Storage of food-derived glucose as muscle and liver glycogen is vital to avoid damaging swings plasma osmolarity. We report that in Type 2 diabetes, muscle glycogen storage was completely inactive although liver glycogen storage was normal.

CLINICAL PERSPECTIVES

  • This study aimed to quantify the extent of day-long storage of food-derived glucose as muscle and liver glycogen in Type 2 diabetic and matched normoglycaemic individuals.

  • Muscle glycogen storage was completely inactive in Type 2 diabetes whereas liver glycogen storage was normal.

  • The data define the extent of postprandial glucose storage as glycogen in health and define an important contributor to postprandial hyperglycaemia.

INTRODUCTION

Normal control of postprandial plasma glucose depends upon timely storage of meal-derived glucose as glycogen. In young healthy humans, approximately 30% of carbohydrate absorbed is stored as skeletal muscle glycogen in the immediate postprandial period [1] and approximately 20% as liver glycogen [2]. If no food is taken over the next 7 h, glycogen levels in both muscle and liver decline, as glucose carbon is glycolysed to lactate and further metabolized. However, when meals are taken at intervals of 4–5 h during the day, there is a sequential rise in muscle and liver glycogen in healthy volunteers [3,4]. Tissue glycogen levels would be expected to be maximal after the evening meal prior to the overnight fall to fasting levels.

These data suggest that muscle and liver act to buffer the plasma changes in glucose concentration after meals by storing the osmotically active metabolite in an inert form, hence preventing large postprandial swings in plasma glucose concentration. If tissue uptake did not occur, a meal containing 80 g of carbohydrate can be calculated to bring about a rise in plasma glucose of 32 mmol/l for a 70 kg person. In Type 2 diabetes, inadequate glucose storage as glycogen has been previously reported, and the postprandial rise in plasma glucose is excessive [46]. However, the extent to which the buffering function of glycogen storage in muscle and liver is abnormal during day-long eating in individuals with Type 2 diabetes has not been defined.

The purpose of this study was to quantify the maximum daily flux of muscle and liver glycogen in people with well-controlled Type 2 diabetes compared with matched glucose-tolerant controls. In order to understand the interrelationship with intramyocellular and liver triacylglycerol concentration, these parameters were also measured.

METHODS

Subjects

We studied 40 (25 males; 15 females) well-controlled Type 2 diabetes subjects with an HbA1c level of 6.4±0.1% (47±0.8 mmol/mol) and mean duration of 5.9±0.8 years. All were taking metformin and no other oral hypoglycaemic agent or insulin. Fourteen (8 males; 6 females) normal glucose tolerance subjects matched for age, weight and body mass index (BMI), with no first-degree relative with diabetes were recruited as controls to provide comparative data. Normal glucose tolerance was demonstrated in all control subjects by normal response to a 75 g oral glucose tolerance test (fasting plasma glucose 5.3 mmol compared with 2 h 5.5 mmol). The characteristics of both groups are shown in Table 1. Informed written consent was obtained from all volunteers. The study was approved by the Newcastle and North Tyneside 2 Research Ethics Committee.

Table 1
Characteristics of study subjects
Type 2 diabetes groupControl
Age (years) 61.8±1.0 59.0±2.2 
Weight (kg) 87.7±2.1 86.7±3.7 
BMI (kg/m230.0±0.5 29.6±1.0 
Body fat (%) 31.6±1.2 27.9±2.6 
Waist/hip ratio (WHR) 0.95±0.01 0.90±0.02* 
Fasting plasma glucose (mmol/l) 7.7±0.2 5.1± 0.1** 
4 h post third meal plasma glucose (mmol/l) 6.5±0.2 5.7±0.2* 
HbA1c (%) 6.4±0.1 – 
HbA1c (mmol/mol) 46.5±0.8 – 
Fasting plasma insulin (mU/l) 13.8±1.2 7.9±0.9** 
4 h post third meal plasma insulin (mU/l) 31.9±2.9 18.8±3.3 
HOMA-IR (μU/mol/l35.0±0.4 1.9±0.2** 
HOMA-β (%) 72.0±7.1 97.3.0±9.5* 
Type 2 diabetes groupControl
Age (years) 61.8±1.0 59.0±2.2 
Weight (kg) 87.7±2.1 86.7±3.7 
BMI (kg/m230.0±0.5 29.6±1.0 
Body fat (%) 31.6±1.2 27.9±2.6 
Waist/hip ratio (WHR) 0.95±0.01 0.90±0.02* 
Fasting plasma glucose (mmol/l) 7.7±0.2 5.1± 0.1** 
4 h post third meal plasma glucose (mmol/l) 6.5±0.2 5.7±0.2* 
HbA1c (%) 6.4±0.1 – 
HbA1c (mmol/mol) 46.5±0.8 – 
Fasting plasma insulin (mU/l) 13.8±1.2 7.9±0.9** 
4 h post third meal plasma insulin (mU/l) 31.9±2.9 18.8±3.3 
HOMA-IR (μU/mol/l35.0±0.4 1.9±0.2** 
HOMA-β (%) 72.0±7.1 97.3.0±9.5* 

Note: Data are shown as means±SEM. *P ≥ 0.05; **P ≥ 0.01.

Study design

All subjects were asked to avoid vigorous exercise and take no alcohol for 3 days prior to assessment. On the morning of the first day, assessment was carried out at 08:00, after a 10-h overnight fast. Subjects were transported to the study centre by taxi. Blood samples were taken and magnetic resonance (MR) measurements of muscle and liver glycogen and triacylglycerol were made. For the second day, all food had been previously provided and subjects were asked to consume all of the meals but nothing in addition to that provided. Meals were eaten at 4 h intervals: 08:00 (carbohydrate 132.2 g; fat 12.4 g; protein 28.1 g); 12:00 (carbohydrate 133.3 g; fat 21.9 g; protein 45.3 g) and 16:00 (carbohydrate 120.1 g; fat 34.2 g; protein 38.0 g). Metformin tablets were taken 5 min before meals by the Type 2 diabetes subjects. Blood samples were taken at 20:00 and MR measurements were carried out immediately thereafter. This ensured that muscle and liver post-meal glycogen concentrations were measured 4 h after the third meal, when maximal glycogen storage from the three meals would be expected. Liver and muscle triacylglycerol concentrations were measured only in the fasting state as no diurnal fluctuation was observed in normal subjects (EYH Khoo, Stephenson MC, Leverton E, Eriksson JW, Poucher SM, Morris PG, Taylor R, MacDonald IA and Mansell P unpublished work).

Measurement of muscle and liver glycogen

Glycogen concentration was determined from the magnitude of the natural abundance signal from the C-1 carbon of glycogen at a frequency of 100.3 p.p.m. A Philips 3 T Achieva scanner (Philips Healthcare) was used with a 6 cm diameter 13C surface coil with integral 1H decoupling surface coil (PulseTeq) to measure muscle glycogen concentration and an in-house built 12 cm 13C/1H surface coil to measure liver glycogen concentration.

For muscle glycogen measurements, the surface coil was placed over the widest part of the gastrocnemius and was held in position with fabric straps to prevent movement. Pulse power was calibrated to a nominal value of 80° by observing the power-dependent variation in signal from a fiducial marker located in the coil housing, containing a sample exhibiting 13C signal with short T1 (213 mM [2-13C]acetone and 25 mM GdCl3 in water). Automated shimming was carried out to ensure that the magnetic field within the scanner was uniform over the active volume of the 13C coil. The 13C spectra were acquired over 15 min using a non-localized 1H-decoupled 13C pulse-acquire sequence (TR 120 ms, spectral width 8 kHz, 7000 averages, WALTZ decoupling). 1H decoupling was applied for 60% of the 13C signal acquisition to allow a relatively fast TR of 120 ms to be used within the Specific Absorption Rate safety limitations.

For liver glycogen measurements the 13C/1H surface coil was placed over the right lobe of the liver. Spectra were acquired over 15 min using non-localized 1H-decoupled 13C pulse acquisition sequences (TR 300 ms, spectral width 8 kHz, 2504 averages, WALTZ decoupling, nominal 13C tip angle of 80°). Scout images were obtained at the start of each study to ensure optimal coil position.

Tissue glycogen concentration was calculated from the amplitude of the C1-glycogen 13C signal using Java-based magnetic resonance user interface (jMRUI) version 3.0 and the AMARES algorithm [7]. For each subject the separation between radiofrequency (RF) coil and muscle/liver tissue was measured from 1H images, and 13C coil loading assessed from 13C flip angle calibration data. Tissue glycogen concentration was determined by comparison of glycogen signal amplitude with spectra acquired from liver- and leg-shaped phantoms filled with aqueous solutions of glycogen (100 mM) and potassium chloride (70 mM). Phantom data were acquired at a range of flip angles and separation distances between coil and phantom. Quantification of each human 13C spectrum employed a phantom dataset matched to body geometry and achieved a flip angle so that account differences in coil sensitivity profile and loading were taken into account for each subject.

Measurement of intramyocellular triacylglycerol

A 12 cm 1H transmitter/receiver coil was used to obtain 1H spectra to measure intramyocellular lipid content in muscle. The PRESS (Point Resolved Spectroscopy) [8] sequence was used to acquire 1H spectra from the gastrocnemius muscle with a 2 cm×2 cm×2 cm voxel, using an echo time of 25 ms, spectral resolution of 1 Hz and repetition time of 5000 ms with 32 acquisitions. Spectra were analysed with JMRUI version 3.0 using the least-squares fitting AMARES algorithm [7]. The inter-observer bias was 0.09 mmol/l with a 95% limit of agreement of 0.8 mmol/l (P>0.05).

Measurement of hepatic triacylglycerol

A six-channel cardiac coil (Philips Healthcare) was used to measure liver triacylglycerol content. Data were acquired using a three point Dixon method [9] with three gradient-echo scans acquired with adjacent out-of-phase and in-phase echoes during a 17-s breath hold (repetition time (TR)/echo time/averages/flip angle=50 ms/3.45, 4.60, 5.75 ms/1/5°). A matrix size of 160×109, and field view of 400–480 mm according to volunteer size were used. The fat and water contribution of the MRI images were separated using an in-house programme written in MATLAB (MathWorks), with the triacylglycerol content in the images expressed as a percentage of the total signal from fat and water in each pixel. The polygon region of interest (ROI) tool in the imaging software ImageJ (NIH; http:://imagej.nih.gov./ij/) was used to carefully draw ROIs within the homogenous liver parenchyma on five separate slices of each scan, avoiding contamination of data from blood vessels, gall bladder or any peripheral tissue so that liver fat data solely represented intrahepatic triacylglycerol. Measurements of the five slices were then averaged. The inter-scan Bland Altman repeatability for hepatic triacylglycerol in our laboratory is 0.5% [10].

Metabolite and Hormone Assay

Plasma glucose concentration was measured with a Yellow Springs glucose analyser (YSI). Plasma insulin concentration was measured with Dako insulin enzyme linked immunoabsorbent assay using a spectrophotometric analyser. Glucagon concentration was measured with a Glucagon Radioimmunoassay (RIA) kit (Millipore Corporation), which is specific for pancreatic glucagon. Plasma triacylglycerol concentration was measured with a Triacylglycerol GPO-PAP spectrophotometric assay (Roche Diagnostics), using Roche/Hitachi MODULARa analyser. HbA1c was measured by high-performance liquid chromatography (Bio-Rad Laboratories).

Insulin resistance and β-cell function were assessed using homoeostatic model assessment (HOMA-IR and HOMA-β) [11].

Data analysis

Statistical analysis was performed with GraphPad Prism 6.0d software. Data are expressed as means±S.E.M. Student's means two-tailed paired t-test or the Mann–Whitney test was used to compare groups as appropriate. Pearson correlation was used to assess the relationship between variables. Statistical significance was defined as P<0.05.

RESULTS

Muscle glycogen

Mean fasting muscle glycogen concentration did not differ between the Type 2 diabetes and control groups (68.3±2.6 vs. 68.1±4.8 mmol/l; P=0.82). Four hours after the third meal, muscle glycogen remained unchanged in the Type 2 diabetes group (68.3±2.6 to 67.1±2.0 mmol/l; P=0.62). In the control group, muscle glycogen rose by 17.3% over the same period (68.1±4.8 to 79.7± 4.2 mmol/l; P=0.0006; Figure 1). The change in diurnal muscle glycogen inversely correlated with HOMA-IR only in the control group (r=−0.56; P=0.02; Figure 2).

Diurnal change in muscle glycogen concentration

Figure 1
Diurnal change in muscle glycogen concentration

Fasting and post-three meal concentrations are shown for the normoglycaemic control and Type 2 diabetes groups.

Figure 1
Diurnal change in muscle glycogen concentration

Fasting and post-three meal concentrations are shown for the normoglycaemic control and Type 2 diabetes groups.

Correlation between HOMA-IR and change in muscle glycogen concentration in the normoglycaemic control subjects

Figure 2
Correlation between HOMA-IR and change in muscle glycogen concentration in the normoglycaemic control subjects
Figure 2
Correlation between HOMA-IR and change in muscle glycogen concentration in the normoglycaemic control subjects

Liver glycogen

Fasting liver glycogen concentration was similar in the Type 2 diabetes and control group (296.1±16.0 vs. 325.9±25.0 mmol/l; P=0.24). Four hours after the third meal, liver glycogen was higher in both groups (296.1±16.0 to 350.5±6.7 mmol/l; P<0.0001 and 325.9±25.0 to 388.1±30.3 mmol/l; P=0.005) respectively; and there was no significant difference between the groups (P=0.31). The percentage diurnal glycogen change was comparable between the Type 2 diabetes and control groups (18% vs. 19%; Figure 3). There was no correlation between liver glycogen and plasma glucose either fasting or post-meal.

Diurnal change in liver glycogen concentration

Figure 3
Diurnal change in liver glycogen concentration

Fasting and post-three meal concentrations are shown for the normoglycaemic control and Type 2 diabetes groups.

Figure 3
Diurnal change in liver glycogen concentration

Fasting and post-three meal concentrations are shown for the normoglycaemic control and Type 2 diabetes groups.

Liver and intramyocellular triacylglycerol

Liver triacylglycerol concentration in the Type 2 diabetes group was found to be approximately double that of the controls (6.7±0.7 vs. 3.4±0.8%; P=0.0008). Liver triacylglycerol concentration did not correlate with the diurnal change in liver glycogen concentration in either the Type 2 diabetes or control groups (r=0.11, P=0.36; r=−0.18, P=0.15).

Intramyocellular triacylglycerol was 38% greater in the diabetes group (22.3±2.0 vs. 16.2±1.1 mmol/l; P=0.01). The lack of change in muscle glycogen from fasting to 4 h after the third meal in the diabetes group dictated that this parameter was not related to intramyocellular lipid (IMCL) (r=0.18, P=0.18), although in the controls there was a tendency for lower IMCL to be associated with higher amplitude of diurnal change in muscle glycogen (r=−0.43, P=0.08).

Plasma glucose, metabolites and hormones

Plasma glucose concentration was higher in the Type 2 diabetes group, both fasting (7.7±0.2 vs. 5.1±0.1 mmol/l; P<0.0001) and 4 h after the third meal (6.5±0.2 vs. 5.7±0.2 mmol/l; P=0.03).

In the control group, plasma insulin was higher than fasting 4 h after the third meal (7.9±0.9 to 18.8±3.3 mU/l; P=0.003). Similarly in the Type 2 diabetes group, plasma insulin rose from 13.8±1.2 to 31.9±2.9 mU/l (P<0.0001). Plasma insulin was higher in the Type 2 diabetes group both fasting (13.8±1.2 vs. 7.9±0.9 mU/l; P=0.01) and 4 h after the third meal (31.9±2.9 vs. 18.8±3.3 mU/l; P=0.03).

Plasma glucagon was higher in the Type 2 diabetes group than in the control group both fasting (70.4±4.7 vs. 48.4±4.8 ng/l; P=0.01) and 4 h after the third meal (108.5±7.5 vs. 67.7±7.3 ng/l; P<0.001). Plasma glucagon levels rose both in the Type 2 diabetes group (70.4±4.7 to 108.5±7.5 ng/l; P<0.0001) and in the control group (48.4±4.8 to 67.7±7.3 ng/l; P=0.001) respectively.

Fasting plasma triacylglycerol levels rose both in the Type 2 diabetes (1.4±0.1 to 2.2±0.1 mmol/l; P<0.0001) and in the control groups (1.5±0.3 to 2.3±0.3 mmol/l; P=0.007) respectively. There was no significant difference between the groups at either time point. In both Type 2 diabetes and control groups, fasting triacylglycerol positively correlated with HOMA-IR (r=0.30; P=0.03); (r=0.59; P=0.02).

DISCUSSION

This study has defined the extent of daylong glycogen storage in muscle and liver during normal eating in both people with or without Type 2 diabetes. The data demonstrate that skeletal muscle does not contribute to glycogen storage in people with early well-controlled Type 2 diabetes. In contrast, day-long storage of glycogen in liver was similar between those with and without Type 2 diabetes. In normal glucose-tolerant individuals, muscle and liver were shown to act together as dynamic buffers, storing osmotically active glucose temporarily and allowing redistribution of carbon energy overnight.

Under euglycaemic hyperinsulinaemic conditions, the rate of muscle glucose uptake is subnormal in subjects with Type 2 diabetes [12]. Limb balance studies of oral glucose administration showed that the higher blood glucose concentration in diabetic subjects only partially compensated for this and overall rates of glucose uptake by muscle were observed to be up to 30% lower [6,1315]. The present data demonstrate that under everyday conditions postprandial storage of glucose as glycogen in muscle is markedly impaired in Type 2 diabetes. This observation is consistent with the subnormal rate of insulin stimulated glycogen synthesis in type 2 diabetic subjects using 13C-radiolabelled glucose [16].

Natural abundance 13C MR spectroscopy has previously been applied to measure change in muscle glycogen in Type 2 diabetes [17]. In a group of people with Type 2 diabetes who had similar BMI but less tight blood glucose control than the present experimental group, muscle glycogen did not rise significantly after two successive meals, although there appeared to be a modest upward trend. The present data demonstrate that there is no net contribution of muscle glycogen to meal-derived glucose storage during a full day of eating. The complete lack of increase in muscle glycogen following three meals is all the more striking when the ambient plasma insulin levels are considered. Both fasting and 4 h after the third meal plasma insulin levels were over 50% higher in Type 2 diabetes, illustrating the practical everyday extent of muscle insulin resistance of Type 2 diabetes.

In skeletal muscle, glucose uptake is the rate-limiting step for glycogen storage [1820]. Resistance to insulin-stimulated glucose uptake by muscle has been shown to precede the onset of Type 2 diabetes by decades [21,22]. The defect specifically involves muscle glycogen synthesis as shown by clamp studies using 13C-radiolabelled glucose infusion and MR spectroscopy [17]. As exercise training and weight loss both have the potential to ameliorate this defect, these have been intensively studied. Although 6 weeks of intensive aerobic exercise increased insulin-dependent muscle glucose uptake, this did not change relative to the non-diabetic control group [23]. Weight loss sufficient to restore durable normoglycaemia had no effect upon insulin sensitivity measured by insulin clamp [10]. The defect in muscle glycogen synthesis appears to be a central aspect of the aetiopathogenesis of Type 2 diabetes [24].

The normality of the capacity of the liver to store glycogen in Type 2 diabetes is striking in contrast with the absence of insulin stimulation of glycogen synthesis in muscle. This is likely to represent the very different glucose handling in this tissue compared with muscle. Unlike in skeletal muscle, where glucose uptake via Glut4 glucose transporters is the rate-limiting step for glycogen storage and the impact of insulin resistance on cellular glucose uptake is great [1820], there is no insulin regulation of cellular glucose uptake by the hepatocyte. Liver glucose concentration inside the hepatocyte reflects that of the extracellular fluid, and hepatic glycogen synthesis is responsive to glucose concentration such that hyperglycaemia increases the rate of this process by mass action [25]. Glucokinase has a high Km and glucose phosphorylation is largely substrate-driven. As a result, the overall rate of glucose storage in Type 2 diabetes will be expedited by the hyperglycaemia following each meal, in additional to any effect brought about by insulin. Hence the normal diurnal increment in hepatic glycogen concentration cannot be interpreted as indicating normal insulin sensitivity of this metabolic pathway. When plasma glucose is clamped, the hepatic glycogen synthesis rate has been shown to be subnormal [5]. The same paper reported decreased rates of hepatic glycogen storage following a test meal in Type 2 diabetes, and the reasons for the lack of accord with the present data must be considered. This is most likely to relate to the different groups studied. In the previously reported study [5] the Type 2 diabetic subjects (n=7) were close to normal weight (BMI 26.9 kg/m2) with an HbA1c level of 7.1% whereas in the present study the Type 2 diabetic group (n=41) had a more typical mean BMI of 30 kg/m2 and were at an earlier stage of the condition with an HbA1c level of 6.4%. The latter point is important illustrating that Type 2 diabetes develops whilst postprandial hepatic glycogen metabolism is normal.

Knowledge of the physiological mechanisms involved in substrate storage and glucose homoeostasis is central to the appreciation of normal energy metabolism in humans [6]. The present data quantify the ranges between which the glycogen concentration in muscle and liver varies during diurnal substrate ingestion. The nocturnal usage of glycogen stores needs to be considered. In liver, the stored glucose will directly contribute to the ongoing hepatic glucose production overnight by glycogenolysis. In muscle, the carbon energy cannot be exported as glucose due to the lack of glucose-6-phosphatase and is exported as lactate or pyruvate. The process of peripheral export of these three-carbon fragments and utilization for gluconeogenesis is described as Cori cycle activity [26]. Although classically this is regarded as a futile cycle, moving glucose carbon to peripheral tissues and recycling this to liver, it can be seen that the temporal shift in storage and export allows a vital function to be fulfilled. Hence, in normal health, glycogen stores in muscle and liver are built up during daytime eating and underpin the essential overnight maintenance of plasma glucose, which is essential to support brain glucose oxidation.

All of the Type 2 diabetes subjects were treated with metformin, and it must be considered whether this could have altered liver glycogen concentration, as the therapy will have decreased fasting hepatic glucose output [27,28]. However, direct examination of this question has established that the metformin effect is mediated via inhibition of gluconeogenesis [29]. The drug has no effect upon muscle glycogen synthase activation by insulin [27].

In the non-diabetic controls, there was a correlation bet-ween IMCL and insulin sensitivity as quantified by HOMA, and higher IMCL tended to be associated with a lower postprandial increase in muscle glycogen as previously reported [30]. In Type 2 diabetes, IMCL was raised and there was no net increase in muscle glycogen. As a consequence there was no correlation between lipid stores and extent of diurnal glycogen cycling in muscle. In the liver, there was no relationship between lipid stores and extent of postprandial storage of glycogen in either group, most probably reflecting the lack of lipid effect upon control of cellular glucose uptake in the liver. This contrasts with the close relationship between intrahepatic lipid and insulin sensitivity of hepatic glucose production [10,31,32]. Additionally, when the mass action effect of glucose on hepatic uptake is removed a significant inverse relationship can be demonstrated during hyperglycaemic hyperinsulinaemic clamps in Type 2 diabetes and control subjects. The difference between this experimental fixed condition, designed to give insight into mechanisms, and the everyday condition investigated in the present study is important to appreciate.

Abbreviations

     
  • BMI

    body mass index

  •  
  • HbA1c

    glycated haemoglobin

  •  
  • HOMA-IR

    homoeostatic model assessment of insulin resistance

  •  
  • IMCL

    intramyocellular lipid

  •  
  • jMRUI

    Java-based magnetic resonance user interface

  •  
  • MR

    magnetic resonance

  •  
  • ROI

    region of interest

  •  
  • TR

    repetition time

AUTHOR CONTRIBUTION

M. Macauley collected the data, performed the analysis and wrote the manuscript, F.E. Smith, P.E. Thelwall and K.G. Hollingsworth developed the methodology and contributed to the manuscript, R. Taylor designed the study, contributed to the methodology and data analysis and wrote the manuscript.

We are grateful to the volunteers for their time and co-operation, to Dr J.E. Foley for helpful discussion and to Louise Ward, Tim Hodgson and Dorothy Wallace, research radiographers. Input and assistance from Dr Matthew Clemence, Philips Healthcare Clinical Science, regarding implementation of 13C MR spectroscopy methodology is gratefully acknowledged.

FUNDING

The studies were funded by an investigator initiated grant (2419T) from the Novartis International AG to study subjects without diabetes and the subjects with diabetes were studied at baseline in a contemporaneous Novartis-sponsored study (http:://ClinicalTrials.gov Identifier: NCT01356381; the tRiglycEride accumulation anD Insulin REsistanCe assessment on vildaglipTin–REDIRECT Study).

References

References
1
Taylor
 
R.
Price
 
T.B.
Katz
 
L.D.
Shulman
 
R.G.
Shulman
 
G.I.
 
Direct measurement of change in muscle glycogen concentration after a mixed meal in normal subjects
Am. J. Physiol.
1993
, vol. 
265
 (pg. 
E224
-
E229
)
[PubMed]
2
Taylor
 
R.
Magnussen
 
I.
Rothman
 
D.L.
Cline
 
G.W.
Caumo
 
A.
Cobelli
 
C.
Shulman
 
G.I.
 
Direct assessment of liver glycogen storage by 13C-nuclear magnetic resonance spectroscopy and regulation of glucose homeostasis after a mixed meal in normal subjects
J. Clin. Invest.
1996
, vol. 
97
 (pg. 
126
-
132
)
[PubMed]
3
Hwang
 
J.-H.
Perseghin
 
G.
Rothman
 
D.L.
Cline
 
G.W.
Magnusson
 
I.
Falk-Petersen
 
K.
Shulman
 
G.I.
 
Impaired net hepatic glycogen synthesis in insulin-dependent diabetic subjects during mixed meal ingestion
J. Clin. Invest.
1995
, vol. 
95
 (pg. 
783
-
787
)
[PubMed]
4
Carey
 
P.E.
Halliday
 
J.
Snaar
 
J.E.M.
Morris
 
P.G.
Taylor
 
R.
 
Direct assessment of muscle glycogen storage after mixed meals in normal and type 2 diabetic subjects
Am. J. Physiol. Endocrinol. Metab.
2003
, vol. 
284
 (pg. 
E688
-
E694
)
[PubMed]
5
Krssak
 
M.
Brehm
 
A.
Bernroider
 
E.
Anderwald
 
C.
Nowotny
 
P.
Dalla Man
 
C.
Cobelli
 
C.
Cline
 
G.W.
Shulman
 
G.I.
Waldhausl
 
W.
Roden
 
M.
 
Alterations in postprandial hepatic glycogen metabolism in type 2 diabetes
Diabetes
2004
, vol. 
53
 (pg. 
3048
-
3056
)
[PubMed]
6
Mitrakou
 
A.
Kelley
 
D.
Veneman
 
T.
Jenssen
 
T.
Pangburn
 
T.
Reilly
 
J.
Gerich
 
J.
 
Contribution of abnormal muscle and liver glucose metabolism to postprandial hyperglycemia in NIDDM
Diabetes
1990
, vol. 
39
 (pg. 
1381
-
1390
)
[PubMed]
7
Naressi
 
A.
Couturier
 
C.
Castang
 
I.
de Beer
 
R.
Graveron-Demilly
 
D.
 
Java-based graphical user interface for MRUI, a software package for quantitation of in vivo/medical magnetic resonance spectroscopy signals
Comput. Biol. Med.
2001
, vol. 
31
 (pg. 
269
-
286
)
[PubMed]
8
Bottomley
 
P.A.
 
Spatial localization in NMR spectroscopy in vivo
Ann. N.Y. Acad. Sci.
1987
, vol. 
508
 (pg. 
333
-
348
)
[PubMed]
9
Glover
 
G.H.
Schneider
 
E.
 
Three-point Dixon technique for true water/fat decomposition with B0 inhomogeneity correction
Magn. Reson. Med.
1991
, vol. 
18
 (pg. 
371
-
383
)
[PubMed]
10
Lim
 
E.L.
Hollingsworth
 
K.G.
Aribisala
 
B.S.
Chen
 
M.J.
Mathers
 
J.C.
Taylor
 
R.
 
Reversal of type 2 diabetes: normalisation of beta cell function in association with decreased pancreas and liver triacylglycerol
Diabetologia
2011
, vol. 
54
 (pg. 
2506
-
2514
)
[PubMed]
11
Matthews
 
D.R.
Hosker
 
J.P.
Rudenski
 
A.S.
Naylor
 
B.A.
Treacher
 
D.F.
Turner
 
R.
 
Homeostasis model assessment: insulin resistance and β-cell function from fasting plasma glucose and insulin
Diabetologia
1985
, vol. 
28
 (pg. 
412
-
419
)
[PubMed]
12
Basu
 
A.
Basu
 
R.
Shah
 
P.
Vella
 
A.
Johnson
 
C.M.
Jensen
 
M.
Nair
 
K.S.
Schwenk
 
W.F.
Rizza
 
R.A.
 
Type 2 diabetes impairs splanchnic uptake of glucose but does not alter intestinal glucose absorption during enteral feeding. Type 2 diabetes impairs splanchnic uptake of glucose but does not alter intestinal glucose absorption during enteral glucose feeding: additional evidence for a defect in hepatic glucokinase activity
Diabetes
2001
, vol. 
50
 (pg. 
1351
-
1362
)
[PubMed]
13
Firth
 
R.G.
Bell
 
P.M.
Marsh
 
H.M.
Hansen
 
I.
Rizza
 
R.A.
 
Postprandial hyperglycaemia in patients with non-insulin dependent diabetes mellitus
J. Clin. Invest.
1986
, vol. 
77
 (pg. 
1525
-
1532
)
[PubMed]
14
Kelley
 
D.E.
Mandarino
 
L.J.
 
Hyperglycemia normalizes insulin-stimulated skeletal muscle glucose oxidation and storage in noninsulin-dependent diabetes mellitus
J. Clin. Invest.
1990
, vol. 
86
 (pg. 
1999
-
2007
)
[PubMed]
15
Kelley
 
D.E.
Simoneau
 
J.-A.
 
Impaired FFA utilisation by skeletal muscle in NIDDM
J. Clin. Invest.
1994
, vol. 
35
 (pg. 
671
-
682
)
16
Shulman
 
G.I.
Rothman
 
D.L.
Jue
 
T.
Stein
 
P.
DeFronzo
 
R.A.
Shulman
 
R.G.
 
Quantitation of muscle glycogen synthesis in normal subjects and subjects with non-insulin dependent diabetes by 13C nuclear magnetic resonance spectroscopy
N. Engl. J. Med.
1990
, vol. 
322
 (pg. 
223
-
228
)
[PubMed]
17
Carey
 
P.E.
Gerrard
 
J.
Cline
 
G.W.
Dalla Man
 
C.
English
 
P.T.
Firbank
 
M.J.
Cobelli
 
C.
Taylor
 
R.
 
Acute inhibition of lipolysis does not affect postprandial suppression of endogenous glucose production
Am. J. Physiol. Endocrinol. Metab.
2005
, vol. 
289
 (pg. 
E941
-
E947
)
[PubMed]
18
Cline
 
G.W.
Petersen
 
K.F.
Krssak
 
M.
Shen
 
J.
Hundal
 
R.S.
Trajanoski
 
Z.
Inzucchi
 
S.
Dresner
 
A.
Rothman
 
D.L.
Shulman
 
G.I.
 
Decreased glucose transport as a cause of decreased insulin-stimulated muscle glycogen synthesis in Type 2 diabetes
N. Engl. J. Med.
1999
, vol. 
341
 (pg. 
240
-
246
)
[PubMed]
19
Roden
 
M.
Price
 
T.B.
Perseghin
 
G.
Petersen
 
K.F.
Rothman
 
D.L.
Cline
 
G.W.
Shulman
 
G.I.
 
Mechanism of free fatty acid-induced insulin resistance in humans
J. Clin. Invest.
1996
, vol. 
97
 (pg. 
2859
-
2865
)
[PubMed]
20
Yki-Jarvinen
 
H.
 
Acute and chronic effects of hyperglycaemia on glucose metabolism
Diabetologia
1990
, vol. 
33
 (pg. 
579
-
585
)
[PubMed]
21
Lillioja
 
S.
Mott
 
D.M.
Howard
 
B.V.
Bennett
 
P.H.
Yki-Jarvinen
 
H.
Freymond
 
D.
Nyomba
 
B.L.
Zurlo
 
F.
Swinburn
 
B.
Bogardus
 
C.
 
Impaired glucose tolerance as a disorder of insulin action: longitudinal and cross sectional studies in Pima Indians
N. Engl. J. Med.
1988
, vol. 
318
 (pg. 
1217
-
1225
)
[PubMed]
22
Warram
 
J.H.
Martin
 
B.C.
Krolewski
 
A.S.
Soeldner
 
J.S.
Kahn
 
C.R.
 
Slow glucose removal rate and hyperinsulinemia precede the development of Type II diabetes in the offspring of diabetic parents
Ann. Int. Med.
1990
, vol. 
113
 (pg. 
909
-
915
)
23
Perseghin
 
G.
Price
 
T.B.
Petersen
 
K.F.
Roden
 
M.
Cline
 
G.W.
Gerow
 
K.
Rothman
 
D.L.
Shulman
 
G.I.
 
Increased glucose transport-phosphorylation and muscle glycogen synthesis after exercise training in insulin-resistant subjects
N. Engl. J. Med.
1996
, vol. 
335
 (pg. 
1357
-
1362
)
[PubMed]
24
Taylor
 
R.
 
Type 2 diabetes: etiology and reversibility
Diabetes Care
2013
, vol. 
36
 (pg. 
1047
-
1055
)
[PubMed]
25
Agius
 
L.
Peak
 
M.
Alberti
 
K.G.M.M.
 
Regulation of glycogen synthesis from glucose and gluconeogenic precursors by insulin in periportal and perivenous rat hepatocytes
Biochem. J.
1990
, vol. 
266
 (pg. 
91
-
102
)
[PubMed]
26
Dunn
 
A.
Chenoweth
 
M.
Schaeffer
 
L.D.
 
Estimation of glucose turnover and the Cori cycle using glucose-6-t-14C
Biochemistry
1967
, vol. 
6
 (pg. 
6
-
11
)
[PubMed]
27
Johnson
 
A.B.
Webster
 
J.M.
Sum
 
C.F.
Heseltine
 
L.
Argyraki
 
M.
Cooper
 
B.G.
Taylor
 
R.
 
The impact of metformin therapy on hepatic glucose production and skeletal muscle glycogen synthase activity in overweight Type II diabetic patients
Metabolism
1993
, vol. 
42
 (pg. 
1217
-
1222
)
[PubMed]
28
Inzucchi
 
S.E.
Maggs
 
D.G.
Spollett
 
G.R.
Page
 
S.L.
Rife
 
F.S.
Walton
 
V.
Shulman
 
G.I.
 
Efficacy and metabolic effects of metformin and troglitazone in type II diabetes mellitus
N. Engl. J. Med.
1998
, vol. 
338
 (pg. 
867
-
872
)
[PubMed]
29
Hundal
 
R.S.
Krssak
 
M.
Dufour
 
S.
Laurent
 
D.
LeBon
 
V.
Chandramouli
 
V.
Inzucchi
 
S.E.
Schumann
 
W.C.
Petersen
 
K.F.
Landau
 
B.R.
Shulman
 
G.I.
 
Mechanism by which metformin reduces glucose production in type 2 diabetes
Diabetes
2000
, vol. 
49
 (pg. 
2063
-
2069
)
[PubMed]
30
Krssak
 
M.
Petersen
 
K.F.
Dresner
 
A.
DiPietro
 
L.
Vogel
 
S.M.
Rothman
 
D.L.
Shulman
 
G.I.
Roden
 
M.
 
Intramyocellular lipid concentrations are correlated with insulin sensitivity in humans: a H-1 NMR spectroscopy study
Diabetologia
1999
, vol. 
42
 (pg. 
113
-
116
)
[PubMed]
31
Seppala-Lindroos
 
A.
Vehkavaara
 
S.
Hakkinen
 
A.M.
Goto
 
T.
Westerbacka
 
J.
Sovijarvi
 
A.
Halavaara
 
J.
Yki-Jarvinen
 
H.
 
Fat accumulation in the liver is associated with defects in insulin suppression of glucose production and serum free fatty acids independent of obesity in normal men
J. Clin. Endocrinol. Metab.
2002
, vol. 
87
 (pg. 
3023
-
3028
)
[PubMed]
32
Petersen
 
K.F.
Dufour
 
S.
Befroy
 
D.
Lehrke
 
M.
Hendler
 
R.E.
Shulman
 
G.I.
 
Reversal of nonalcoholic hepatic steatosis, hepatic insulin resistance, and hyperglycemia by moderate weight reduction in patients with type 2 diabetes
Diabetes
2005
, vol. 
54
 (pg. 
603
-
608
)
[PubMed]

Author notes

1

Roy Taylor has served on a Novartis advisory board and has lectured at Novartis-supported events.