Effects of pioglitazone versus glimepiride exposure on hepatocellular fat content in type 2 diabetes.

AIMS: Thiazoledinediones decrease blood glucose by their insulin-sensitizing properties. Here, we examined whether pioglitazone plus nateglinide (PIO) interferes with hepatocellular lipid (HCL) content and/or improves insulin sensitivity in well-controlled non-obese patients with type 2 diabetes mellitus (T2DM). METHODS: Sixteen patients [body mass index (BMI): 28 ± 1 kg/m(2) ; HbA1c: 7.1 ± 0.6%] were studied in a randomized, double-blind, 12-week parallel group trial, whereas matched healthy humans [non-diabetic control subjects (CON), BMI: 26 ± 1 kg/m(2)] were studied once. Treatment with pioglitazone (30 mg/day) plus nateglinide (PIO arm) to control for glimepiride-induced insulin secretion was compared to treatment with glimepiride (2 mg/day) plus placebo (GLI arm). Multinuclei magnetic resonance spectroscopy (MRS) was combined with pancreatic normoglycaemic-two-step-insulin clamps and stable isotopes to assess glucose turnover, glucose transport/phosphorylation, HCL and intramyocellular lipid (IMCL) contents, non-esterified fatty acids (NEFA) and adipokines. RESULTS: At baseline, HCL was approximately 5.6-fold higher in T2DM (p < 0.05 vs. CON). This was paralleled by approximately doubled leptin : adiponectin ratios (p < 0.05). HCL decreased by approximately 39% (p < 0.05) after PIO and only tended to decrease after GLI (p = 0.12). Treatment with PIO did not affect leptin : adiponectin ratios, but slightly improved (p < 0.05) insulin-mediated NEFA suppression, which related to lower HCL. PIO further prevented the insulin-induced increase in IMCL content of soleus and tibialis anterior muscles. Peripheral and hepatic insulin sensitivity, glucose transport and glycaemic control did not change in both groups. CONCLUSION: Short-term, low-dose thiazolidendione treatment improves insulin sensitivity of lipolysis and HCL, without affecting muscle and liver insulin sensitivity. It appears that metabolic PIO action in T2DM is primarily mediated via a decline in HCL associated with greater sensitivity of lipolysis to insulin.

Impaired insulin stimulation of muscular ATP production in patients with type 1 diabetes.

OBJECTIVE: in type 2 diabetic patients and their first-degree relatives, insulin resistance (IR) is associated with impairment of insulin-stimulated myocellular glucose-6-phosphate (g6p) and unidirectional flux through ATP synthase (fATP), suggesting the presence of inherited abnormal mitochondrial oxidative fitness. We hypothesized that patients with long-standing type 1 diabetes may also exhibit insulin resistance as well as lower fATP. DESIGN: this single-centre trial was registered at ClinicalTrials.gov (NCT00481598). SUBJECTS: we included eight nonobese type 1 diabetic patients (mean diabetes duration: 17 years) with near-target glycaemic control [haemoglobin A1c (HbA1c): 6.8 ± 0.4%] during treatment with continuous subcutaneous insulin infusion pumps and eight healthy volunteers (HbA1c: 5.4 ± 0.2%) of comparable age, body mass and level of physical activity. OUTCOME MEASURES: myocellular fATP, g6p and intramyocellular lipid content (IMCL) were measured with (1) H/(31) P magnetic resonance spectroscopy during fasting and hyperinsulinaemic-euglycaemic clamp tests. RESULTS: fasting fATP, g6p and IMCL did not differ between groups. During stimulation by insulin, type 1 diabetic patients exhibited approximately 50% (P < 0.001) lower whole-body glucose disposal along with approximately 42% (P = 0.003) lower intramyocellular g6p and approximately25% (P = 0.024) lower fATP. Insulin-stimulated fATP correlated positively with whole-body insulin sensitivity (R = 0.706, P = 0.002) and negatively with HbA1c (R = -0.675, P = 0.004). CONCLUSIONS: despite documented near-target glycaemic control for 1 year, nonobese patients with long-standing type 1 diabetes can exhibit insulin resistance. This associates with lower insulin-stimulated flux through muscular ATP synthase which could result from glucose toxicity.

Acute elevation of plasma lipids does not affect ATP synthesis in human skeletal muscle.

Prolonged elevation of plasma triglycerides and free fatty acids (FFA) reduces insulin-stimulated glucose disposal and myocellular flux through ATP synthase (fATPase). However, the early effects of lipids per se on fATPase are as yet unclear. Thus, this study examined glucose disposal and fATPase during 3 h of FFA elevation in the presence of low plasma insulinemia. Euglycemic pancreatic clamps with low-dose insulin supplementation (6 mU.m body surface area(-2).min(-1)) were performed in eight healthy men with (LIP) or without (CON) lipid infusion to measure whole body glucose disposal. (31)P/(1)H magnetic resonance spectroscopy of calf muscle was applied to quantify fATPase and concentrations of glucose 6-phosphate (G6P), inorganic phosphate (P(i)), phosphocreatine (PCr), ADP, pH, and IMCL before and during the clamps. Lipid infusion increased plasma FFA approximately twofold and decreased glucose disposal by approximately 50% (110-180 min: LIP 0.87 +/- 0.45 vs. CON 1.75 +/- 0.42 mg.kg(-1).min(-1), P = 0.002; means +/- SD). Intramyocellular G6P tended to rise only under control conditions, whereas PCr, ADP, pH, and IMCL remained unchanged from fasting in LIP and CON. Although P(i) concentrations increased by approximately 18%, fATPase remained unchanged from fasting during the clamps (LIP 10.2 +/- 2.2 vs. CON 10.5 +/- 2.6 micromol.g muscle(-1).min(-1), P = not significant). We conclude that 3 h of lipid elevation fail to affect ATP synthesis despite marked reduction of whole body glucose uptake. This suggests that lipid-induced insulin resistance results primarily from mechanisms decreasing glucose uptake rather than from direct interference of fatty acid metabolites with mitochondrial function.

Assessment of (31)P relaxation times in the human calf muscle: a comparison between 3 T and 7 T in vivo.

Phosphorus ((31)P) T(1) and T(2) relaxation times in the resting human calf muscle were assessed by interleaved, surface coil localized inversion recovery and frequency-selective spin-echo at 3 and 7 T. The obtained T(1) (mean +/- SD) decreased significantly (P < 0.05) from 3 to 7 T for phosphomonoesters (PME) (8.1 +/- 1.7 s to 3.1 +/- 0.9 s), phosphodiesters (PDE) (8.6 +/- 1.2 s to 6.0 +/- 1.1 s), phosphocreatine (PCr) (6.7 +/- 0.4 s to 4.0 +/- 0.2 s), gamma-NTP (nucleotide triphosphate) (5.5 +/- 0.4 s to 3.3 +/- 0.2 s), alpha-NTP (3.4 +/- 0.3 s to 1.8 +/- 0.1 s), and beta-NTP (3.9 +/- 0.4 s to 1.8 +/- 0.1 s), but not for inorganic phosphate (Pi) (6.9 +/- 0.6 s to 6.3 +/- 1.0 s). The decrease in T(2) was significant for Pi (153 +/- 9 ms to 109 +/- 17 ms), PDE (414 +/- 128 ms to 314 +/- 35 ms), PCr (354 +/- 16 ms to 217 +/- 14 ms), and gamma-NTP (61.9 +/- 8.6 ms to 29.0 +/- 3.3 ms). This decrease in T(1) with increasing field strength of up to 62% can be explained by the increasing influence of chemical shift anisotropy on relaxation mechanisms and may allow shorter measurements at higher field strengths or up to 62% additional signal-to-noise ratio (SNR) per unit time. The fully relaxed SNR increased by +96%, while the linewidth increased from 6.5 +/- 1.2 Hz to 11.2 +/- 1.9 Hz or +72%. At 7 T (31)P-MRS in the human calf muscle offers more than twice as much SNR per unit time in reduced measurement time compared to 3 T. This will facilitate in vivo (31)P-MRS of the human muscle at 7 T.

Three-dimensional high-resolution magnetic resonance spectroscopic imaging for absolute quantification of 31P metabolites in human liver.

Liver dysfunction correlates with alterations of intracellular concentrations of (31)P metabolites. Localization and absolute quantification should help to trace regional hepatic metabolism. An improved protocol for the absolute quantification of (31)P metabolites in vivo in human liver was developed by employing three-dimensional (3D) k-space weighted spectroscopic imaging (MRSI) with B(1)-insensitive adiabatic excitation. The protocol allowed for high spatial resolution of 17.8 +/- 0.22 cm(3) in 34 min at 3 T. No pulse adjustment prior to MRSI measurement was necessary due to adiabatic excitation. The protocol geometry was identical for all measurements so that one calibration data set, acquired from phantom replacement measurement, was applied for all quantifications. The protocol was tested in 10 young, healthy volunteers, for whom 57 +/- 7 spectra were quantified. Concentrations per liter of liver volume (reproducibilities) were 2.24 +/- 0.10 mmol/L (1.8%) for phosphomonoesters (PME), 1.37 +/- 0.07 mmol/L (7.9%) for inorganic phosphate (Pi), 11.40 +/- 0.96 mmol/L (2.9%) for phosphodiesters (PDE), and 2.14 +/- 0.10 mmol/L (1.6%) for adenosine triphosphate (ATP), respectively. Taken together, this approach provides fast, simple, and reproducible high-resolution absolute quantification and detailed mapping of the spatial distribution of hepatic (31)P metabolites. This method allows for examination of regional deviations of energy metabolism in human liver diseases.

Quantitative ATP synthesis in human liver measured by localized 31P spectroscopy using the magnetization transfer experiment.

The liver plays a central role in intermediate metabolism. Accumulation of liver fat (steatosis) predisposes to various liver diseases. Steatosis and abnormal muscle energy metabolism are found in insulin-resistant and type-2 diabetic states. To examine hepatic energy metabolism, we measured hepatocellular lipid content, using proton MRS, and rates of hepatic ATP synthesis in vivo, using the 31P magnetization transfer experiment. A suitable localization scheme was developed and applied to the measurements of longitudinal relaxation times (T1) in six healthy volunteers and the ATP-synthesis experiment in nine healthy volunteers. Liver 31P spectra were modelled and quantified successfully using a time domain fit and the AMARES (advanced method for accurate, robust and efficient spectral fitting of MRS data with use of prior knowledge) algorithm describing the essential components of the dataset. The measured T1 relaxation times are comparable to values reported previously at lower field strengths. All nine subjects in whom saturation transfer was measured had low hepatocellular lipid content (1.5 +/- 0.2% MR signal; mean +/- SEM). The exchange rate constant (k) obtained was 0.30 +/- 0.02 s(-1), and the rate of ATP synthesis was 29.5 +/- 1.8 mM/min. The measured rate of ATP synthesis is about three times higher than in human skeletal muscle and human visual cortex, but only about half of that measured in perfused rat liver. In conclusion, 31P MRS at 3 T provides sufficient sensitivity to detect magnetization transfer effects and can therefore be used to assess ATP synthesis in human liver.

[Histometric studies of placental villi in Rh incompatibility]. / Histometrische Untersuchungen an Plazentazotten bei Rh-Inkompatibilität.

The delay in maturation of placental terminal villi in cases of Rh-incompatibility has often been described. The terminal villi of sixteen placentas from pregnancies complicated by Rh-incompatibility histometric were studied. Twenty-two placentas of healthy mothers who had born healthy babies after normal pregnancies served as a control group. The severity of the Rh-incompatibility was classified according to the clinical histories of the newborns and the criteria of Liley. The average area of a cross-section of a terminal villus was 3402 microns 2 (control group 2150 micronsoff; the degree of vascularization was calculated to be 6.5% (control group 30.5%). The other villous parameters measured or calculated show that maturation of the placenta is extremely retarded in cases of Rh-incompatibility. This failure in differentiation can not be compensated for by an enlargement of the organ as a whole. It was shown that the changes in the placental villi are closely correlated to the severity of the fetal disease.