Synthesis and Biologic Evaluation of ¹¹C-Methyl-d-Glucoside, a Tracer of the Sodium-Dependent Glucose Transporters

Radiosynthesis and HPLC Separation

¹¹C-Methyl triflate was obtained by passage of ¹¹C-CH₃I over a Ag-triflate column at 180°C. ¹¹C-CH₃I was synthesized from ¹¹CO₂ that was first reduced to ¹¹CH₄ and subsequently reacted with I₂ vapor at 650°C using a home-built recirculation synthesis module. ¹¹C-Methyl triflate was bubbled through a solution of 3 mg of d-glucose (or l-glucose) in a mixture of 0.5 mL of 0.05 mol/L NaOH and 1 mL of acetone for 5 min at room temperature. The solution was then neutralized with 0.2 mL phosphate buffer, pH 6 (0.5 mol/L), and acetone was evaporated by heating (90°C) under a stream of helium over 4 min. The resulting solution was diluted with saline to an appropriate radioactive concentration for the biodistribution study with the mixture of α- and β-anomers of ¹¹C-methylglucoside.

For biodistribution of the individual α- and β-anomers, the anomeric mixture was separated using high-performance liquid chromatography (HPLC) with an Aminex HPX-87c column (Bio-Rad) (300 × 7.8 mm) eluted with water at a flow rate of 0.5 mL/min. The α- and β-anomers of ¹¹C-methyl-d-glucoside were found to coelute with authentic α-MDG (retention time [Rt] = 9.7 min) and authentic βMDG (Rt = 8.7 min), respectively. The isolated peaks were reanalyzed 60 min after isolation by the same HPLC system. The purified α- and β-anomers were diluted with saline to an appropriate radioactive concentration for the biodistribution studies. Plasma was analyzed on a carbohydrate-NH₂ column (300 × 4.1 mm; Alltech) eluted with a mixture of CH₃CN:H₂O (90:10) at a flow rate of 1.5 mL/min. In this system both anomers elute as 1 peak at Rt = 4.8 min.

The column eluates were passed through a refractive index and a radiometric detector, and the detector signals were processed using a Rachel acquisition system (LabLogic). The recovery of radioactivity in the eluate (relative to the injected activity) was 98%.

Biodistribution in Mice and Autoradiography

The biodistribution experiments were performed in compliance with national laws related to the conduct of animal experimentation. Male NMRI mice (body mass, 25–30 g; fed ad libitum) were sedated by intramuscular injection of 0.05 mL of a 1:4 diluted solution of Hypnorm (fentanylcitrate [0.315 mg/mL] and fluanisone [10 mg/mL]; Janssen Pharmaceutica). Then, 0.1 mL of the tracer solution (3.7 MBq) was injected in each of 4 mice via a tail vein. For pretreatment with phlorizin, sedated mice were injected intraperitoneally with 800 mg of phlorizin/kg of body mass. Phlorizin was dissolved to a concentration of 200 mg/mL in 1,3-propanediol and was injected 1 h before the tracer.

The mice were killed by decapitation at 10 or 30 min after injection. Blood was collected in a tared tube and weighed. All organs and other body parts were dissected and their radioactivity was determined in a γ-counter (Wizard 1480; Wallac). Corrections were made for background radiation and physical decay during counting. For calculation of total blood activity, blood mass was assumed to be 7% of body mass. The hematocrit value of a blood sample obtained 30 min after injection was determined by centrifugation of a hematocrit microcapillary for 5 min in a dedicated centrifuge (Damon). A blood sample (about 1 mL) was weighed and counted for radioactivity. The sample was centrifuged for 10 min at 1,500g and plasma was pipetted off, weighed, and counted in a γ-sample changer. Four-tenths milliliter of the plasma sample was added to 1.5 mL of acetonitrile to denaturate the plasma proteins. The sample was centrifuged and the supernatant was filtered through a 0.22-μm membrane filter and injected on an HPLC column. The HPLC eluate was collected in 1-mL fractions and their radioactivity was determined in a γ-counter. The activity associated with the denaturated plasma proteins in the sediment obtained after centrifugation was also determined by counting in a γ-counter.

To determine the in vivo intrarenal distribution, 0.1 mL of a saline solution containing 37 MBq of ¹¹C-methyl-d-glucoside (mixture of isomers) was injected intravenously in a sedated NMRI mouse. The animal was killed by decapitation at 30 min after injection and a kidney was rapidly removed and immediately frozen in 2-methylbutane (cooled to −25°C with liquid nitrogen). Serial frozen coronal sections of 50 μm were cut and mounted on microscope slides. The sections were immediately air dried at 50°C and exposed for 2 h to a high-performance storage phosphor screen (Packard) that was scanned in a Phosphor Imager scanner (Cyclone; Packard).

Expression of hSGLT1 in Xenopus laevis Oocytes and Transport Measurements

To prepare sense complementary RNA (cRNA) of human SGLT1 (21), purified pBluescript II plasmid containing hSGLT1 was linearized with EcoRI and 5′-7meGppp5′-G capped cRNA was synthesized by T3 RNA polymerase (20). The cRNAs were extracted twice with phenol/chloroform/isoamyl alcohol (25:24:1), precipitated with acidic ethanol, and dissolved in double-distilled water treated with diethylpyrocarbonate.

For expression, Xenopus oocytes were defolliculated and injected with 50 nL of water per oocyte without or with 10 ng of hSGLT1-cRNA (20). The oocytes were incubated 3 d at 19°C in Ori buffer (5 mmol/L Hepes-Tris, pH 7.4, 100 mmol/L NaCl, 3 mmol/L KCl, 2 mmol/L CaCl₂, 1 mmol/L MgCl₂) to which 50 mg/L gentamycin was added.

For tracer flux measurements of αMDG in Xenopus oocytes, the oocytes were incubated in the absence or presence of 100 μmol/L phlorizin with Ori buffer (room temperature) containing 10 μmol/L ¹⁴C-αMDG plus different concentrations of βMDG. After a 1-h incubation, the oocytes were washed and radioactivity in the oocytes was analyzed as described (20). The electric properties of glucose uptake were investigated using the 2-microelectrode voltage-clamp technique (22). The oocytes were clamped to −50 mV and continuously perfused at room temperature (∼3 mL/min) with Ori buffer. Different concentrations of αMDG or βMDG were added to the buffer, and the inward current induced by cotransported sodium was measured.

The uptake of ¹⁴C-αMDG was calculated as the difference of the uptake in 7–10 oocytes in the absence of phlorizin and the uptake in 7–10 oocytes in the presence of phlorizin per data point. Mean values ± SE values are presented. The apparent inhibitory constant (K_i) for βMDG inhibition of ¹⁴C-αMDG uptake was calculated by fitting the Michaelis–Menten equation to the data. The electrical measurements were performed with at least 4 oocytes from 2 different oocyte batches. The Michaelis–Menten equation was fitted to the currents that were induced by different glucose concentrations.

Radiosynthesis

The reaction of glucose with ¹¹C-methyl triflate in alkaline conditions yields not only ¹¹C-methylglucoside but also ¹¹C-methanol as a side product. The latter is easily removed by evaporation after neutralization of the reaction mixture.

The described production method yields a mixture of 2 anomers that coelute during HPLC analysis with the authentic α- and β-anomers of methyl-d-glucoside, respectively (detected with a refractive index detector). Reanalysis of the HPLC-isolated individual anomers showed that they do not interconvert in aqueous solution, in agreement with literature data (23). The anomers of ¹¹C-methyl-d-glucoside are obtained in a ratio of 25% α-anomer and 75% β-anomer. This ratio was not influenced by changes in the reaction conditions (concentration of NaOH, glucose, reaction temperature, reaction duration). The amount of radiochemical impurities in the final solution was <5%. The overall synthesis procedure (starting from ¹¹CO₂) takes 30 min and provides ¹¹C-methylglucoside with a decay-corrected yield of 30% (relative to starting ¹¹C-CO₂). Because of the low sensitivity of the refractive index detector, the specific activity could not be calculated but can be assumed to be in the same range (20 GBq/μmol) as that of other ¹¹C tracers that were synthesized using the same methylation module. The mixture of anomers of ¹¹C-methyl-l-glucoside was obtained using an identical synthesis procedure but starting from l-glucose.

Biodistribution in Mice and Autoradiography

The biodistribution of the standard preparation that contained a mixture of the α- and β-anomers of ¹¹C-methyl-d-glucoside was investigated in mice at 10 and 30 min after injection (Table 1). The kidney is the organ with the highest concentration of ¹¹C-methyl-d-glucoside, and the activity accumulating in the kidneys still continues to increase up to 30 min after injection. In mice pretreated with phlorizin, the renal accumulation of ¹¹C-methyl-d-glucoside observed at 30 min after injection was inhibited by 78% compared with normal mice. In the absence of phlorizin, the clearance from blood and carcass was slow, and only a small amount of ¹¹C-methyl-d-glucoside was excreted with the urine. However, after pretreatment with phlorizin the urinary excretion at 30 min after injection of the tracer was increased about 4-fold and a lower percentage of the injected dose was observed in the blood as compared with the nonpretreated animals. The data indicate that transport systems that can be inhibited by phlorizin are required for renal accumulation and reabsorption of ¹¹C-αMDG or ¹¹C-βMDG. A more detailed analysis of the blood at 30 min after injection showed that >98% of the activity in blood was associated with plasma but was not bound to proteins (protein binding, <1%). This indicates that ¹¹C-αMDG or ¹¹C-βMDG is poorly transported or not transported by the passive glucose transporter GLUT1, which is expressed in erythrocytes (1). Also, the low brain uptake observed for methyl-d-glucoside at both 10 and 30 min after injection is in agreement with poor transport by GLUT1. HPLC analysis of the plasma indicated that all ¹¹C is in the chemical form of methylglucoside, excluding significant degradation or metabolism of ¹¹C-methyl-d-glucoside during the experiment.

Table 2 presents separate results of the biodistribution studies 30 min after intravenous injection of the individual α- and β-anomers of ¹¹C-methyl-d-glucoside. A higher accumulation in the kidneys was observed for the β-anomer compared with the α-anomer. This corresponds to a lower concentration of the β-anomer in blood and carcass 30 min after intravenous injection. Due to the effective reabsorption in the kidney, the urinary excretion of ¹¹C-αMDG and ¹¹C-βMDG is low. The urinary excretion is somewhat higher for the β-anomer that is accumulated in proximal tubular cells. The data suggest that the second step in transtubular movement of ¹¹C-methyl-d-glucoside (the basolateral efflux from cell to interstitium, which is supposed to be mediated by the passive glucose transporter GLUT2 (2)) is slower for the β-anomer compared with the α-anomer.

For a mixture of α- and β-anomers of methyl-l-glucoside, a 20–40 times higher excretion into urine was observed compared with the α- and β-anomers of methyl-d-glucoside. This indicates that both methyl-d-glucoside anomers are reabsorbed from the glomerular filtrate in contrast to the anomers of methyl-l-glucoside. Because the renal accumulation of methyl-l-glucoside was low, it may be concluded that methyl-l-glucoside was not accepted as a substrate of the sodium-d-glucose cotransporters SGLT1 and SGLT2 that are localized in renal brush border membranes. The results of the biodistribution study suggest that the renal accumulation of methyl-d-glucoside represents its accumulation in proximal tubular cells. Both the α- and β-anomers may be transported into the cell, but the β-anomer may be released more slowly at the basolateral site, resulting in a higher renal accumulation and a lower concentration in blood.

The intrarenal distribution of the anomeric mixture of ¹¹C-methyl-d-glucoside was determined by autoradiography on a coronal kidney slice obtained 30 min after injection (Fig. 1). An accumulation of activity was observed throughout the renal cortex reaching into the outer medulla. SGLT2 is expressed in S1 and S2 segments of renal proximal tubules in the outer cortex and midcortex and SGLT1 is expressed in S3 segments of renal proximal tubules that are localized in the inner cortex and outer medulla. Therefore, these data strongly suggest that ¹¹C-methyl-d-glucoside is transported by both SGLT1 and SGLT2.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Intrarenal distribution of ¹¹C-methyl-d-glucoside at 30 min after injection. Autoradiographic image of 50-μm-thick coronal kidney slice is shown. Image is scaled to maximum pixel intensity of image.

Transport of αMDG and βMDG by Human SGLT1

To evaluate the potential human use of ¹¹C-methyl-d-glucoside with PET, we compared the transport of αMDG and βMDG mediated by the human SGLT1 (hSGLT1) transporter. Xenopus oocytes expressing hSGLT1 were clamped to −50 mV, and inward currents that were induced after incubation of αMDG or βMDG in the presence of 100 mmol/L sodium were measured. The current traces in Figure 2A were obtained when the same oocyte was first incubated with 5 mmol/L αMDG and then incubated with 5 mmol/L βMDG. Because maximal currents were induced with these substrate concentrations (see below), they were used to compare the maximal transport velocities for αMDG and βMDG. Figure 2B shows that hSGLT1 translocates both anomers with similar maximal transport velocities. In Figure 3 the substrate dependence of inward currents by αMDG and βMDG were compared at −50 mV. For both anomers, similar apparent Michaelis–Menten constant values were obtained: 1.06 ± 0.03 mmol/L for αMDG and 0.84 ± 0.23 mmol/L for βMDG. The affinity of βMDG to hSGLT1 was also determined by tracer flux measurements with 10 μmol/L ¹⁴C-αMDG, which were performed with varying concentrations of βMDG (Fig. 4). A 50% inhibitory concentration of 0.75 ± 0.12 mmol/L was obtained, which represents a K_i value of 0.74 mmol/L, assuming competitive inhibition. The data show that hSGLT1 transports αMDG and βMDG with the same transport rate and affinity.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Comparison of inward currents by hSGLT1 that were induced by saturating concentrations of αMDG or βMDG. Xenopus oocytes expressing hSGLT1 were clamped at −50 mV, incubated with Ori buffer without sugar or with Ori buffer containing 5 mmol/L αMDG or 5 mmol/L βMDG, and induced inward currents were measured. (A) Typical current trace from 1 oocyte is shown. (B) Data from 5 experiments are summarized.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Substrate dependence of hSGLT1-expressed inward currents induced by αMDG or βMDG. Xenopus oocytes were clamped at −50 mV, incubated with Ori buffer containing different concentrations of αMDG (A) or βMDG (B), and glucose-induced currents were measured. Mean values ± SD from 4 ooyctes are presented. Michelis–Menten equation was fitted to data.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Inhibition of hSGLT1-expressed uptake of ¹⁴C-αMDG uptake by βMDG. Xenopus oocytes expressing hSGLT1 were incubated for 1 h with Ori buffer without and with 100 μmol/L phlorizin that contained 10 μmol/L of ¹⁴C-αMDG plus indicated concentrations of βMDG. Mean values ± SD of phlorizin-inhibited uptake rates measured from pairs of 7–10 oocytes are presented. Michaelis–Menten equation was fitted to data.