CASE REPORT

Three Cases of Gitelman's Syndrome Possibly Caused by Different Mutations in the -Sensitive Na-Cl Kazuhisa Takeuchi, Taro Kato, Yoshihiro Taniyama, Kazuo Tsunoda, Nobuyuki Takahashi, Yukio Ikeda, Ken Omata, Yutaka Imai, Takao Saito, Sadayoshi Ito and Keishi Abe

Three adult Japanese cases of Gitelman's syndromewere characterized by secondary aldosteronism, hypokalemic alkalosis, hypomagnesemia, and hypocalciuria. Twowere revealed to be familial cases. Amutation in the thiazide-sensitive Na-Cl cotransporter gene, which had already been confirmed in one family (Takeuchi et al. J Clin Endocrinol Metab 81: 4496, 1996), was not detected in the other twocases. These observations maypossibly support the previous report (Simon et al. Nature Genet 12: 24, 1996) that Gitelman's syndrome is caused by a variety of mutations in the thiazide-sensitive Na-Cl cotransporter. (Internal Medicine 36: 582-585, 1997) Key words: Bartter' s syndrome, secondary aldosteronism, missense mutation, restriction fragment length polymorphism

Introduction Case Reports "Bartter's syndrome" has recently been divided into two Case1 subsets, true Bartter's syndrome and Gitelman's syndrome (1 , A 45-year-old Japanese female had noticed mild periodic 2). Bartter' s syndrome is characterized by hypokalemic alkalo- and since the age of28. In March1994, she felt sis, and secondary aldosteronism without hypertension, and discomfort in her lower abdomen, and a tumor was felt palpable with blunted pressor responsiveness to angiotensin (Ang) II (3). in the region. Uterine myomawas identified by a gynecologist, True Bartter's syndromealso refers to patients with normal or and she was admitted to her neighboring hospital for resection. hypercalciuria, and typically normal levels. Laboratory examination disclosed her (2.4 mmol/ Gitelman's syndrome (4), however, refers to patients with /), , secondary aldosteronism and normoten- hypocalciuria and hypomagnesemia. Genetic analyses have sion. She had no history of chronic , vomiting, or shownthat Bartter's syndromeis due to mutations in Na-K-2C1 abuse. Bartter's syndromewas suspected, and the cotransporter NKCC2(5) or K+ channel ROMK(6), and that operation was postponed. For the therapy, either 32 Gitelman's syndrome is attributable to mutations in thiazide- mmol/day, non-steroidal anti-inflammatory drug (diclofenac sensitive Na-Cl cotransporter (TSC) (7). Wehave recently 1 00 mg/day), or mineralocorticoid receptor antagonist reported the close association of a mutation in TSCwith ( 50 mg/day) was administered. Hypokalemia familial Gitelman's syndrome (8), which further supports the was however not completely corrected. In February 1995, the report by Simon et al (7) that a mutation in TSCgene (probably patient was referred to our department in TohokuUniversity leading to inhibition of the transporter activity) causes Gitelman ' s Hospital for further investigation of hypokalemia and resection syndrome. Here we report three cases ofGitelman's syndrome, of uterine myoma. and suggest apossible variety of mutations in TSCin Gitelman' s Her parents and grandparents are consanguineous. Individu- syndrome. als in her family had a tendency towards hypotension. Her blood pressure was 106/70 mmHg:she was 155 cm tall and weighed 65 kg. Laboratory data are listed in Table 1. Pressor response to

From the Second Department of Internal Medicine, Tohoku University School of Medicine, Sendai Received for publication January 13, 1997; Accepted for publication April 25, 1997 Reprint requests should be addressed to Dr. Kazuhisa Takeuchi, the Second Department of Internal Medicine, TohokuUniversity School of Medicine, 1-1 Seiryo-machi, Aoba-ku, Sendai 980-77

582 Internal Medicine Vol. 36, No. 8 (August 1997) Gitelman' s Syndromeand Na-Cl Cotransporter Table 1. Representative Laboratory Data in Three Cases of Gitelman's Syndrome C a se 1 C a s e 2 C a s e 3 N o r m a l ra n g e

Se ru m e le c tr ol y te l ev e ls N a 1 4 3 1 4 0 1 3 5 1 3 5 - 1 4 7 m m o l K 2 .6 2 .7 3.4 3 . 4 - 5 . 0 m m o l C l 1 0 1 1 0 5 9 6 9 8 - 1 0 8 m m o l C a 8 .9 9 . 2 9. 1 8 . 6 - 1 0. 5 m g /d l M g 1. 8 1. 1 1.3 1 . 8 -2 . 4 m g / d l U ri na r y el ec tr o ly te e xc r et io ns N a 1 8 1 1 1 2 9 2 K 5 9 6 1 4 7 c ¥ 17 0 1 1 7 1 0 8 C a 2. 3 1 6 6 .0 > 5 0 m g / d a y P la sm a r en i n ac t iv it y 8 3 .6 48 . 6 7 8 .4 5- 3 0 n g A n g l / m l / 6 h P la s m a a l d os t e r on e co n c en t r a ti o n 26 . 9 4 1.4 2 5 .7 2 - 12 n g /d l S e r u m p H 7. 4 7 7. 4 8 7 .4 5 7 . 3 5 - 7 .4 5 S e r um b ic a r bo n a t e l e ve l s 32 . 5 3 0. 3 2 8. 5 2 1 - 2 5 m m o l / /

JZhr-0 0=H2 å I 6 I I I I 6°j*h\ ^ (Tali

Case 1 Case 2 Case 3

Figure 1. Pedigrees of kindred of the three patients with Gitelman's syndrome. Probands are indicated by arrows.

Ang II (9, 10) was normal, and hypomagnesemia and marked she has been followed up in the outpatient clinic. Hypokalemia hypocalciuria were confirmed. Based on these clinical findings, (2.4-3. 1 mmol//) and hypomagnesemia (1.6-1.8 mg/dl) how- Gitelman's syndrome was diagnosed. Familial relationships everpersisted. She had no history of chronic diarrhea, vomiting, are shown in Fig. 1. and diuretics abuse. Her height was 146.7 cm, and weight, 47.4 kg. Blood pressure levels were less than 1 10/70 mmHg. Case2 In February 1996, urinary levels were determined, A 5 1 -year-old Japanese female had suffered from diabetes and marked hypocalciuria was confirmed. Gitelman's syn- mellitus since the age of 40. When diabetes mellitus was drome was therefore diagnosed. Laboratory data during the diagnosed, hypokalemia (2.4 mmol//) was also observed, and admission are listed in Table 1. Pressor response to Ang II was she was admitted to our department. Byintense examinations, normal. Family relationships are shown in Fig. 1. Her father hypotension, hypokalemia, metabolic alkalosis, and suffered from cerebral infarction in August 1994. During his hyperreninemic hyperaldosteronism were identified. Pressor admission in a hospital, persistent hypokalemia (2.4-2.7 mmol/ response to Ang II was normal. Hypomagnesemia ( 1.6-1.8 mg/ /), hyponatremia ( 1 24-1 34 mmol//), hypomagnesemia ( 1.6 mg/ dl) was also observed. A variant of Bartter's syndrome was dl), and hypocalciuria (12 mg/day) were observed. A diagnosis diagnosed because of the lack of blunted pressor response to of Gitelman's syndrome was also likely for him. Ang II. Potassium (32 mmol/day) has been administered, and

Internal Medicine Vol. 36, No. 8 (August 1997) 583 Takeuchi et al

Case3 A 66-year-old Japanese female had suffered from recurrent Discussion . She was first admitted to a neighboring hospital in 1983 because of pyelonephritis. Mild muscle paralysis was Gitelman's syndrome was diagnosed in three patients with felt, and laboratory examination disclosed low serum potas- hypokalemia, secondary aldosteronism, and normotension in sium levels (2.0-3.0 mmol//). Although pyelonephritis im- conjunction with hypomagnesemiaand hypocalciuria. Although proved, hypokalemia and secondary aldosteronism were per- Gitelman ' s syndrome is similar to Bartter' s syndrome, both are sistently observed. Potassium 32 mmol/day and spironolactone distinguished on the basis of urinary calcium excretion because 50 mg/day were then administered, and she was followed up in true Bartter ' s syndrome shows a tendency towards hypercalciuria the outpatient clinic. She had no history of chronic diarrhea, whereasGitelman's syndromeshowsmarkedhypocalciuria. vomiting, and diuretics abuse. Blood pressure levels were Mild secondary aldosteronism was observed in the patients, and within the normal range. In July 1995, hypokalemia worsened pressor response to AngII was not impaired in the three (serum potassium, 1.9 mmol//), and she was referred to our subjects. Onthe other hand, true Bartter's syndrome is known department for further examinations. Representative labora- to showthe blunted pressor response. In the three patients, tory data during admission are listed in Table 1. Pressor re- secondary aldosteronism was not persistent, and the plasma sponse to Angll was normal. Based on these findings, Gitelman' s renin activity was normal in the experimental periods. In syndrome was diagnosed. Familial relationships are shownin Gitelman's syndrome, secondary reninism due to salt wasting Fig.1. is not so severe, and therefore desensitization of AngII re- ceptor may not necessarily occur. Congenital disorder of DNArestriction fragment length polymorphism Gitelman's syndrome was first diagnosed in adults in these Wehave reported (8) a mutation in TSC gene (T to C change patients. Since the clinical findings ofGitelman' s syndrome are at 1,868 nucleic acid position) in the Gitelman's syndrome so mild compared to those of true Bartter' s syndrome, Gitelman ' s kindred including Case 1 , which causes substitution of leucine abnormalities may possibly be discovered by chance in adults for proline at the 623 position, and creates Nci I such as in the present cases, in which the disorder was un- restriction site in the exon 15. Weperformed Nci I digestion of maskedin the course of examinations for other diseases such as the DNAfragment (-250 base pairs) ofPCR-amplified exon 15 uterine myoma, diabetes mellitus, and pyelonephritis. Early of TSC gene based on the previously reported methods (8) in and careful determinations of urinary calcium excretion as well order to determine if the samemutation as found in Case 1 as plasma magnesiumlevels are recommendedfor the diagno- would be identified in the ot^er cases with Gitelman's syn- sis ofGitelman' s syndromein normotensive patients of second- drome (Case 2 and 3). GenomeDNAwas obtained and analyzed ary aldosteronism with hypokalemic alkalosis ( 1). after informed consent was obtained. Figure 2 shows that Case Sodiummetabolism at the renal distal convoluted tubules la or lb is homozygousor heterozygous for the Nci I digestion (DCT) is known to be mediated by a Na-Cl cotransporter, which fragment length polymorphism, respectively, as shown previ- stimulates absorption of NaCl, and is inhibited by thiazide ously (8). This mutation was not detected in Cases 2 and 3. diuretics (ll, 12). Actually, the thiazide-sensitive Na-Cl cotransporter has been shown to be localized to DCT( 1 3). In the administration of thiazide diuretics, hypocalciuria as well as la lb hypokalemia is frequently observed (14). Hypokalemia may likely be due to an increase in the sodium delivery to distal collecting ducts, where potassium secretion occurs in exchange for sodiumabsorption. Hypocalciuria maypossibly be due to a decrease in the cytosolic chloride level by inactivation of the Na-Cl cotransporter in conjunction with chloride channel activ- ity, which activates the calcium channel causing calcium reab- sorption at DCT( 15). Inactivation of this transporter caused by any mutation maytherefore lead to the hypokalemia and hypocalciuria observed in Gitelman's syndromesimilar to those observed with administration of thiazide diuretics. Previ- ously, Simon et al showed by linkage analysis with genetic Figure2. Nci I digestion of exon 15 of thiazide- markers that the locus responsible for Gitelman's syndrome sensitive Na-Cl cotransporter gene in cases of Gitelman's encodes TSC, and a wide variety of mutations in TSCgene were syndrome. r250 base pairs ofexon 15 is amplified, and the identified in patients with Gitelman's syndrome (7). Wehave fragment is digested with Nci I when the mutation (thy- also recently reported that a missense mutation in TSCat 623 mine to cytosine) at 1,868 base position in exon 15 is amino acid position is associated with the clinical findings occurred (8). Case la or lb is a homozygote or heterozy- (especially, hypokalemia, and hypocalciuria) observed in the gote of the previously reported mutation, respectively. kindred including Case 1 (8). However, the mutation detected Cases 2 and 3 lack this mutation. by Nci I digestion of the exon 15 DNAfragment of TSC gene

584 Internal Medicine Vol. 36, No. 8 (August 1997) Gitelman's Syndrome and Na-Cl Cotransporter was not identified in either Case 2 or 3. The results suggest that the Gitelman's syndromeis not only attributed to the same References mutation as identified in Case 1, and may possibly further 1) Bettinelli A, Bianchetti MG, Girardin E, et al. Use of calcium excretion support the report by Simon et al (7) that Gitelman's syndrome values to distinguish two forms of primary renal tubular hypokalemic is possibly be derived by a wide variety of mutations in TSC alkalosis: Bartter and Gitelman syndromes. J Pediatr 120: 38, 1992. gene, although abnormalities in the other functional proteins in 2) Sutton RA, Mavichak V, Halabe A, Wilkins GE. Banter's syndrome: Cases 2 and 3 could not be excluded. evidence suggesting a distal tubular defect in a hypocalciuric variant of the syndrome. Miner Electrolyte Metab 18: 43, 1992. Genetic heterogeneity in Gitelman' s syndrome has recently 3) Bartter FC, Pronove P, Gill JR Jr, MacCardle RC. Hyperplasia of been reported: one type is that with autosomal recessive inher- juxtaglomerular complex with hyperaldosteronism and hypokalemic itance presenting frequent tetanic episodes and lower plasma alkalosis. AmJ Med 33: 811, 1962. potassium and magnesiumlevels, and the other is that with 4) GitelmanHJ, GrahamJB, WeltLG. Anew familial disordercharacterized by hypokalemia and hypomagnesemia. Trans Assoc AmPhysicians 79: autosomal dominant inheritance presenting mild muscular 221, 1966. weakness and normal potassium levels (16). Although it is not 5) Simon DB, Karet FE, Hamdan JM, DiPietro AD, et al. Bartter's syn- determined whether these two types of Gitelman's syndrome drome, hypokalaemic alkalosis with hypercalciuria, is caused by muta- are caused by the same , the clinical findings tions in the Na-K-2C1 cotransporter NKCC2.Nat Genet 13: 183, 1996. observedin the present cases are compatible with the autosomal 6) Simon DB, Karet FE, Rodrigues-Sariano J, et al. Genetic heterogeneity recessive inheritance type of Gitelman's syndrome, as shown of Bartter's syndrome revealed by mutations in the K+ channel, ROMK. Nat Genet 14: 152, 1996. previously (8). Considering the family relationship of Case 2, 7) Simon DB, Nelson-Williams C, Bia MJ, et al. Gitelman's variant of her father, whohas suffered from hyponatremia, hypokalemia Bartter's syndrome, inherited hypokalaemic alkalosis, is caused by mu- and cerebral infarction probably due to salt wasting, is possibly tations in the thiazide-sensitive Na-Cl cotransporter. Nat Genet 12: 24, a homozygotefor a mutation. 1996. 8) Takeuchi K, Kure S, Kato T, et al. Association of a mutation in thiazide- Gitelman's syndromeshowsnormotensionor hypotension sensitive Na-Cl cotransporter with familial Gitelman' s syndrome. J Clin probably due to a defect in the Na-Cl cotransporter. Bartter's Endocrinol Metab 81: 4496, 1996. syndrome is also characterized by normotension/hypotension, 9) Kaplan NM, Silah JG. The effect of angiotensin II on the blood pressure hypokalemic alkalosis, and salt wasting, etc which has recently in humans with hypertensive diseases. J Clin Invest 43: 659, 1964. been shown to be due to mutations of Na-K-2C1cotransporter 10) Takeuchi K, Imai Y, Omata K, et al. A case of secondary aldosteronism similar to Bartter's syndromewith no abnormality in renal chloride NKCC2(5) or potassium channel ROMK(6) by inhibiting these reabsorption. Tohoku J Exp Med169: 141, 1993. activities. In contrast, mutations of the epithelial Na channel at 1 1) Costanzo LS. Localization of action in microperfused rat distal the distal tubules have been identified in patients with Liddle's tubules: Ca and Na transport. AmJ Physiol 248: F527, 1985. syndrome ( 17), and the mutations have been suggested to cause 12) Stanton BA. Cellularactions ofthiazide diuretics in the distal tubule. J Am constitutive stimulation of this channel leading to excessive Soc Nephrol 1: 832, 1990. sodiumabsorption and thereby excessive potassium secretion 1 3) Plotkin MD, Kaplan MR, VerlanderJW, et al. Localization of the thiazide sensitive Na-Cl cotransporter, rTSC l , in the . Kidney Int 50: 174, ( 1 8) resulting in Liddle' s syndrome that shows clinical findings 1996. such as hypertension, hypokalemic alkalosis, and hyporenine- 14) Middler S, Pak CY, Murad F, Bartter FC. Thiazide diuretics and calcium mic hypoaldosteronism. Thus, these channels and Na-Cl trans- metabolism. Metabolism 22: 139, 1973. porters located in the distal tubules are suggested to be impor- 15) Gesek FA, Friedman PA. Mechanism of calcium transport stimulated by chlorothiazide in mouse cells. J Clin Invest 90: tant for blood pressure regulation in terms of renal sodium, 429, 1992. chloride, and potassium . 16) Bettinelli A, Bianchetti M, Borella P, et al. Genetic heterogeneity in Acknowledgements:The present study was supported in part by Grants- tubular hypomagnesemia-hypokalemia with hypocalcuria (Gitelman' s in-Aid (Nos. 09470236 and 09877216) from the Ministry of Education, syndrome). Kidney Int 47: 547, 1995. Science, and Culture, Japan, and a grant for the Promotion of High Grade 17) Shimket RA, Warnock DG, Bositis CM, et al. Liddle's syndrome: Medical Science, Tohoku University School of Medicine. We wish to thank heritable human hypertension caused by mutations in (3 subunit of the Drs. S. Kure and K. Narisawa, Department of Biochemical Genetics, Tohoku epithelial sodium channel. Cell 79: 407, 1994. University, for technical assistance and discussion in the mutation study. 18) SnyderPM, PriceMP, McDonaldFJ, etal. Mechanismby whichLiddle's syndrome mutations increase activity of a human epithelial Na+ channel. Cell 83: 969, 1995.

Internal Medicine Vol. 36, No. 8 (August 1997) 585