- Email: [email protected]
Renal glucosuria due to SGLT2 mutations
Molecular Genetics and Metabolism 82 (2004) 56–58 www.elsevier.com/locate/ymgme
Renal glucosuria due to SGLT2 mutations Robert Kleta,a,b,¤ Caroline Stuart,a Fred A. Gill,c and William A. Gahla,b a
Section on Human Biochemical Genetics, Medical Genetics Branch, National Human Genome Research Institute, National Institutes of Health, Building 10, Room 10C-107, MSC 1851, 10 Center Drive, Bethesda, MD 20892-1851, USA b OYce of Rare Diseases, Intramural Program, OYce of the Director, National Institutes of Health, Bethesda, MD, USA c Internal Medicine Consult Service, Warren Grant Magnuson Clinical Center, National Institutes of Health, Bethesda, MD, USA Received 8 December 2003; received in revised form 28 January 2004; accepted 29 January 2004
Abstract Isolated renal glucosuria results from mutations in SGLT2, which codes for an active transporter speciWc for D-glucose and expressed in the luminal membrane of the renal proximal tubule. In aVected individuals, glucosuria leads to pursuit of hyperglycemia to exclude defects in glucose metabolism, and to investigation of renal proximal tubular function to exclude renal Fanconi syndrome. Here we present clinical and molecular data regarding a 19-year-old woman with isolated glucosuria. She was compound heterozygous for two SGLT2 mutations, i.e., a new missense mutation, T200K, and a known missense mutation, N654S. Published by Elsevier Inc. Keywords: D-Glucose; SGLT1; SGLT2; SGLT3; GLUT1; GLUT2; Renal Fanconi syndrome; Glucosuria
Introduction Renal glucose reabsorption represents an example of Na+-dependent transport across an epithelial membrane. Glucose is Wltered by the glomerulus and more than 99% reabsorbed within the proximal tubule through a sophisticated coupling of luminal secondary active glucose transporters, such as SGLT1 and SGLT2, with basolateral passive glucose transporters, such as GLUT2. SGLT1 and SGLT2 are thought to accomplish renal tubular glucose reabsorption in a sequential manner. In the early proximal tubule, the low aYnity, high capacity transporter SGLT2, with a sodium to glucose coupling of 1:1, reabsorbs the bulk of the Wltered glucose. In the late proximal tubule, the high aYnity, low capacity transporter SGLT1, with a sodium to glucose coupling of 1:2, nearly completes the reabsorption of glucose, removing practically all of this sugar from the urine [1]. Human “knock-outs” of the SGLT1, SGLT2, and GLUT2 genes are represented by the disorders glucose– galactose malabsorption, renal glucosuria, and Fanconi– Bickel syndrome, respectively. Currently, all known ¤
Corresponding author. Fax: 1-301-402-7290. E-mail address: [email protected] (R. Kleta).
1096-7192/$ - see front matter. Published by Elsevier Inc. doi:10.1016/j.ymgme.2004.01.018
cases of isolated renal glucosuria are due to mutations in SGLT2; the role of SGLT3, another sodium-linked glucose cotransporter expressed in the renal proximal tubule, is not fully understood. Transport of a cytostatic glucose derivative by SGLT3 has been reported [2], but we could not conWrm these Wndings using a diVerent and more physiological approach [3]. Mutations in GLUT1, which is expressed in the cells of the blood–brain barrier, give rise to the clinical picture of infantile seizures due to hypoglycorrhachia. This disorder is inherited in an autosomal dominant fashion, in contrast to the autosomal recessively inherited glucose–galactose malabsorption, renal glucosuria, and Fanconi–Bickel syndromes. We now report another case of isolated renal glucosuria, along with molecular evidence of two mutations in the SGLT2 gene, one of which has not been previously reported. Materials and methods The patient was enrolled in a clinical protocol for the diagnosis and treatment of patients with inborn errors of metabolism, approved by the National Human Genome Research Institute Institutional Review Board. She provided written, informed consent for these studies.
R. Kleta et al. / Molecular Genetics and Metabolism 82 (2004) 56–58
The Fanconi syndrome index (FSI), a quantitative measure of aminoaciduria, was determined by measuring the daily urinary excretion of 21 speciWc amino acids utilizing a Biochrom 30 amino acid analyzer (Biochrom, Holliston, MA, USA) [4]. Molecular studies for SGLT2 and SGLT3 were performed using standard procedures. In short, genomic DNA was isolated from whole blood using the Wizard Genomic DNA PuriWcation Kit (Promega, Madison, WI, USA). To document mutations in the coding regions of SGLT2 or SGLT3, genomic DNA was ampliWed (puReTaq Ready-to-Go PCR Beads, Amersham Biosciences, Piscataway, NJ, USA) using standard PCR techniques (PTC-200, MJ Research, Waltham, MA, USA) and sequenced in both directions utilizing a Beckman CEQ8000 (Beckman Coulter, Fullerton, CA, USA). The primers used for SGLT2 were published previously [5] and were obtained from Invitrogen (Carlsbad, CA, USA). The reverse primer for exon 9 was used instead of the published reverse primer for exon 8, which was found to reside within exon 8 itself. Appropriate primers for sequencing of all coding exons of SGLT3 were chosen using the public database of genomic data (NCBI) and are available upon request. Genomic DNA from the Human Variation Panels HD50CAU (50 Caucasians), HD 08 (10 Mexicans), HD23 (10 Northeast European Russians), and HD20 (9 Southeast European Russians) were obtained from Coriell Cell Repositories (Camden, NJ, USA) and used for polymorphism screening by sequencing of the exons of interest. NCBI Accession Nos.: mRNA SGLT2 NM_ 003041.1; mRNA SGLT3 NM_014227.1.
Results and discussion The patient is a 19-year-old college student. She was the product of a normal pregnancy and term delivery. Her parents were unrelated and of Russian/Cuban/ Spanish ancestry. The parents and siblings were not available for studies, but a younger sister was also reported to have glucosuria. In our patient, glucosuria was noted during a well-child visit at age 8 years, but since hyperglycemia was absent, no further measures were taken. The patient had no history of polydipsia, polyuria or hematuria and was taking no medications. She had no major illnesses or surgeries. The review of systems and the physical examination were completely normal. The blood pressure was 110/67 mm Hg, body weight 68 kg, and length 174 cm. Laboratory evaluations revealed no abnormalities of proximal tubular function except renal glucosuria, and no other conditions, which could contribute to hyperglycemia. Non-fasting plasma glucose concentrations ranged from 90 to 112 mg/dl. A 24-h urine collection revealed a glucose excretion of 9.1 g. Urine dipstick analyses showed 2+ to 4+ glucose; ketones, hemoglobin, and
57
protein were negative, and the pH was between 6.0 and 7.0. The speciWc gravities of random urine samples, containing excess glucose, were 1.021–1.027. The creatinine clearance, determined on a 24-h urine collection, was 106 ml/min/1.73 m2 (normal, 90–125). Fractional excretion rates for sodium, chloride, potassium, phosphate, calcium, and magnesium were normal. The urinary excretion of -2-microglobulin in a 24-h urine was 0.08 mg/L (normal, 0–0.30). Quantitative determination of urinary amino acids revealed an FSI of 107 mol/kg/ day (normal, 94 § 45). Plasma or serum values for sodium, potassium, chloride, total CO2, creatinine, glucose, BUN, albumin, calcium, magnesium, phosphorus, ALT, AST, total bilirubin, total protein, CK, osmolality, and uric acid were within normal limits. The sedimentation rate, thyroid function results, and plasma/serum levels of ACTH, cortisol, growth hormone, cholesterol, triglycerides, amylase, and lipase were also within normal limits. A complete blood count was unremarkable. Interestingly, the Hgb A1c was 4.6%, slightly below the reference range (4.8–6.4), and the insulin concentration was 5.7 U/ml, also below the reference range (6–27). The fasting C-peptide level was 1.7 ng/ml (normal, 0.9– 4.0), and the glucagon was 26 pg/ml (normal, 0–60). These Wndings raise the possibility that the isolated genetic condition of renal glucosuria might prevent the development of diabetes mellitus or ameliorate its symptoms. We speculate that speciWc inhibition of the renal transporter SGLT2, by lowering the body’s glucose level, might provide therapy for, or prophylaxis against, type II diabetes. Because of the ambiguous role of SGLT3 in glucose transport across the proximal tubule, we sequenced our patient’s SGLT3 gene. This revealed a normal sequence. Further molecular studies, however, revealed two mutations in the SGLT2 gene of our patient (Fig. 1). In exon 6 at codon 200, a novel heterozygous missense mutation (ACG 1 AAG) changes a conserved threonine to lysine. In exon 14 at codon 654, another heterozygous missense mutation (AAC 1 AGC) changes a conserved asparagine to serine. No other mutations in the coding region or splice sites of SGLT2 were present in this patient. Sequencing of 158 control alleles of Caucasian, Mexican, and Russian ancestry revealed one heterozygous N654S mutation in a Caucasian. No T200K mutation was present in any of the alleles. The rat (Rattus norvegicus) and human SGLT2 proteins contain 670 and 672 amino acids, respectively, and have 90% homology. Both of the mutations found in our patient cause changes in amino acids that are highly conserved among human, rat, and mouse SGLTs. In particular, Tabatabai et al. [6] have shown that both residues 200 (threonine) and 654 (asparagine) reside within conserved regions of the mouse genes SGLT1, SGLT2, SGLT3a, and SGLT3b. Santer et al. [7] considered one of our patient’s mutations (N654S) to be a polymorphism, in part because of its location in transmembrane region
58
R. Kleta et al. / Molecular Genetics and Metabolism 82 (2004) 56–58
Fig. 1. (A, B) Sequence analysis of SGLT2 exons 6 and 14 showing genomic DNA from the patient. (A) Shows the forward sequence from part of exon 6; (B) shows the reverse sequence from part of exon 14. Arrows indicate the c599 position, showing a heterozygous alteration of C 1 A, and the c1961 position, showing a heterozygous alteration of T 1 C (A 1 G). These lead to T200K and N654S missense mutations in SGLT2, respectively.
14, even though it segregated with renal glucosuria in a family in which the proband was compound heterozygous for this mutation with another missense mutation. Interestingly, N654S was present in 1 of 80 control alleles of unknown ethnicity. However, we believe that this alteration is disease causing because the codon is conserved across species, no other mutation was found in our patient despite exhaustive sequencing, and Calado et al. [8] recently reported a patient with isolated renal glucosuria in whom N654S was one of two compound heterozygous SGLT2 mutations. In fact, we speculate that N654S might represent a common mutation in SGLT2. Alternatively, it might represent a mutation, which allows for residual function of the SGLT2 glucose transporter, since heterozygote carriers show almost no glucose wasting [7]. Our patient’s glucose excretion of 9.1 g/day translates into 8.8 g/1.73 m2/day, or 5.7 g/L; patients with homozygous nonsense mutations in SGLT2 showed glucose excretions of 68.7–162.2 g/ 1.73 m2/day [7] and 61.6 g/L [5]. These diVerences may reXect a genotype/phenotype correlation. The consequences of isolated renal glucosuria due to SGLT2 mutations are not profound. Glucose wasting in our patient cost her only 36 kcal/day, and even severely aVected patients lose only a few hundred kilocalories in wasted glucose. Rather, the danger lies in mistaking isolated renal glucosuria for a more serious condition, such as diabetes mellitus, and pursuing it with invasive interventions. The key to avoiding misdiagnosis is recognition of the disorder itself. Acknowledgments We greatly appreciate the professional and dedicated work of the nurses of the 8 West ward and of Stacey
Gordon, NP of the Warren Grant Magnuson Clinical Center of the National Institutes of Health. We thank Isa Bernardini for performing the amino acid analysis.
References [1] Y. Kanai, W.S. Lee, G. You, D. Brown, M.A. Hediger, The human kidney low aYnity Na+/glucose cotransporter SGLT2. Delineation of the major renal reabsorptive mechanism for D-glucose, J. Clin. Invest. 93 (1994) 397–404. [2] M. Veyhl, K. Wagner, C. Volk, V. Gorboulev, K. Baumgarten, W.M. Weber, M. Schaper, B. Bertram, M. Wiessler, H. Koepsell, Transport of the new chemotherapeutic agent beta-D-glucosylisophosphoramide mustard (D-19575) into tumor cells is mediated by the Na+-D-glucose cotransporter SAAT1, Proc. Natl. Acad. Sci. USA 17 (95) (1998) 2914–2999. [3] R. Kleta, B.C. Burckhardt, N.A. WolV, E. Schlatter, Unexpected electrophysiological eVects of D-19575, a new cytostatic drug, Nephrol. Dial. Transplant. 14 (Suppl. 4) (1999) 18–20. [4] L.R. Charnas, I. Bernardini, D. Rader, J.M. Hoeg, W.A. Gahl, Clinical and laboratory Wndings in the oculocerebrorenal syndrome of Lowe, with special reference to growth and renal function, N. Engl. J. Med. 324 (1991) 1318–1325. [5] L.P. van den Heuvel, K. Assink, M. Willemsen, L. Monnens, Autosomal recessive renal glucosuria attributable to a mutation in the sodium glucose cotransporter (SGLT2), Hum. Genet. 111 (2002) 544–547. [6] N.M. Tabatabai, S.S. Blumenthal, D.L. Lewand, D.H. Petering, Mouse kidney expresses mRNA of four highly related sodium-glucose cotransporters: regulation by cadmium, Kidney Int. 64 (2003) 1320–1330. [7] R. Santer, M. Kinner, C.L. Lassen, R. Schneppenheim, P. Eggert, M. Bald, J. Brodehl, M. Daschner, J.H. Ehrich, M. Kemper, S. Li Volti, T. Neuhaus, F. Skovby, P.G. Swift, J. Schaub, D. Klaerke, Molecular analysis of the SGLT2 gene in patients with renal glucosuria, J. Am. Soc. Nephrol. 14 (2003) 2873–2882. [8] J. Calado, K. Soto, C. Clemente, P. Correia, J. RueV, Novel compound heterozygous mutations in SLC5A2 are responsible for autosomal recessive renal glucosuria, Hum. Genet. 114 (2004) 314–316.
Renal glucosuria due to SGLT2 mutations Robert Kleta,a,b,¤ Caroline Stuart,a Fred A. Gill,c and William A. Gahla,b a
Section on Human Biochemical Genetics, Medical Genetics Branch, National Human Genome Research Institute, National Institutes of Health, Building 10, Room 10C-107, MSC 1851, 10 Center Drive, Bethesda, MD 20892-1851, USA b OYce of Rare Diseases, Intramural Program, OYce of the Director, National Institutes of Health, Bethesda, MD, USA c Internal Medicine Consult Service, Warren Grant Magnuson Clinical Center, National Institutes of Health, Bethesda, MD, USA Received 8 December 2003; received in revised form 28 January 2004; accepted 29 January 2004
Abstract Isolated renal glucosuria results from mutations in SGLT2, which codes for an active transporter speciWc for D-glucose and expressed in the luminal membrane of the renal proximal tubule. In aVected individuals, glucosuria leads to pursuit of hyperglycemia to exclude defects in glucose metabolism, and to investigation of renal proximal tubular function to exclude renal Fanconi syndrome. Here we present clinical and molecular data regarding a 19-year-old woman with isolated glucosuria. She was compound heterozygous for two SGLT2 mutations, i.e., a new missense mutation, T200K, and a known missense mutation, N654S. Published by Elsevier Inc. Keywords: D-Glucose; SGLT1; SGLT2; SGLT3; GLUT1; GLUT2; Renal Fanconi syndrome; Glucosuria
Introduction Renal glucose reabsorption represents an example of Na+-dependent transport across an epithelial membrane. Glucose is Wltered by the glomerulus and more than 99% reabsorbed within the proximal tubule through a sophisticated coupling of luminal secondary active glucose transporters, such as SGLT1 and SGLT2, with basolateral passive glucose transporters, such as GLUT2. SGLT1 and SGLT2 are thought to accomplish renal tubular glucose reabsorption in a sequential manner. In the early proximal tubule, the low aYnity, high capacity transporter SGLT2, with a sodium to glucose coupling of 1:1, reabsorbs the bulk of the Wltered glucose. In the late proximal tubule, the high aYnity, low capacity transporter SGLT1, with a sodium to glucose coupling of 1:2, nearly completes the reabsorption of glucose, removing practically all of this sugar from the urine [1]. Human “knock-outs” of the SGLT1, SGLT2, and GLUT2 genes are represented by the disorders glucose– galactose malabsorption, renal glucosuria, and Fanconi– Bickel syndrome, respectively. Currently, all known ¤
Corresponding author. Fax: 1-301-402-7290. E-mail address: [email protected] (R. Kleta).
1096-7192/$ - see front matter. Published by Elsevier Inc. doi:10.1016/j.ymgme.2004.01.018
cases of isolated renal glucosuria are due to mutations in SGLT2; the role of SGLT3, another sodium-linked glucose cotransporter expressed in the renal proximal tubule, is not fully understood. Transport of a cytostatic glucose derivative by SGLT3 has been reported [2], but we could not conWrm these Wndings using a diVerent and more physiological approach [3]. Mutations in GLUT1, which is expressed in the cells of the blood–brain barrier, give rise to the clinical picture of infantile seizures due to hypoglycorrhachia. This disorder is inherited in an autosomal dominant fashion, in contrast to the autosomal recessively inherited glucose–galactose malabsorption, renal glucosuria, and Fanconi–Bickel syndromes. We now report another case of isolated renal glucosuria, along with molecular evidence of two mutations in the SGLT2 gene, one of which has not been previously reported. Materials and methods The patient was enrolled in a clinical protocol for the diagnosis and treatment of patients with inborn errors of metabolism, approved by the National Human Genome Research Institute Institutional Review Board. She provided written, informed consent for these studies.
R. Kleta et al. / Molecular Genetics and Metabolism 82 (2004) 56–58
The Fanconi syndrome index (FSI), a quantitative measure of aminoaciduria, was determined by measuring the daily urinary excretion of 21 speciWc amino acids utilizing a Biochrom 30 amino acid analyzer (Biochrom, Holliston, MA, USA) [4]. Molecular studies for SGLT2 and SGLT3 were performed using standard procedures. In short, genomic DNA was isolated from whole blood using the Wizard Genomic DNA PuriWcation Kit (Promega, Madison, WI, USA). To document mutations in the coding regions of SGLT2 or SGLT3, genomic DNA was ampliWed (puReTaq Ready-to-Go PCR Beads, Amersham Biosciences, Piscataway, NJ, USA) using standard PCR techniques (PTC-200, MJ Research, Waltham, MA, USA) and sequenced in both directions utilizing a Beckman CEQ8000 (Beckman Coulter, Fullerton, CA, USA). The primers used for SGLT2 were published previously [5] and were obtained from Invitrogen (Carlsbad, CA, USA). The reverse primer for exon 9 was used instead of the published reverse primer for exon 8, which was found to reside within exon 8 itself. Appropriate primers for sequencing of all coding exons of SGLT3 were chosen using the public database of genomic data (NCBI) and are available upon request. Genomic DNA from the Human Variation Panels HD50CAU (50 Caucasians), HD 08 (10 Mexicans), HD23 (10 Northeast European Russians), and HD20 (9 Southeast European Russians) were obtained from Coriell Cell Repositories (Camden, NJ, USA) and used for polymorphism screening by sequencing of the exons of interest. NCBI Accession Nos.: mRNA SGLT2 NM_ 003041.1; mRNA SGLT3 NM_014227.1.
Results and discussion The patient is a 19-year-old college student. She was the product of a normal pregnancy and term delivery. Her parents were unrelated and of Russian/Cuban/ Spanish ancestry. The parents and siblings were not available for studies, but a younger sister was also reported to have glucosuria. In our patient, glucosuria was noted during a well-child visit at age 8 years, but since hyperglycemia was absent, no further measures were taken. The patient had no history of polydipsia, polyuria or hematuria and was taking no medications. She had no major illnesses or surgeries. The review of systems and the physical examination were completely normal. The blood pressure was 110/67 mm Hg, body weight 68 kg, and length 174 cm. Laboratory evaluations revealed no abnormalities of proximal tubular function except renal glucosuria, and no other conditions, which could contribute to hyperglycemia. Non-fasting plasma glucose concentrations ranged from 90 to 112 mg/dl. A 24-h urine collection revealed a glucose excretion of 9.1 g. Urine dipstick analyses showed 2+ to 4+ glucose; ketones, hemoglobin, and
57
protein were negative, and the pH was between 6.0 and 7.0. The speciWc gravities of random urine samples, containing excess glucose, were 1.021–1.027. The creatinine clearance, determined on a 24-h urine collection, was 106 ml/min/1.73 m2 (normal, 90–125). Fractional excretion rates for sodium, chloride, potassium, phosphate, calcium, and magnesium were normal. The urinary excretion of -2-microglobulin in a 24-h urine was 0.08 mg/L (normal, 0–0.30). Quantitative determination of urinary amino acids revealed an FSI of 107 mol/kg/ day (normal, 94 § 45). Plasma or serum values for sodium, potassium, chloride, total CO2, creatinine, glucose, BUN, albumin, calcium, magnesium, phosphorus, ALT, AST, total bilirubin, total protein, CK, osmolality, and uric acid were within normal limits. The sedimentation rate, thyroid function results, and plasma/serum levels of ACTH, cortisol, growth hormone, cholesterol, triglycerides, amylase, and lipase were also within normal limits. A complete blood count was unremarkable. Interestingly, the Hgb A1c was 4.6%, slightly below the reference range (4.8–6.4), and the insulin concentration was 5.7 U/ml, also below the reference range (6–27). The fasting C-peptide level was 1.7 ng/ml (normal, 0.9– 4.0), and the glucagon was 26 pg/ml (normal, 0–60). These Wndings raise the possibility that the isolated genetic condition of renal glucosuria might prevent the development of diabetes mellitus or ameliorate its symptoms. We speculate that speciWc inhibition of the renal transporter SGLT2, by lowering the body’s glucose level, might provide therapy for, or prophylaxis against, type II diabetes. Because of the ambiguous role of SGLT3 in glucose transport across the proximal tubule, we sequenced our patient’s SGLT3 gene. This revealed a normal sequence. Further molecular studies, however, revealed two mutations in the SGLT2 gene of our patient (Fig. 1). In exon 6 at codon 200, a novel heterozygous missense mutation (ACG 1 AAG) changes a conserved threonine to lysine. In exon 14 at codon 654, another heterozygous missense mutation (AAC 1 AGC) changes a conserved asparagine to serine. No other mutations in the coding region or splice sites of SGLT2 were present in this patient. Sequencing of 158 control alleles of Caucasian, Mexican, and Russian ancestry revealed one heterozygous N654S mutation in a Caucasian. No T200K mutation was present in any of the alleles. The rat (Rattus norvegicus) and human SGLT2 proteins contain 670 and 672 amino acids, respectively, and have 90% homology. Both of the mutations found in our patient cause changes in amino acids that are highly conserved among human, rat, and mouse SGLTs. In particular, Tabatabai et al. [6] have shown that both residues 200 (threonine) and 654 (asparagine) reside within conserved regions of the mouse genes SGLT1, SGLT2, SGLT3a, and SGLT3b. Santer et al. [7] considered one of our patient’s mutations (N654S) to be a polymorphism, in part because of its location in transmembrane region
58
R. Kleta et al. / Molecular Genetics and Metabolism 82 (2004) 56–58
Fig. 1. (A, B) Sequence analysis of SGLT2 exons 6 and 14 showing genomic DNA from the patient. (A) Shows the forward sequence from part of exon 6; (B) shows the reverse sequence from part of exon 14. Arrows indicate the c599 position, showing a heterozygous alteration of C 1 A, and the c1961 position, showing a heterozygous alteration of T 1 C (A 1 G). These lead to T200K and N654S missense mutations in SGLT2, respectively.
14, even though it segregated with renal glucosuria in a family in which the proband was compound heterozygous for this mutation with another missense mutation. Interestingly, N654S was present in 1 of 80 control alleles of unknown ethnicity. However, we believe that this alteration is disease causing because the codon is conserved across species, no other mutation was found in our patient despite exhaustive sequencing, and Calado et al. [8] recently reported a patient with isolated renal glucosuria in whom N654S was one of two compound heterozygous SGLT2 mutations. In fact, we speculate that N654S might represent a common mutation in SGLT2. Alternatively, it might represent a mutation, which allows for residual function of the SGLT2 glucose transporter, since heterozygote carriers show almost no glucose wasting [7]. Our patient’s glucose excretion of 9.1 g/day translates into 8.8 g/1.73 m2/day, or 5.7 g/L; patients with homozygous nonsense mutations in SGLT2 showed glucose excretions of 68.7–162.2 g/ 1.73 m2/day [7] and 61.6 g/L [5]. These diVerences may reXect a genotype/phenotype correlation. The consequences of isolated renal glucosuria due to SGLT2 mutations are not profound. Glucose wasting in our patient cost her only 36 kcal/day, and even severely aVected patients lose only a few hundred kilocalories in wasted glucose. Rather, the danger lies in mistaking isolated renal glucosuria for a more serious condition, such as diabetes mellitus, and pursuing it with invasive interventions. The key to avoiding misdiagnosis is recognition of the disorder itself. Acknowledgments We greatly appreciate the professional and dedicated work of the nurses of the 8 West ward and of Stacey
Gordon, NP of the Warren Grant Magnuson Clinical Center of the National Institutes of Health. We thank Isa Bernardini for performing the amino acid analysis.
References [1] Y. Kanai, W.S. Lee, G. You, D. Brown, M.A. Hediger, The human kidney low aYnity Na+/glucose cotransporter SGLT2. Delineation of the major renal reabsorptive mechanism for D-glucose, J. Clin. Invest. 93 (1994) 397–404. [2] M. Veyhl, K. Wagner, C. Volk, V. Gorboulev, K. Baumgarten, W.M. Weber, M. Schaper, B. Bertram, M. Wiessler, H. Koepsell, Transport of the new chemotherapeutic agent beta-D-glucosylisophosphoramide mustard (D-19575) into tumor cells is mediated by the Na+-D-glucose cotransporter SAAT1, Proc. Natl. Acad. Sci. USA 17 (95) (1998) 2914–2999. [3] R. Kleta, B.C. Burckhardt, N.A. WolV, E. Schlatter, Unexpected electrophysiological eVects of D-19575, a new cytostatic drug, Nephrol. Dial. Transplant. 14 (Suppl. 4) (1999) 18–20. [4] L.R. Charnas, I. Bernardini, D. Rader, J.M. Hoeg, W.A. Gahl, Clinical and laboratory Wndings in the oculocerebrorenal syndrome of Lowe, with special reference to growth and renal function, N. Engl. J. Med. 324 (1991) 1318–1325. [5] L.P. van den Heuvel, K. Assink, M. Willemsen, L. Monnens, Autosomal recessive renal glucosuria attributable to a mutation in the sodium glucose cotransporter (SGLT2), Hum. Genet. 111 (2002) 544–547. [6] N.M. Tabatabai, S.S. Blumenthal, D.L. Lewand, D.H. Petering, Mouse kidney expresses mRNA of four highly related sodium-glucose cotransporters: regulation by cadmium, Kidney Int. 64 (2003) 1320–1330. [7] R. Santer, M. Kinner, C.L. Lassen, R. Schneppenheim, P. Eggert, M. Bald, J. Brodehl, M. Daschner, J.H. Ehrich, M. Kemper, S. Li Volti, T. Neuhaus, F. Skovby, P.G. Swift, J. Schaub, D. Klaerke, Molecular analysis of the SGLT2 gene in patients with renal glucosuria, J. Am. Soc. Nephrol. 14 (2003) 2873–2882. [8] J. Calado, K. Soto, C. Clemente, P. Correia, J. RueV, Novel compound heterozygous mutations in SLC5A2 are responsible for autosomal recessive renal glucosuria, Hum. Genet. 114 (2004) 314–316.