A novel SGLT2 mutation in a patient with autosomal recessive renal glucosuria

Jean Francis 2 3 Junhui Zhang 1 2 Anita Farhi 0 2 Hugh Carey 2 3 David S. Geller 1 2 0 Department of Genetics, Yale University School of Medicine , CT, USA 1 Section of Nephrology 2 1 Gilbert Street, Section of Nephrology, Yale University School of Medicine , New Haven, CT 06510, USA 3 Department of Internal Medicine, Hospital of St. Raphael - The patient is an 82-year-old healthy woman, the second child of consanguineous Italian parents. Her parents were first cousins. The patient was first noted to have glucosuria in the absence of hyperglycaemia at the age of 6. Her past medical history is pertinent for hypertension diagnosed 1 year ago and pyelonephritis diagnosed 2 years ago. She has no history of diabetes, heart, lung, neurological or musculoskeletal diseases, and denies any history of hypoglycaemia. Repeated evaluation in the past failed to define the cause of her persistent glucosuria. To her knowledge, her parents, brothers, sisters, cousins and children have no history of glucosuria. The patient came to our attention recently. An extensive laboratory investigation revealed no abnormalities with the exception of glucosuria, which was quantitated at >30 g/day, with a fractional glucose excretion of 55%. In particular, there was no proteinuria, b-2 microglobulinuria, acidosis, hyperglycaemia, amino aciduria or phosphaturia. Creatinine clearance was 49 ml/min. A blood sample for DNA analysis was obtained from our index patient after she signed an informed written consent form, in accordance with institutional review board-approved protocols, and genomic DNA was extracted as described previously [5]. All coding exons and flanking intronic regions of SLC5A2 were amplified and sequenced directly using the previously described primers [3]. Sequence analysis of SLC5A2 genomic DNA in our patient revealed only one mutation: a homozygous substitution of two adenines for two guanines within exon 8 (G910A, G911A, numbered from the start codon) altering Gly304 to lysine (Figure 1A and B); the identified mutation in our patient was confirmed by repeat polymerase chain reaction (PCR) amplification and sequencing of the opposite strand. Of interest, we detected no heterozygosity in the >5000 bp we analysed, suggesting that the patient had inherited the identical allele from both parents due to the history of consanguinety The kidney plays a central role in the regulation of plasma glucose levels. Glucose is freely filtered at the glomerulus and is normally reabsorbed via specific apical glucose transporters. In recent years, the identity of these transporters has been established. The sodiumglucose co-transporter type 2 (SGLT2) is a 672 amino acid high-capacity low-affinity transporter expressed in the S1 segment of the proximal tubule which is believed to mediate the majority of renal glucose reabsorption. The type 1 sodiumglucose co-transporter SGLT1, primarily expressed in the S3 segment of the proximal tubule and the small intestine, is a high-affinity, low-capacity glucose transporter saturated at or near physiological glucose concentrations [1]. SGLT1 has a 10-fold greater affinity for galactose and its deficiency is responsible for glucosegalactose malabsorption [2]. Primary renal glucosuria (OMIM #233100) is a disorder characterized by renal glucose wasting in the absence of hyperglycaemia or other proximal tubular dysfunction. Deficiency of SGLT2 recently has been shown to cause autosomal recessive renal glucosuria [3,4]. Here, we describe a patient with autosomal recessive renal glucosuria attributable to a homozygous mutation in SLC5A2. This case is unique in that our patient is quite healthy and fit at the age of 82. Her wellbeing clarifies the benign nature of SGLT2 deficiency, which has implications for the treatment of diabetes and obesity. Fig. 1. Sequence alteration in SLC5A2 in a patient with primary renal glucosuria. Genomic DNA sequence of the coding strand of exon 8 in SGLT2. (A) Sequence of the selected region of exon 8 in a wild-type individual. (B) Sequence of the same region of exon 8 in our index patient with primary glucosuria. Our patient has a homozygous substitution of adenines for guanines at the indicated position (nucleotides 910 and 911 numbered from the start codon) resulting in a substitution of lysine for glycine. (C and D) Conservation of G304 in SGLT2 in other species (C) and in related sodiumsolute co-transporters in humans (D). The amino acid sequences of selected sodiumsolute transporters were aligned using Clustal W [9], and the residues corresponding to Gly304 are highlighted. Members of the sodiumsolute co-transporter family all have either glycine or the structurally similar alanine at this position. in her family. We screened exon 8 in 180 control individuals whose DNA had been referred to our laboratory for evaluation of other genetic diseases; these individuals were primarily American of Caucasian descent with a small number of HispanicAmericans as well. Using denaturing high-performance liquid chromatography (DHPLC) as described previously [6], we did not observe this allele in any other individuals (data not shown), indicating that the G304K mutation is not a common polymorphism. Previous studies have documented that patients heterozygous for disease-causing mutations in SLC5A2 may have elevated levels of glucosuria as well [4,7]. We screened the two children of our index patient, each of whom is heterozygous for the G304K mutation (data not shown), and found that they had 52 and 215 mg glucosuria/24 h, respectively. These amounts are within the normal range for 24 h glucose excretion. The identified mutation substitutes the bulky, charged amino acid lysine for the small, hydrophobic glycine at residue 304. Gly304 is thought to lie intracellularly just beyond the seventh transmembrane domain [1]. This is a highly conserved region within SGLT2: in SGLT2 from other mammalian species, this position is always occupied by either glycine or the structurally similar alanine (Figure 1C). Furthermore, this position is occupied exclusively by either glycine or alanine in the broader family of human sodiumsolute co-transporters (Figure 1D). The identification of a homozygous non-conservative mutation in a highly conserved residue which is not seen in 360 control chromosomes in a gene previously implicated in primary renal glucosuria strongly suggests that this is indeed the disease-causing mutation. Definitive proof of this point would require a functional assay for SGLT2, which currently is not available [1]. This report joins a growing number of reports of primary renal glucosuria attributable to mutations SGLT2 mutation in recessive renal glucosuria in SLC5A2, and it further confirms that mutations in SLC5A2 cause autosomal recessive glucosuria. As with the previously reported patients with recessive renal glucosuria attributable to mutations in SLC5A2, our patient has massive glucosuria in the absence of hyperglycaemia, met (...truncated)