2017 Volume 40 Issue 8 Pages 1268-1274
Although magnetic resonance imaging (MRI) has potential in assessments of formulations, few studies have been conducted because of the size and expense of the instrument. In the present study, the processes of in vitro and in vivo release in a gelatin capsule formulation model were visualized using a compact MRI system with 1.5 T permanent magnets, which is more convenient than the superconducting MRI systems typically used for clinical and experimental purposes. A Gd-chelate of diethylenetriamine-N,N,N′,N″,N″-pentaacetic acid, a contrast agent that markedly enhances proton signals via close contact with water, was incorporated into capsule formulations as a marker compound. In vitro experiments could clearly demonstrate the preparation-dependent differences in the release/disintegration of the formulations. In some preparations, the penetration of water into the formulation and generation of bubbles in the capsule were also observed prior to the disintegration of the formulation. When capsule formulations were orally administered to rats, the release of the marker into the stomach and its transit to the duodenum were visualized. These results strongly indicate that the compact MRI system is a powerful tool for pharmaceutical studies.
Capsule formulations are convenient drug delivery systems that control the stability and release of drugs in the gastrointestinal tract; therefore, they are used in a wide range of drugs that are administered orally. The performance of formulations is influenced by the swelling and/or disintegration of capsules, the penetration of water in formulations, and the release of drug contents in the gastrointestinal tract. Since these processes influence the bioavailability of drugs, visualizing these processes may provide direct information for assessing capsule formulations.
Magnetic resonance imaging (MRI) is a non-invasive imaging technique that provides unique information that is useful for diagnostic purposes. MRI is sensitive to the local concentrations and physical state of protons, mainly those in water in biological specimens. The advantages of MRI over other non-invasive imaging techniques such as γ-scintigraphy are its high spatial and temporal resolutions and representation of anatomical structures. Paramagnetic compounds such as chelate compounds of Gd(III) are typically used as contrast agents.1) Gd-chelates shorten the longitudinal relaxation time (T1) of protons in water by making close contact with water, which markedly enhances proton signals on T1-weighted magnetic resonance (MR) images. In contrast, the dry form of Gd-chelates provides no signal on MR images without contact with water.2,3) The release of substances from tablets and capsules has been visualized in rats3,4) and humans2) using this technique. Despite its usefulness, few studies have employed MRI in pharmaceutical analyses because of the size and expense of superconducting MRI systems.
Compact MRI systems were recently developed with 0.5–1.0 T permanent magnets and used in pathological studies on mice and human hands5,6) as well as in analyses of the biodistribution of paramagnetic substances in rats and mice.7,8) We previously investigated the pharmacokinetics of nitroxyl radicals, paramagnetic probes sensitive to biological redox, in mice using compact MRI with 1.5 T permanent magnets.9) Compact MRI systems only need a small amount of space, and a shield is virtually unnecessary. In addition, the expense of compact MRI systems is much less than that of superconducting MRI systems typically used for clinical and experimental purposes. Therefore, this system is expected to become a powerful tool in pharmaceutical studies.
In the present study, a Gd-chelate of diethylenetriamine-N,N,N′,N″,N″-pentaacetic acid (Gd-DTPA) was incorporated into capsule formulations as a marker compound, and the processes of the in vitro and in vivo release of the compound were visualized with the compact MRI system using 1.5 T permanent magnets.
Gd-DTPA and meglumine were purchased from Sigma Chemical Co. (St. Louis, MO, U.S.A.), starch from potato and gelatin from bovine bone were from Wako Pure Chemical Industries, Ltd. (Osaka, Japan), and size 9 gelatin capsules (2.7 mm o.d.×8.4 mm long) were from Torpac Inc. (Fairfield, NJ, U.S.A.). Pure water was freshly prepared with a Millipore Milli-Q Labo (Bedford, MA, U.S.A.).
AnimalsMale Wistar rats (age 6 weeks, 150–155 g) were purchased from Kyudo Co., Ltd. (Saga, Japan). They were housed in a temperature- and humidity-controlled room and fed a commercial diet (CLEA Japan, Inc., Tokyo, Japan) and water ad libitum. Rats weighing 300 g were fasted for 16 h, but allowed free access to water before experiments. Animal experiments were performed according to the guidelines of the Laboratory Protocol of Animal Handling, Sojo University under the permission of the Committee of Laboratory Animal Experiments in our university.
Calibration CurvesSmall glass tubes (5 mm i.d.×45 mm long) were filled with aqueous solutions of 0, 0.125, 0.25, 0.5, 1, 2, and 3 mmol/L Gd-DTPA. T1-weighted slice MR images were obtained with either a spin echo (SE) or gradient echo (GE) sequence at room temperature (approximately 25°C).
In Vitro Release ExperimentTo increase the water solubility of Gd-DTPA, Gd-DTPA was mixed with meglumine which is usually used as a base/pH adjuster. Gd-DTPA (102 mg) and meglumine (100 mg) were dissolved in distilled water and lyophilized. The residue (31 mg) was mixed well with starch in a mortar, and added to size 9 gelatin capsules. The amount of the filled material was 19.4±1.4 mg (mean±standard deviation (S.D.)). The prepared capsule formulations were immersed in distilled water in a 2-mL polypropylene tube (8 mm i.d.) with a screw cap. The capped tube was placed in an MRI coil unit (solenoid coil, 30 mm i.d.) together with two 0.5-mL polypropylene sample tubes (6 mm i.d.); one containing an aqueous solution of 0.5 mmol/L Gd-DTPA as a standard and another containing pure water. T1-weighted multi slice MR images were taken with a SE sequence at room temperature (approximately 25°C) 5, 10, 15, 20, 25, 30, 40, 50, 60, and 70 min after immersion.
In Vivo Release ExperimentGd-DTPA (25 mg), meglumine (18 mg), and starch (500 mg) were mixed well in a mortar, and 18 mg of the mixture was added to size 9 gelatin capsules. Rats were orally administered fractured gelatin gel (7%, 0.5 mL) using a feeding needle. Rats were then anesthetized with isoflurane (1.5% in air, 0.7 L/min), taped on a teflon bed in a prone position, and set in an MRI coil unit (solenoid coil, 50 mm i.d.). T1-weighted multi slice MR images were taken with the SE sequence at room temperature (approximately 25°C). Rats were loosened and orally administered capsule formulations using a dedicated dosing apparatus (Torpac Inc.). Rats were anesthetized again with isoflurane (1.5% in air, 0.7 L/min), and T1-weighted multi slice MR images were taken with the SE sequence 8, 20, 34, 44, 60, 70, 100, 105, and 125 min after the administration of the capsule formulations.
MRI ExperimentMRI was performed with a MR VivoLVA compact MRI system (DS Pharma Biomedical Co., Ltd., Osaka, Japan), which is based on 1.5 T permanent magnets with an 80-mm gap. Parameters for the SE sequence were as follows; repetition time of a radio frequency pulse (TR), 540 ms; echo time (TE), 14 ms; number of excitations, 4 for calibration and in vivo experiments or 3 for in vitro experiments, while those for the GE sequence were as follows; TR, 250 ms; TE, 6 ms; flip angle, 40°; number of excitations, 4. The field of view was 30×60 mm for the calibration and in vitro experiments and 50×100 mm for in vivo experiments. Image resolution was 128×256 in all cases. Slice thickness was set to 1.0 mm for the calibration, 1.5 mm for in vitro experiments, and 2.0 mm for in vivo experiments. Real thickness was slightly widened by eddy currents; approximately 1.6 and 2.2 mm were estimated for the setting values 1.5 and 2.0 mm, respectively, from reductions in the magnitude of the field gradient.
Estimation of the Amount of Gd-DTPA ReleasedImages were analyzed using ImageJ 1.48 (ver. 5) software. Intensity enhancement (IE) was defined as follows;
 |
 |
GE and SE sequences are widely used for MRI. In order to identify suitable sequences for this experiment, various concentrations of Gd-DTPA aqueous solutions were imaged with the compact MRI system. When image intensity was compared with and without Gd-DTPA, the contrast obtained with the SE sequence was markedly higher than that obtained with the GE sequence (Fig. 1A). Intensity expressed with IE increased with elevations in the concentration of Gd-DTPA, and then reached a plateau in both sequences (Fig. 1B). In the SE sequence, the increase observed in the IE value was linear within the range of 0–0.5 mmol/L Gd-DTPA. These results indicate that the SE sequence is suitable for this experiment, and that a quantitative analysis is possible in the range of 0–0.5 mmol/L Gd-DTPA using IE values.
(A) Small glass tubes containing various concentrations of Gd-DTPA were set in a compact MRI system. T1-weighted images were obtained with either the SE or GE sequence. Concentrations of Gd-DTPA were schematically indicated at the bottom. (B) IE for the SE sequence (open circle) or GE sequence (filled circle) was plotted against the concentration of Gd-DTPA.
Capsule formulations containing Gd-DTPA were immersed in distilled water, and multi slice images were repeatedly taken with the SE sequence. Three slice images including capsule formulations were shown in the time course (Fig. 2). The dry content of the formulation gave no signal in images taken in the early stages, and capsules were delineated with high intensity signals. The release of Gd-DTPA gave high intensity. The images rendered showed that Gd-DTPA was increasingly released and diffused three dimensionally in the aqueous phase with time.
Capsule formulations containing Gd-DTPA dispersed in starch powder were placed in pure water in plastic tubes. T1-weighted images of the tubes were obtained with the SE sequence. (A) Schematic illustrations of a phantom in in vitro experiments and the position of sagittal slices against capsule formulations in a tube filled with water. (B) Slice images including capsule formulations (slices 2–4) were shown with time after immersion in water. Bright and dark circles at the top of the images are 0.5 mmol/L Gd-DTPA aqueous solution and water, respectively. Images at 4 time points were presented. The release of Gd-DTPA from capsule formulations was clearly visualized with time as an enhancement in signal intensity.
Images of 9 preparations of capsule formulations 70 min after immersion were shown in Fig. 3. The extent of marker release and capsule disintegration varied depending on the preparations. The results of 9 experiments were roughly divided into the following two patterns: gradual release without a large change in the shape of the capsule (a–d in Fig. 3), and rapid release accompanied by the disintegration of the capsule (e–i in Fig. 3). Even 70 min after immersion, a large amount of dry substances remained in capsules in all preparations. In some cases of disintegration, the penetration of water and generation of bubbles were observed as a high intensity signal (open arrowheads) and no signal (filled arrowheads), respectively, within the content of capsules, as indicated in Fig. 3(g, h). The time course of MR images clarified that the penetration of water and generation of bubbles occurred prior to the swelling of capsule formulations to reach disintegration (Fig. 4).
T1-weighted images were taken as described in the legend of Fig. 2, and slices including capsule formulations (slice 3) were shown in each preparation. Release was roughly divided into 2 patterns: gradual release without a marked change in the shape of the capsule (preparations a–d) and rapid release accompanying the disintegration of the capsule (preparations e–i). Open and filled arrowheads indicate water penetration into the content of the capsule and the generation of bubbles in the capsule, respectively.
T1-weighted images for preparation h in Fig. 3 were presented with time. Open and filled arrowheads indicate water penetration into the content of the capsule and the generation of bubbles within the capsule, respectively.
In order to compare marker release more precisely, the amount of Gd-DTPA released was estimated. The amounts of Gd-DTPA in the aqueous phase were calculated in each image of 3 slices (slices 2–4 indicated in Fig. 2) and summed. The total amount of Gd-DTPA was plotted against time (Fig. 5). The results obtained demonstrated that the rate of release was different, even within groups with similar release patterns.
The amount of Gd-DTPA released was estimated by the method indicated in Materials and Methods. Estimations were conducted using images up to 25 min after immersion. Slices 2–4 were used for estimations because the thickness of the aqueous phase for slices 1 and 5 was not homogeneous with the curved edge of the tube. Values out of the linear range of the calibration curve were included in order to calculate data points indicated with an asterisk.
A capsule formulation containing Gd-DTPA was orally administered to rats, and the release of the marker from the formulation was visualized using the compact MRI system. The release of Gd-DTPA was rendered with a high intensity signal, as observed in in vitro experiments (Fig. 6). The area of the high intensity signal increased within the stomach up to 44 min, and the stomach was then filled with Gd-DTPA. The efflux of some Gd-DTPA into the duodenum was observed after 100 min (open arrowheads in Fig. 6). The dry content of the capsule formulation was observed in the stomach as a no signal region (filled arrowhead in Fig. 6). The dry part and most of the Gd-DTPA released were retained within the stomach, even 125 min after the administration of the capsule formulation.
Rats were orally administered capsule formulations containing Gd-DTPA, and T1-weighted MR images of rats were obtained with the SE sequence. Two coronal slices including the stomachs of individual rats were shown along with the time after the administration of capsule formulations. Filled and open arrowheads indicate the dry part of the formulation and efflux of the marker into the duodenum, respectively.
In order to clarify the release profile in the stomach quantitatively, the amount of Gd-DTPA released in the stomach was estimated. The time course indicates that the amount of Gd-DTPA released increased up to approximately 40 min after the administration of the capsule formulation, and this increase then became very slow (Fig. 7). The efflux of Gd-DTPA into the duodenum occurred after the increase in the amount of the marker in the stomach became slow. Thus, quantitative analysis of the gastrointestinal transit of a drug was possible with a compact MRI system.
The amounts of Gd-DTPA released in the stomach (open circle) and duodenum (open diamond) were estimated by the method indicated in Materials and Methods using three slice images including the stomach. Values out of the linear range of the calibration curve were included in order to calculate data points indicated with an asterisk.
This study clearly demonstrated the visualization of release from and the disintegration of capsule formulations using the compact MRI system. In order to visualize the behavior of a drug in capsule formulations, the contrast agent Gd-DTPA was used as a model drug. Gd(III) strongly enhances the proton signal of water on T1-weighted MR images once it makes close contact with water, whereas it gives no MRI signal in a solid or dry state. Therefore, Gd-DTPA may act as a marker of release. The visualization of Gd-DTPA release was possible not only in vitro, but also in vivo because MRI is non-invasive and images the concentration and physical state of protons despite the opacity of specimens. The signal enhancement by Gd-DTPA should be pH-independent. Thus, pharmaceutical studies may represent one of the suitable applications of MRI, in which its advantages are used. Recently Curley et al.10) visualized disintegration of tablets in vitro and in the human stomach using a clinical MRI system without any contrast agents. However, they did not go far enough to visualize release of contained materials. Thus, use of contrast agent should be necessary to monitor process of release.
In the present study, a compact MRI system with 1.5 T permanent magnets was used to assess capsule formulations. The magnet gap of this instrument is 80 mm. Therefore, the sample space of this instrument is markedly smaller than that of clinical MRI, whereas their magnetic fields are similar. However, this space is sufficient to carry out in vitro experiments and experiments with small laboratory animals such as rats.
Some in vitro MR images clearly visualized the processes of the disintegration of and release from capsule formulations. The penetration of water was noted prior to the swelling and disintegration of the formulations, and these processes occurred under the continuous release of the marker Gd-DTPA. The outer diameter of the capsule used was 2.7 mm. The compact MRI system employed visualized events occurring within this size. The penetration and swelling of tablets were previously analyzed in vitro and with subcutaneous implantation in rats using a 7 T superconducting MRI system.11) The present study showed that the spatial resolution of a compact MRI system with 1.5 T permanent magnets is sufficient for analyzing the processes of the disintegration of and release from capsule formulations in vitro; however, the MR images obtained under a high magnetic field provided high spatial resolution. In some cases of disintegration, the generation of bubbles was observed in the content of the capsule. These bubbles may be formed with air in the void space of powder in capsules. Air may be pushed out by water with the increased penetration of water into capsule formulations.
In our parameter settings, the time spent on data collection per scan was 3.6 min in vitro and 4.8 min in vivo. On the other hand, the in vitro experiments showed that the marker compound Gd-DTPA continued to be released from gelatin capsules for more than 1 h. In addition, the capsule formulation remained in the stomach for more than 2 h, which is consistent with previous findings reported by Kremser et al.12) with the clinical MRI system. They showed that gelatin capsules and tablets remained in the rat stomach for approximately 3 h. The time spent on data collection per image was markedly shorter than the time scales of capsule release and gastrointestinal transit. Therefore, the time resolution of MRI is sufficiently high to analyze the in vitro and in vivo release processes from slow-release formulations, such as capsule formulations.
The presence of food in the gastrointestinal tract may modify drug release, capsule disintegration, and retention times of formulations in the stomach. Furthermore, a previous study reported that food in the gastrointestinal tract frequently presented as a hyperintense signal, which resulted in the misinterpretation of images.12) Therefore, the fasting of experimental animals is essential for MRI experiments. In the present study, rats were fasted for 16 h. Fractured gelatin gel was orally administered to rats prior to the administration of capsule formulations in order to mimic fed conditions and maintain enough space to detect the released marker in the stomach.
Shape of capsules was delineated with signal with high intensity after immersion (Fig. 2). Whereas the signal was relatively weak 5 min after immersion, it became remarkable along with time. Empty capsule immersed in water gave signal whose intensity was comparable to that of signal for capsule formulation containing Gd-DTPA at 5 min after immersion, but no remarkable increase was observed with time (data not shown). Neither capsule before immersion nor capsule immersed in deuterium oxide gave no signal (data not shown). These observations indicate that signal delineating capsule wall results from proton in water infiltrating into gelatin gel of capsule at early stage, and that the signal is strongly enhanced by Gd-DTPA dissolved around capsule wall at later stage.
The calibration curve of signal enhancement against Gd-DTPA concentrations slightly arched in the concentration range of 0–0.5 mM, but was linear in that range. Data for Figs. 5 and 7 include values out of the linear range of the calibration curve in order to understand tendency of release. It should be noted that data with asterisk should be underestimated. The linearity of the calibration curve may be improved by using nitroxyl radicals instead of Gd-chelates. Nitroxyl radicals are durable organic free radicals used as spin labels and redox probes.13–18) The calibration curve was linear up to at least 3 mmol/L of nitroxyl radicals, which is consistent with previous findings by Hyodo et al.17); however, the slope of the calibration curve was less than one-fifteenth that for Gd-DTPA (data not shown). This difference in sensitivity may result from differences in the number of unpaired electrons in a molecule, 7 electrons for Gd(III) vs. 1 electron for nitroxyl radicals. The use of nitroxyl radicals may increase the accuracy of quantitative analyses.
In conclusion, the processes of the disintegration of and release from capsule formulations were assessed using a compact MRI system with 1.5 T permanent magnets and the contrast agent Gd-DTPA. The leakage of formulation contents, penetration of water into formulations, and swelling and disintegration of formulations were clearly visualized. This technique may be applicable to the assessment of other solid formulations such as tablet formulations. The compact MRI system with permanent magnets is more convenient than superconducting MRI systems in the terms of cost, facilities for installation, and handling. The compact MRI system is expected to make a significant contribution to pharmaceutical studies.
This study was partially supported by JSPS KAKENHI Grant Numbers 25460053 and 15K07912. We thank our student, Ms. Yuriko Hirose, for her excellent technical assistance with the experiments.
The authors declare no conflict of interest.
