Advertisement

Diabetologia

, Volume 62, Issue 2, pp 286–291 | Cite as

A variant of the glucose transporter gene SLC2A2 modifies the glycaemic response to metformin therapy in recently diagnosed type 2 diabetes

  • Wolfgang RathmannEmail author
  • Klaus Strassburger
  • Brenda Bongaerts
  • Oliver Kuss
  • Karsten Müssig
  • Volker Burkart
  • Julia Szendroedi
  • Jörg Kotzka
  • Birgit Knebel
  • Hadi Al-Hasani
  • Michael Roden
  • for the GDS Group
Short Communication

Abstract

Aims/hypothesis

The aim of this study was to investigate the modifying effect of the glucose transporter (GLUT2) gene SLC2A2 (rs8192675) variant on the glycaemic response to metformin in individuals recently diagnosed with type 2 diabetes.

Methods

Individuals with type 2 diabetes (n = 508) from the prospective German Diabetes Study (age [mean ± SD] 53 ± 10 years; 65% male; BMI 32 ± 6 kg/m2, metformin use 57%) underwent detailed metabolic characterisation (hyperinsulinaemic–euglycaemic clamp, IVGTT) during the first year after diagnosis. Participants provided self-reported data from the time of diagnosis. The change in fasting glucose was assessed in relation to SLC2A2 genotype and glucose-lowering treatment using two-way ANCOVA with gene×treatment interactions adjusted for age, sex, BMI and diabetes duration.

Results

The C variant allele of rs8192675 was associated with a higher prevalence of diabetes symptoms at diabetes diagnosis. In the metformin monotherapy group only, patients with a C allele showed a larger adjusted blood glucose reduction during the first year after diabetes diagnosis than patients with the TT genotype (6.3 mmol/l vs 3.9 mmol/l; genotype difference 2.4 mmol/l, p = 0.02; p value for genotype interaction [metformin monotherapy vs non-pharmacological therapy] <0.01). The greater decline in fasting glucose (CC/CT vs TT) in metformin monotherapy persisted after further adjusting for glucose values at diagnosis (genotype difference 1.0 mmol/l, p = 0.01; genotype×treatment interaction p = 0.06).

Conclusions/interpretation

The variant rs8192675 in the SLC2A2 gene (C allele) is associated with an improved glucose response to metformin monotherapy during the first year after diagnosis in type 2 diabetes.

Trial registration

ClinicalTrials.gov NCT01055093

Keywords

Gene interaction Metformin Personalised medicine Type 2 diabetes 

Abbreviations

GDS

German Diabetes Study

iAUC

C-peptide secretion incremental area under the curve (IVGTT)

M value

Insulin sensitivity (from the hyperinsulinaemic–euglycaemic clamp)

NPT

Non-pharmacological therapy

SNP

Single-nucleotide polymorphism

Introduction

The glucose transporter GLUT2 is expressed in the liver, intestine, kidney, pancreatic beta cells and in the central nervous system [1]. GLUT2 variants are associated with increased hyperglycaemia and type 2 diabetes development [2, 3]. The Metformin Genetics (MetGen) Consortium has shown that the single-nucleotide polymorphism (SNP) rs8192675 in the intron of the SLC2A2 gene encoding GLUT2 has both pre- and post-treatment effects on HbA1c in type 2 diabetes patients on metformin treatment [4]. The C variant allele was associated with higher pre-treatment HbA1c levels [4]. The effect of the variant allele was subsequently influenced by type of glucose-lowering treatment, with lower on-treatment HbA1c values on metformin, but not on the sulfonylurea class of drugs [4].

It is unclear whether the improved response of HbA1c to metformin in the presence of the C variant can be explained by a lower hepatic glucose production or by improved glucose disposal to the liver or peripheral tissues [5].

The aim of the present study was to investigate the interaction between the C allele of the SLC2A2 gene and metformin therapy in the reduction of hyperglycaemia in patients with recently diagnosed type 2 diabetes.

Methods

Study population and measures

The German Diabetes Study (GDS) is a cohort study investigating the natural history of recent-onset diabetes in individuals aged 18 to 69 years [6] and has been approved by the ethics committee of Heinrich Heine University, Düsseldorf, Germany. All participants provided written informed consent. Within the first year after diagnosis, patients undergo detailed metabolic phenotyping, including the Botnia clamp, a 1 h IVGTT followed by a hyperinsulinaemic–euglycaemic clamp [6].

A structured interview was carried out to assess symptoms at the time of diabetes diagnosis and the patient’s current glucose-lowering medication. Glucose-stimulated total incremental AUC (iAUC) for C-peptide during the 1 h IVGTT was calculated using the trapezoidal rule, subtracting basal C-peptide from the AUC. Glucose disappearance rate (%/min) represents the net elimination rate after glucose injection and is calculated as the slope of the interval from 6 to 40 min. Whole-body insulin sensitivity (M value) is given as space-corrected mean glucose infusion rate during steady state of the clamp divided by body weight in (kg) [6].

Blood glucose concentration was measured by the hexokinase method using an EPOS 5060 Analyser (Eppendorf, Germany). C-peptide concentrations were analysed using a Siemens Immulite Analyser (Eschborn, Germany). HbA1c was determined by the DCCT method using HPLC (Bio-Rad Laboratories, Munich, Germany). SLC2A2 genotyping was conducted using real-time polymerase chain reaction-based allelic discrimination with predeveloped gene-specific assays (Thermofisher, Darmstadt, Germany).

Statistical analyses

Data are presented as mean (SD), median (interquartile range) or percentages. Variables with skewed distributions were loge-transformed. Logistic and linear multiple regression adjusted for potential confounders (age, sex, BMI and diabetes duration) were used to compare treatment and genotype groups, respectively. Age- and sex-adjusted least squares means were estimated to calculate adjusted genotype effects within each treatment group.

The modifying effect of the SLC2A2 genotype on the association between glucose-lowering treatment and reduction of fasting blood glucose levels, including interaction terms for SLC2A2 genotype and glucose-lowering therapy, was analysed in a two-way ANCOVA adjusting for age, sex, BMI, diabetes duration and self-reported fasting glucose at diagnosis. A p value <5% was considered to be statistically significant. Analyses were performed with SAS (version 9.4; SAS Institute, Cary, NC, USA).

Results

Clinical and genetic characteristics

A total of 508 individuals with type 2 diabetes (age [mean ± SD]: 53 ± 10 years; 65% males; BMI: 32 ± 6 kg/m2) were genotyped (rs8192675 C allele carriers: allele frequency 24%; exact Hardy–Weinberg p value 0.71). Overall, 227 (45%) were treated with metformin monotherapy, 61 (12%) received metformin in combination with other glucose-lowering drugs (mostly dipeptidyl peptidase-4 inhibitors) and 220 (43%) did not receive pharmacological therapy (NPT). The clinical characteristics of participants at time of diagnosis and at baseline (study visit) stratified by genotype are shown in Table 1. Self-reported data on BMI and fasting glucose at diagnosis was not available for all patients. However, the characteristics of patients with and without self-reported data were similar (data not shown).
Table 1

Clinical characteristics of newly diagnosed type 2 diabetes patients in the GDS according to SLC2A2 genotype (TT vs TC/CC)

Clinical characteristics

TT

TC/CC

p value

CC/TC vs TT

n

292

216

Sex (% male)

66

65

0.840

At time of diabetes diagnosis (self-reported)

  Age (years)

53 ± 10

53 ± 11

0.860

  BMI (kg/m2) (n = 482)

33 ± 6

34 ± 7

0.206

  Fasting blood glucose (FBG) (mmol/l) (n = 255)

10.9 ± 5.1

11.9 ± 5.8

0.221

  Symptoms (at diabetes diagnosis):

   

    Polyuria (%)

25

35

0.022*

    Nocturia (%)

21

28

0.097

    Increased thirst (%)

29

38

0.033*

    Blurred vision (%)

15

21

0.084

    Weight loss (%)

11

17

0.053

At time of baseline study visit (structured interview and physical examination)

  Days since diabetes diagnosis

182 ± 98

175 ± 96

0.479

  BMI (kg/m2)

31 ± 6

32 ± 6

0.220

  FBG, baseline investigation (mmol/l) (n = 508)

7.1 ± 1.6

7.0 ± 1.6

0.282

  FBG, baseline investigation (mmol/l) (subgroup with self-report at diagnosis: n = 255)

7.3 ± 1.8

7.0 ± 1.6

0.107

  Improvement in FBG (FBG diagnosis − FBG baseline) (mmol/l) (n = 225)

3.6 ± 5.1

4.9 ± 5.7

0.079

  HbA1c, baseline (%) (mmol/mol)

6.4 ± 0.9 (46)

6.4 ± 0.9 (46)

0.584

  M value (mg [kg body weight]−1 min−1)

6.0 (4.3, 7.6)

6.1 (4.3, 7.8)

0.331

  C-peptide (iAUC)a; (nmol/l × min)

35 (23, 50)

35 (22, 56)

0.670

  Glucose disappearance rateb (%/min)

0.79 ± 0.27

0.82 ± 0.28

0.261

  Metformin user (%)

55

59

0.520

  Metformin dosage (mg/day) (users only)

(1515 ± 706)

(1462 ± 582)

0.320

Data are percentages, mean ± SD or median (25th; 75th percentile)

Blood glucose values, BMI and symptoms at diabetes diagnosis are based on self-reports. The p values were adjusted for age, sex, BMI and diabetes duration

Conversion factor for glucose (mg/dl to mmol/l): 0.0555

aThe iAUC is for 1 h IVGTT

bIVGTT

*p < 0.05

Participants with the TC and CC genotypes more frequently had diabetes-related symptoms at time of diagnosis (p < 0.05) (Table 1). No significant differences were found for age, sex, BMI, fasting glucose at diagnosis, baseline fasting glucose, HbA1c or metabolic measures (M value, C-peptide iAUC, glucose disappearance rate) between the genotype groups.

Association between genotype and metabolic variables

Figure 1 shows the age- and sex-adjusted change in fasting glucose between diabetes diagnosis and baseline stratified by genotype and treatment groups. Patients on combination therapy had the largest change in glucose values followed by those on metformin monotherapy and non-pharmacological therapy. In metformin monotherapy only, patients with a C allele showed a larger blood glucose reduction than patients with the TT genotype (6.2 mmol/l vs 3.9 mmol/l; genotype difference: 2.3 mmol/l; p = 0.02; p value for genotype interaction (metformin monotherapy vs NPT) = 0.01) (Fig. 1a).
Fig. 1

Change in (a) fasting glucose (from diabetes diagnosis until the baseline investigation during first year after diagnosis); (b) glucose disappearance rate (baseline); and (c) C-peptide secretion (iAUC at baseline) by SLC2A2 genotype and treatment in patients with recently diagnosed type 2 diabetes in the GDS according to least squares means (95% CI) adjusted for age and sex. The key in (a) applies to all parts. Genotype×treatment interaction for metformin monotherapy vs NPT: (a) p=0.01, (b) p=0.11, (c) p=0.08

Glucose disappearance rate during the IVGTT was highest in the NPT group, followed by the metformin monotherapy and combination therapy groups (Fig. 1b). When analysed according to genotype, only the metformin monotherapy group showed a statistically significant difference (0.82%/min vs 0.75%/min; genotype difference: +0.07%/min; p = 0.04; p value for genotype interaction [metformin monotherapy vs NPT] = 0.11). C-peptide secretion was also highest in the NPT group, followed by the metformin monotherapy and combination therapy groups. C-peptide secretion was significantly higher in patients with TC and CC genotypes than in those with the TT genotype in the metformin monotherapy group (36 vs 29 iAUC [nmol/l × min]; relative change +25%, p = 0.04; p value for genotype interaction [metformin monotherapy vs NPT] = 0.08), but not in the other groups (Fig. 1c).

After adjusting for age, sex, BMI and diabetes duration, two-way ANCOVA confirmed that the C allele of the SLC2A2 gene was associated with a reduction in fasting blood glucose in the metformin monotherapy group (genotype difference: 2.4 mmol/l, p = 0.02; genotype×treatment interaction p < 0.01). After additionally adjusting for self-reported fasting glucose at diagnosis, the effect diminished, but was still statistically significant (genotype difference: 1.0 mmol/l; p = 0.01; genotype×treatment interaction p = 0.06).

In the fully adjusted model, there was no difference between genotypes within the metformin monotherapy group for the glucose disappearance rate (genotype difference +0.07%/min, p = 0.05; genotype×treatment interaction 0.11). In fully adjusted analyses, C allele carriers in the metformin monotherapy group also showed no difference in C-peptide secretion (relative change: +21%, p = 0.07; genotype×treatment interaction p = 0.14). Finally, there were no interactions of genotype and glucose-lowering therapy with regard to measures of insulin sensitivity (M value) or baseline HbA1c (data not shown).

Discussion

This study shows that the SNP rs8192675 in the SLC2A2 gene (C allele) was associated with an improved glucose response to metformin monotherapy in patients with newly diagnosed type 2 diabetes. These findings are in line with a three-stage genome-wide association study including 13,123 type 2 diabetes patients from the MetGen Consortium, which reported that having a C allele was related to higher baseline HbA1c values independent of the type of glucose-lowering therapy, and to a significantly larger on-treatment HbA1c reduction for metformin users only [4].

SLC2A2 encodes the glucose transporter isoform GLUT2, which is expressed in liver, kidney, intestine, pancreatic islet beta cells and the central nervous system (neurons, astrocytes) [1]. The glucose-lowering effect of metformin is believed to result mainly from decreased hepatic glucose output through inhibition of gluconeogenesis, but also from increased muscle glucose uptake [7]. Furthermore, metformin may also act through alterations in the gut microbiome [7]. Gene expression data from 1226 human liver samples revealed that SLC2A2 expression is decreased in people with the C allele, which most likely results in less GLUT2 activity in the liver [4]. This genetic alteration may be positively modulated by metformin. A possible explanation for the present finding could be that glucose clearance is reduced in C allele carriers, including a decreased ability of glucose to enter the liver, which is improved by metformin therapy. Furthermore, GLUT2 is also responsible for hepatic glucose release [8]. It is conceivable that people with the C allele variant of the SLC2A2 gene may be more sensitive to the effects of metformin on the liver.

In early type 2 diabetes, insulin secretion is commonly increased because of the prevailing insulin resistance [9]. Over time, insulin secretion begins to decline and glycaemic control worsens [9].

In the present study, neither C-peptide stimulation nor glucose disappearance rate was increased in C allele carriers with metformin monotherapy. It is noteworthy that is has previously been reported that metformin improves glucose disappearance as assessed by IVGTT in women with polycystic ovary syndrome [10].

Our results should be interpreted in the context of some limitations. First, data on diabetes-related symptoms and fasting glucose values at time of diabetes diagnosis relied on self-reports only. Second, metformin blood concentrations were not measured. Third, while the primary endpoint was the change of blood glucose from diagnosis to baseline, for various other variables there were no data available at time of diabetes diagnosis. Therefore, the corresponding tests and p values should be interpreted as hypotheses-generating.

In conclusion, in this study we found a modifying effect of the SLC2A2 gene on reduction of fasting glucose by metformin monotherapy in patients with recently diagnosed type 2 diabetes.

Notes

Acknowledgements

The authors appreciate the voluntary contribution of all study participants. They also thank the staff of the Clinical Research Center of the German Diabetes Center (DDZ) for excellent technical assistance and taking care of the patients.

Some of the data were presented at the 53rd Annual Conference of the DDG (German Diabetes Society) in 2018.

Contribution statement

WR designed the study and wrote the manuscript. KS designed the study and analysed the data. BK, JK and HAH carried out the genetic analyses. All authors acquired and interpreted the data, critically reviewed the manuscript, provided technical or material support and approved the final version to be published. WR and KS are the guarantors of this work and, as such, had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Funding

The GDS was initiated and is performed by the German Diabetes Center (DDZ), which is funded by the German Federal Ministry of Health and the Ministry of Innovation, Science, Research and Technology of the State North Rhine-Westphalia. The GDS study is supported by a grant from the German Federal Ministry of Education and Research (BMBF) to the German Center for Diabetes Research (DZD e.V.).

Duality of interest

WR reports receiving consulting fees for attending educational sessions or advisory boards from AstraZeneca, Boehringer Ingelheim and NovoNordisk and institutional research grants from NovoNordisk. MR reports receiving consulting fees from Boehringer Ingelheim, Eli Lilly, Merck, NovoNordisk, Poxel and Sanofi. No other potential conflicts of interest relevant to this article were reported.

References

  1. 1.
    Thorens B (2015) GLUT2, glucose sensing and glucose homeostasis. Diabetologia 58(2):221–232.  https://doi.org/10.1007/s00125-014-3451-1 CrossRefPubMedGoogle Scholar
  2. 2.
    Soranzo N, Sanna S, Wheeler E et al (2010) Common variants at 10 genomic loci influence hemoglobin A1c levels via glycemic and nonglycemic pathways. Diabetes 59(12):3229–3239.  https://doi.org/10.2337/db10-0502 CrossRefPubMedPubMedCentralGoogle Scholar
  3. 3.
    Scott RA, Lagou V, Welch RP et al (2012) Large-scale association analyses identify new loci influencing glycemic traits and provide insight into the underlying biological pathways. Nat Genet 44(9):991–1005.  https://doi.org/10.1038/ng.2385 CrossRefPubMedPubMedCentralGoogle Scholar
  4. 4.
    Zhou K, Yee SW, Seiser EL et al (2016) Variation in the glucose transporter gene SLC2A2 is associated with glycemic response to metformin. Nat Genet 48(9):1055–1059.  https://doi.org/10.1038/ng.3632 CrossRefPubMedPubMedCentralGoogle Scholar
  5. 5.
    Florez JC (2017) Pharmacogenetics in type 2 diabetes: precision medicine or discovery tool? Diabetologia 60(5):800–807.  https://doi.org/10.1007/s00125-017-4227-1 CrossRefPubMedGoogle Scholar
  6. 6.
    Szendroedi J, Saxena A, Weber KS et al (2016) Cohort profile: the German Diabetes Study (GDS). Cardiovasc Diabetol 15(1):59.  https://doi.org/10.1186/s12933-016-0374-9 CrossRefPubMedPubMedCentralGoogle Scholar
  7. 7.
    Rena G, Hardie DG, Pearson ER (2017) The mechanisms of action of metformin. Diabetologia 60(9):1577–1585.  https://doi.org/10.1007/s00125-017-4342-z CrossRefPubMedPubMedCentralGoogle Scholar
  8. 8.
    Liang X, Giacomini KM (2017) Transporters involved in metformin pharmacokinetics and treatment response. J Pharm Sci 106(9):2245–2250.  https://doi.org/10.1016/j.xphs.2017.04.078 CrossRefPubMedGoogle Scholar
  9. 9.
    Ferrannini E, Mari A (2014) β-Cell function in type 2 diabetes. Metabolism 63(10):1217–1227.  https://doi.org/10.1016/j.metabol.2014.05.012 CrossRefPubMedGoogle Scholar
  10. 10.
    Pau CT, Keefe C, Duran J, Welt CK (2014) Metformin improves glucose effectiveness, not insulin sensitivity: predicting treatment response in women with polycystic ovary syndrome in an open-label, interventional study. J Clin Endocrinol Metab 99(5):1870–1878.  https://doi.org/10.1210/jc.2013-4021 CrossRefPubMedPubMedCentralGoogle Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Authors and Affiliations

  • Wolfgang Rathmann
    • 1
    • 2
    Email author
  • Klaus Strassburger
    • 1
    • 2
  • Brenda Bongaerts
    • 1
    • 2
  • Oliver Kuss
    • 1
    • 2
    • 3
  • Karsten Müssig
    • 2
    • 4
    • 5
  • Volker Burkart
    • 2
    • 4
  • Julia Szendroedi
    • 2
    • 4
    • 5
  • Jörg Kotzka
    • 2
    • 6
  • Birgit Knebel
    • 2
    • 6
  • Hadi Al-Hasani
    • 2
    • 6
  • Michael Roden
    • 2
    • 4
    • 5
  • for the GDS Group
  1. 1.Institute of Biometrics and Epidemiology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich Heine UniversityDüsseldorfGermany
  2. 2.German Center for Diabetes Research (DZD)München-NeuherbergGermany
  3. 3.Institute of Medical Statistics, Heinrich Heine University, Medical FacultyDüsseldorfGermany
  4. 4.Institute for Clinical Diabetology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich Heine UniversityDüsseldorfGermany
  5. 5.Division of Endocrinology and Diabetology, Medical FacultyHeinrich Heine UniversityDüsseldorfGermany
  6. 6.Institute for Clinical Biochemistry and Pathobiochemistry, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich Heine UniversityDüsseldorfGermany

Personalised recommendations