Skip Navigation



NDT Advance Access published online on July 12, 2008
Nephrology Dialysis Transplantation, doi:10.1093/ndt/gfn386
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
23/12/3874    most recent
gfn386v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Calado, J.
Right arrow Articles by Santer, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Calado, J.
Right arrow Articles by Santer, R.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author [2008]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org


Twenty-one additional cases of familial renal glucosuria: absence of genetic heterogeneity, high prevalence of private mutations and further evidence of volume depletion

Joaquim Calado1,2, Yves Sznajer3, Daniel Metzger4, Ana Rita2, Marie C. Hogan5, Antonis Kattamis6, Mauro Scharf7, Velibor Tasic8, Johann Greil9, Florian Brinkert10, Markus J. Kemper10 and René Santer10

1 Department of Genetics, Faculty of Medical Sciences, New University of Lisbon 2 Department of Nephrology, Hospital Curry Cabral, Lisbon, Portugal 3 Unité de Génétique Clinique Pédiatrique, Hôpital Universitaire des Enfants Reine Fabiola et Centre de Génétique Hôpital Erasme, Université Libre de Bruxelles, Bruxelles, Belgium 4 Pediatric Endocrinology, BC Children's Hospital, Vancouver, BC, Canada 5 Division of Nephrology, Mayo Clinic, Rochester, MN, USA 6 First Department of Pediatrics, University of Athens, School of Medicine, ‘Agia Sofia’ Children's Hospital, Athens, Greece 7 Department of Pediatric Endocrinology, Diabetes Center, Curitiba, Brazil 8 Department of Pediatric Nephrology, University Children's Hospital, Skopje, Former Yugoslavian Republic of Macedonia 9 Department of Pediatrics, University of Heidelberg, Heidelberg 10 Department of Pediatrics, University Medical Center Hamburg-Eppendorf, Hamburg, Germany

Correspondence and offprint requests to: Joaquim Calado, Departamento de Genética da Faculdade de Ciências Médicas, Instituto de Higiene e Medicina Tropical, Universidade Nova de Lisboa, Rua da Junqueira no. 96, 1349-008, Lisbon, Portugal. Tel: +351-2136-10297; Fax: +351-2136-22018; E-mail: jcalado.gene{at}fcm.unl.pt



   Abstract
Top
 Abstract
Introduction
 Material and methods
 Results
 Discussion
 References
 
Introduction. Familial renal glucosuria (FRG) is a rare renal tubular disorder caused by mutations within the SLC5A2 gene. It is characterized by persistent glucosuria in the absence of hyperglycaemia and any other signs of generalized tubular dysfunction. In small series of patients previously reported, the molecular and phenotypic findings in FRG families, including first hints of extracellular volume depletion and activation of the renin–angiotensin–aldosterone system induced by natriuresis, have been described. We have now extended this analysis to another 21 consecutive cases from 17 pedigrees, including 11 cases with severe glucose excretion.

Methods. Mutation analysis was performed by direct sequencing of the genomic coding segments of the SLC5A2 gene. In two cases with severe glucosuria, basal plasma renin activity and serum aldosterone concentrations were determined.

Results. Within the 17 pedigrees, we have identified a total of 20 different SLC5A2 mutations. Fifteen have not been previously reported. In all glucosuric individuals tested, at least one SLC5A2 mutation could be identified. Heterozygous individuals were found to have only mild glucose excretion whereas homozygous or compound heterozygous patients had severe glucosuria, ranging from 10 to 86.5 g/1.73 m2/24 h. In two patients of the latter group, basal plasma renin activity and serum aldosterone concentration were determined and found to be raised to an average of 4.6-fold and 3.1-fold of the upper limit of the normal range, respectively.

Discussion. The identification of at least one mutated allele in every affected individual in this cohort of 17 consecutively investigated families strongly suggests that genetic heterogeneity is not prevalent in FRG. Although 5 of the detected alleles have been described previously, 15 are novel, confirming that most mutations in FRG are private. Our finding of an activation of compensatory mechanisms for salt loss may warrant more detailed studies of long-term hormonal and metabolic imbalances in patients with FRG.

Keywords: glucose; renal; SGLT2; sodium



   Introduction
Top
 Abstract
 Introduction
Material and methods
 Results
 Discussion
 References
 
Familial renal glucosuria (FRG, McKusick 233100) is characterized by persistent glucosuria in the presence of a normal serum glucose concentration and the absence of any other impairment of tubular function. It is caused by mutations in the SLC5A2 gene, mapped to 16p11.2 and coding for the low-affinity sodium/glucose co-transporter SGLT2. This transporter is responsible for tubular reabsorption of the bulk of filtered glucose. Selected as a major candidate gene for FRG [1], the first report of a SLC5A2 mutation in FRG was presented in 2000 [2]. Recently published series with larger numbers of cases have confirmed that SLC5A2 mutations are indeed responsible for the vast majority of FRG cases [3,4]. Additional support came from reports each describing few or single cases [5–9]. To our knowledge, 29 different mutations have been reported to date, most of them being private (Figure 1). At least three patients have been reported in whom no mutation was detected even after sequencing the entire coding region and adjacent intronic segments of SLC5A2 [3,4], raising the possibility of genetic heterogeneity. Although long considered to be inherited as an autosomal recessive disorder, the mode of inheritance that best fits FRG has been suggested to be one of co-dominance with incomplete penetrance [3]. Many heterozygous individuals display mild glucosuria (<10 g/1.73 m2/24 h), while homozygous or compound heterozygous patients usually have severe renal glucosuria in excess of 10 g/1.73 m2/24 h. In a previous report we have described one such patient with massive glucosuria (65.6 g/1.73 m2/24 h) in whom we found evidence for activation of the renin–angiotensin–aldosterone system (RAAS) by extracellular volume depletion induced by natriuresis [4]. This observation, however, was based on a single patient, and clearly data from additional patients are required for confirmation.


Figure 1
View larger version (18K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Structure of the human sodium/glucose co-transporter SGLT2. Transmembrane domains are shown as dark grey boxes and are numbered 1–14. The novel mutations reported in this work are displayed above, while previously published ones are presented below. For references see the Introduction section.

 
Here, we describe 21 additional individuals with glucosuria of variable severity and 15 novel SLC5A2 mutations, thus extending the allelic heterogeneity in FRG, and report hormonal studies that elucidate pathophysiologic mechanisms in this condition.



   Material and methods
Top
 Abstract
 Introduction
 Material and methods
Results
 Discussion
 References
 
Nomenclature
Gene mutation nomenclature used in this article is in accordance with recommendations of den Dunnen and Antonarakis [10]; usage of gene symbols follows recommendations of the HUGO Gene Nomenclature Committee [11].

Patients and phenotype evaluation
Seventeen not previously reported and unrelated families were investigated. During the study period no other individuals with glucosuria were evaluated. The study was part of the medical evaluation for glucosuria. After obtaining informed consent from participating individuals or their legal guardians, 26 individuals could be included. Whenever possible, glucosuria was quantified by a 24-h urine collection. Relatives were screened for the presence of glucosuria by dipstick urine analysis. Eleven individuals presented with severe glucosuria (as defined previously by a glucose excretion ≥10 g/1.73 m2/24 h). In two of them basal plasma renin activity and serum aldosterone concentration were determined as previously reported [4].

Mutation analysis
Genomic DNA was isolated from peripheral blood leukocytes according to standard methods. The coding region and adjacent intronic segments of the SLC5A2 gene were screened for mutations by direct sequencing of polymerase chain reaction products in one of the two laboratories involved (J.C., R.S.), and the presence of mutations was confirmed by bidirectional sequencing.



   Results
Top
 Abstract
 Introduction
 Material and methods
 Results
Discussion
 References
 
Mutation and pedigree analysis
We have screened 26 individuals from 17 families; in 4 of them consanguinity was known from the medical history. Findings in individual patients and their relatives, when available, are presented in Table 1. The amount of glucose excreted ranged from 2.2 to 86.5 g/1.73 m2/ 24 h. For index cases of families 8, 10 and 13, glucosuria was not quantified. Twenty different mutations were identified in the study group. Five have been previously reported: IVS7 + 5G>A, p.R368W, p.F453L, IVS12 + 1G>A and p.N654S [3,4,6]. Within the 17 families, we identified a total of 15 novel mutations of the SLC5A2 gene: p.M44K, p.V116M, p.Y128X p.D201N, p.G224R, p.R300C, p.T323R, p.G356S, p.M382T, p.L387M, p.A469T, p.F476L, p.E482D, p.L539P and p.R558C (Figure 1 and 2).


View this table:
[in this window]
[in a new window]

  		Table 1 Mutation analysis and glucose excretion in affected individuals and their relatives

 

Figure 2
View larger version (19K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Multiple alignments of SGLT2 proteins from different species and comparison to human and zebrafish SGLT1 and human SGLT3. Missense mutations are depicted with bold characters. Highly conserved amino acids that are identical in all transporter proteins aligned are highlighted in grey.

 
Affected individuals from families 1, 2, 3 and 12 had mild glucosuria in the range of 2.7–6.5 g/1.73 m2/24 h. In each of these cases, only one mutated allele was identified: individuals from family 1, 2 and 12 were found to be heterozygous for three distinct novel mutations, p.V116M, p.R558C and p.R300C, respectively. Also the index patient from family 3, having one of the smallest amount of glucose excretion of the entire cohort, carried only one SLC5A2 mutation. Interestingly, she was heterozygous for the p.N654S allele, previously considered to be a polymorphism [3].

The most severe forms of renal glucosuria were found in homozygous or compound heterozygous individuals. Homozygosity for an SLC5A2 mutation (IVS7 + 5G>A, p.G224R, p.L539P, p.Y128X) was found in index cases of families 4, 5, 13, 14 and 17 while cases from families 6, 7, 8, 9, 10, 11, 15 and 16 were found to be compound heterozygous.

Renin–angiotensin–aldosterone system activation
Within the group of 10 families with documented severe glucosuria, we had the opportunity to evaluate the RAAS activation in two of them. In index cases from families 4 and 6, basal plasma renin activity was markedly elevated (5.7 and 3.4 times normal, respectively). Likewise, serum aldosterone concentrations were increased to 4.4 and 1.8 fold of the upper reference range, respectively (Table 2).


View this table:
[in this window]
[in a new window]

  		Table 2 Basal plasma renin activity and serum aldosterone in FRG patients

 


   Discussion
Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
References
 
In the renal proximal tubular epithelium two sodium/ glucose co-transporters are responsible for the reabsorption of filtered glucose. Among them, SGLT2 that contains 672 amino acid residues and is expressed almost exclusively in the kidney [12] is a low-affinity sodium/glucose co-transporter and responsible for the bulk of glucose reabsorption in this organ. The 664 amino acid transporter protein SGLT1, a high affinity co-transporter that is strongly expressed in the small intestine and the kidney [13], accounts for the reabsorption of the remaining glucose. These two co-transporters show significant sequence homology with 59% of amino acids being identical after alignment [12]. The SLC5A1 gene, coding for SGLT1, was found to be responsible for familial intestinal glucose–galactose malabsorption (GGM, McKusick 606824) [14], while the gene coding for SGLT2, SLC5A2, is mutated in FRG [2]. The SLC5 family comprises over 230 members, including 11 members of the human genome [15]. Most members share a common core of 13 transmembrane helices (TMH), although the number may vary among homologues as in the case of human SGLT1 and SGLT2, with 14 [16].

Among the 28 pedigrees of the two previously published cohorts, three cases had no identifiable mutation in the coding sequence of SLC5A2. Mutations in the promoter region or heterozygosity for large deletions not detectable by PCR may account for these findings. Alternatively, FRG might be a genetically heterogeneous entity. Interestingly, a locus on chromosome 6 detected by linkage analysis and termed GLYS1 has been reported [17]. In addition, other SLC5 sodium/glucose co-transporter members, like SGLT4, SGLT6 and SMIT1, are also expressed in the human kidney [15] and can be expected to play a role in glucose reabsorption. They are candidate genes for FRG in families without an SLC5A2 mutation. Our current findings, however, argue against significant genetic heterogeneity, given that in this new series of 17 consecutive families evaluated for glucosuria, SLC5A2 mutations were always found. Nevertheless, those co-transporters may still act as modifier genes and account for the incomplete penetrance described in FRG.

Among the 20 SLC5A2 mutations now identified, 15 were novel, which is evidence for remarkable allelic heterogeneity (Table 1, Figure 1). This is in agreement with previous studies suggesting that most mutations in FRG are private. Among those that are not, IVS7 + 5G>A, meanwhile reported in nine unrelated pedigrees, is the most frequent one and might be a mutational hot spot ([3], current report), followed by p.Q167RfsX20 (c.500delA; Q167fs) in three families [3,4,6]. Furthermore, the sequence variation p.N654S, originally referred to as a polymorphism since it was detected in 1 out of 80 anonymous chromosomes [3], has been reported in the compound heterozygous state in five different families ([3,4,7], current report). In this regard it is interesting that the index case from family 3, who presented with very mild glucosuria, was found to be heterozygous for p.N654S. Since this was the only sequence variation detected, this provides some evidence that p.N654S alters SGLT2 function although the last TMH, the region where this mutation is located, has been reported not to be essential for proper function of SGLTs. This transmembranous domain is missing in some members of the SGLT family, and in vitro deletion does not eliminate sugar transport [16,18]. If at all present, the impairment of transporter function, however, seems to be mild, since the index case 11-1 who is compound heterozygous for p.N654S and another missense mutation had a glucose excretion of 10 g/1.73 m2/24 h, exactly the borderline between mildly affected (typically heterozygous) and severely affected (typically homo- or compound heterozygous) individuals.

Among the 15 novel SLC5A2 mutations there was only one nonsense mutation, p.Y128X, whose pathogenic significance is obvious. Her massive glucosuria of 72.7 g/ 1.73 m2/24 h in the presence of severe polyuria is in accordance with the notion that the bulk of filtered glucose is lost with the urine if SGLT2 is non-functioning and allows the classification as so-called type 0 glucosuria [19]. Most of the other novel mutations are substitutions that affect residues that are relatively conserved among SGLT2 proteins of different species and among other members of the human SLC5 family (Figure 2). Interestingly, for SGLT1, mutations of residues R300 and R558 (numbering is identical for corresponding amino acids at positions 255 to 605 in SGLT1 and SGLT2) are associated with GGM [20]. The only exceptions are p.V116M, p.T323R and p.L539P. For p.V116M, there is some evidence that it may be a disease-causing allele, since a similar mutation, p.V105M, also in the third TMH, has been previously reported in a FRG heterozygous individual [3]. Less clear is the situation for p.T323R and p.L539P sequence variations that are neither conserved between SGLT2 and SGLT1 and 3, nor among SGLT2s of different species. When the Polyphen software (http://coot.embl.de/PolyPhen) was used, in order to evaluate the potential pathogenicity of these two amino acid substitutions, both were predicted to be possibly damaging. Nevertheless, expression studies will be required, which, however, have not been successful to date for SGLT2 proteins.

According to previous observations, polyuria is the major functional change in FRG [21]. While the impact of unabsorbed glucose is well acknowledged, the role of sodium that may escape both proximal and distal absorption has arrested less attention. Based on a single previous observation [4], we evaluated the RAAS in severe forms of FRG by an indirect assessment of volume depletion resulting from natriuresis. Results have consistently demonstrated that compensatory mechanisms are activated as a response to sodium losses. On average, and considering the three individuals that have been investigated so far, basal plasma renin activity and serum aldosterone concentration were 4.2- and 2.7-fold the upper reference range, respectively (Table 2). This is of particular interest in the context of the use of specific SGLTs inhibitors. Synthetic agents derived from phlorizin were shown to increase urinary glucose excretion, reduce blood glucose and HbA1c levels and prevent the development of microalbuminuria and neuropathy in several rodent diabetic models [22,23]. Specifically, sergliflozin and dapagliflozin, two selective SGLT2 inhibitors meanwhile developed, are being considered as a new approach to the treatment of type 2 diabetes mellitus [24,25]. For the highest dose of sergliflozin used in these studies, a temporary osmotic diuresis was reported and an increase in sodium excretion (although not statistically significant) was observed [24]. Long-term effects of electrolyte metabolism and secondary altered hormonal regulation will have to be considered not only in FRG patients but also in patients with type 2 diabetes who might be treated with SGLT2 inhibitors some time in the future.

In conclusion, we have investigated FRG index cases and some of their relatives, a total of 26 individuals, for SLC5A2 mutations. Given that at least one mutated allele was identified in each individual with FRG of this cohort, we can rule out significant genetic heterogeneity in FRG. Twenty SLC5A2 mutations were identified, including 15 novel alleles, confirming that in FRG most mutations are private. Finally, further evidence for moderate volume depletion in FRG is provided.



   Acknowledgments
 
We thank all members of affected families for their participation. Furthermore, Martina Kinner and Juliane Bergmann are kindly acknowledged for excellent technical assistance. Finally, we thank Christine Fereiro and Maria Mesquita (Bruxelles, Belgium), Peter F. Hoyer (Essen, Germany), Heinz E. Leichter (Stuttgart, Germany), Manfred Teufel (Böblingen, Germany) and Rudolph P. Wüthrich (St Gallen, Switzerland) for referring blood/DNA samples of single patients. This work was partly supported by a grant from the Portuguese Society of Nephrology.

Conflict of interest statement. We have had no involvements that might raise the question of bias in the work reported or in the conclusions, implications or opinions stated.



   References
Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 References
 

  1. Kanai Y, Lee WS, You G, et al. The human kidney low affinity Na+/glucose cotransporter SGLT2. Delineation of the major renal reabsorptive mechanism for D-glucose. J Clin Invest (1994) 93:397–404.[Web of Science][Medline]
  2. Santer R, Kinner M, Schneppenheim M, et al. The molecular basis of renal glucosuria: mutations in the gene for a renal glucose transporter (SGLT2). J Inherit Metab Dis (2000) 23(Suppl_1):178.
  3. Santer R, Kinner M, Lassen CL, et al. Molecular analysis of the SGLT2 gene in patients with renal glucosuria. J Am Soc Nephrol (2003) 14:2873–2882.[Abstract/Free Full Text]
  4. Calado J, Loeffler J, Sakallioglu O, et al. Familial renal glucosuria: SLC5A2 mutation analysis and evidence of salt-wasting. Kidney Int (2006) 69:852–855.[CrossRef][Web of Science][Medline]
  5. van den Heuvel LP, Assink K, Willemsen M, et al. Autosomal recessive renal glucosuria attributable to a mutation in the sodium glucose cotransporter (SGLT2). Hum Genet (2002) 111:544–547.[CrossRef][Web of Science][Medline]
  6. Calado J, Soto K, Clemente C, et al. Novel compound heterozygous mutations in SLC5A2 are responsible for autosomal recessive renal glucosuria. Hum Genet (2004) 114:314–316.[CrossRef][Web of Science][Medline]
  7. Kleta R, Stuart C, Gill FA, et al. Renal glucosuria due to SGLT2 mutations. Mol Genet Metab (2004) 82:56–58.[CrossRef][Web of Science][Medline]
  8. Francis J, Zhang J, Farhi A, et al. A novel SGLT2 mutation in a patient with autosomal recessive renal glucosuria. Nephrol Dial Transplant (2004) 19:2893–2895.[Free Full Text]
  9. Magen D, Sprecher E, Zelikovic I, et al. A novel missense mutation in SLC5A2 encoding SGLT2 underlies autosomal-recessive renal glucosuria and aminoaciduria. Kidney Int (2005) 67:34–41.[CrossRef][Web of Science][Medline]
  10. den Dunnen JT, Antonarakis SE. Nomenclature for the description of human sequence variations. Hum Genet (2001) 109:121–124.[CrossRef][Web of Science][Medline]
  11. Povey S, Lovering R, Bruford E, et al. The HUGO Gene Nomenclature Committee (HGNC). Hum Genet (2001) 109:678–680.[CrossRef][Web of Science][Medline]
  12. Wells RG, Pajor AM, Kanai Y, et al. Cloning of a human kidney cDNA with similarity to the sodium-glucose cotransporter. Am J Physiol Renal Physiol (1992) 263:F459–F465.[Abstract/Free Full Text]
  13. Pajor AM, Wright EM. Cloning and functional expression of a mammalian Na+/nucleoside cotransporter, a member of the SGLT family. J Biol Chem (1992) 267:3557–3560.[Abstract/Free Full Text]
  14. Turk E, Zabel B, Mundlos S, et al. Glucose/galactose malabsorption caused by a defect in the Na+/glucose cotransporter. Nature (1991) 350:354–356.[CrossRef][Medline]
  15. Wright EM, Hirayama BA, Loo DF. Active sugar transport in health and disease. J Intern Med (2007) 261:32–43.[CrossRef][Web of Science][Medline]
  16. Turk E, Wright EM. Membrane topology motifs in the SGLT cotransporter family. J Membrane Biol (1997) 159:1–20.[CrossRef][Web of Science][Medline]
  17. de Marchi S, Cecchin E, Basile A, et al. Close genetic linkage between HLA and renal glycosuria. Am J Nephrol (1984) 4:280–286.[Web of Science][Medline]
  18. Wright EM, Loo DD, Panayotova-Heiermann M, et al. Structure and function of the Na+/glucose cotransporter. Acta Physiol Scand (1998) 643(Suppl):257–264.
  19. Oemar BS, Byrd DJ, Brodehl J. Complete absence of tubular glucose reabsorption: a new type of renal glucosuria (type 0). Clin Nephrol (1987) 27:156–160.[Web of Science][Medline]
  20. Wright EM, Turk E, Martin MG. Molecular basis for glucose-galactose malabsorption. Cell Biochem Biophys (2002) 36:115–121.[CrossRef][Web of Science][Medline]
  21. Scholl-Burgi S, Santer R, Ehrich JH. Long-term outcome of renal glucosuria type 0: the original patient and his natural history. Nephrol Dial Transplant (2004) 19:2394–2396.[Free Full Text]
  22. Oku A, Ueta K, Arakawa K, et al. T-1095, an inhibitor of renal Na+-glucose cotransporters, may provide a novel approach to treating diabetes. Diabetes (1999) 48:1794–1800.[Abstract]
  23. Ueta K, Ishihara T, Matsumoto Y. Long-term treatment with the Na+-glucose cotransporter inhibitor T-1095 causes sustained improvement in hyperglycemia and prevents diabetic neuropathy in Goto-Kakizaki Rats. Life Sci (2005) 76:2655–2668.[CrossRef][Web of Science][Medline]
  24. Katsuno K, Fujimori Y, Takemura Y, et al. Sergliflozin, a novel selective inhibitor of low-affinity sodium glucose cotransporter (SGLT2), validates the critical role of SGLT2 in renal glucose reabsorption and modulates plasma glucose level. J Pharmacol Exp Ther (2007) 320:323–330.[Abstract/Free Full Text]
  25. Han S, Hagan DL, Taylor JR, et al. Dapagliflozin, a selective SGLT2 inhibitor, improves glucose homeostasis in normal and diabetic rats. Diabetes (2008) 57:1723–1729.[CrossRef][Web of Science][Medline]
Received for publication: 15. 4.08
Accepted in revised form: 18. 6.08


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
23/12/3874    most recent
gfn386v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Calado, J.
Right arrow Articles by Santer, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Calado, J.
Right arrow Articles by Santer, R.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?