −
absorptive capacity of Henle’s loop. Recently, an electro-neutral, Na-dependent <SPAN role=presentation tabIndex=0 id=MathJax-Element-10-Frame class=MathJax style="POSITION: relative" data-mathml='HCO3−’>HCO−3HCO−3
HCO 3 −
-co-transporter (NBCn1) has been cloned and localized at the basolateral side of the mTAL. Several results show NBCn1 is fundamental for <SPAN role=presentation tabIndex=0 id=MathJax-Element-11-Frame class=MathJax style="POSITION: relative" data-mathml='NH4+’>NH+4NH+4
NH 4 +
reabsorption in the TAL. <SPAN role=presentation tabIndex=0 id=MathJax-Element-12-Frame class=MathJax style="POSITION: relative" data-mathml='NH4+’>NH+4NH+4
NH 4 +
entering the mTAL cell via the furosemide-sensitive NKCC2 transporter dissociates into H + and NH 3 . The NH 3 leaves the cell via non-ionic diffusion preferentially through the basolateral membrane. The remaining proton may either be transported directly via a basolateral NHE1 and/or could be buffered by basolateral import of <SPAN role=presentation tabIndex=0 id=MathJax-Element-13-Frame class=MathJax style="POSITION: relative" data-mathml='HCO3−’>HCO−3HCO−3
HCO 3 −
through NBCn1. During hypokalemia, NBCn1 protein abundance is strongly upregulated in ISOM. In addition, in vitro perfusion of isolated mTAL shows hypokalaemia induced a three-fold upregulation of Na-coupled <SPAN role=presentation tabIndex=0 id=MathJax-Element-14-Frame class=MathJax style="POSITION: relative" data-mathml='HCO3−’>HCO−3HCO−3
HCO 3 −
influx ( Figure 50.5 ).