Continuous Monitoring of Pseudopolymorphic Transition in Ezetimibe Using T₁ Relaxation with Time-Domain NMR

Abstract

The purpose of this study was to continuously monitor the pseudopolymorphic transition from anhydrate to monohydrate by measuring the NMR relaxation using time-domain NMR (TD-NMR). Taking advantage of the simplicity of the low-field NMR instrument configuration, which is an advantage of TD-NMR, the NMR instrument was connected to a humidity controller to monitor the pseudopolymorphic transition. First, ezetimibe (EZT) monohydrate was prepared from its anhydrate using a saturated salt solution method, and T₁ relaxation of EZT monohydrate and anhydrate was measured without a humidity controller. The T₁ relaxation results confirmed that EZT anhydrate and monohydrate could be distinguished using T₁ relaxation measurement. Next, continuous monitoring was conducted by TD-NMR and connected to a humidity controller. Anhydrous EZT was placed in an NMR glass tube and the T₁ relaxation measurement was repeated while maintaining the humidity on the side entering the NMR tube at 80% relative humidity. The T₁ relaxation became gradually faster from the initial to middle monitoring phases. The final T₁ relaxation was then recovered fully and these T₁ relaxation times were the same as the T₁ relaxation of EZT monohydrate. This study successfully monitored the pseudopolymorphic transition from EZT anhydrate to monohydrate via NMR relaxation.

Introduction

Pseudopolymorphism is an important phenomenon in pharmaceutical sciences. One-third of active pharmaceutical ingredients (APIs) have the ability to form a hydrate,¹⁾ which is one of the pseudopolymorphs, and pseudopolymorphism affects dissolution behavior and bioavailability in drug formulation. Thus, there have been several attempts to monitor the pseudopolymorphic transition of APIs. For example, near-IR,²⁾ IR,³⁾ and Raman spectroscopy⁴⁾ analyses have been used to monitor the pseudopolymorphic transition of APIs. Solid-state NMR spectroscopy provides valuable information on pseudopolymorphism from the perspective of molecular mobility. However, solid-state NMR spectroscopy has yet to be used for monitoring pseudopolymorphism, presumably because solid-state NMR spectroscopy is measured using high-field instruments, which are generally expensive and require expertise. Connecting high-field instruments to other equipment (e.g., humidification or drying unit) is difficult.

The purpose of this study was to continuously monitor the pseudopolymorphic transition by measuring the NMR relaxation using time-domain NMR (TD-NMR), which is a method for measuring the relaxation of NMR and is mainly conducted by low-field NMR instruments.^1,5) Taking advantage of TD-NMR (i.e., the simplicity of the equipment configuration compared to the high-field instruments used by solid-state NMR spectroscopy), several techniques have been reported to measure NMR relaxation by connecting the low-field NMR instrument used for TD-NMR to other equipment. For example, a melt-drawing device⁶⁾ for solid samples and rheometer^7,8) for semi-solid samples were introduced to the low-field NMR instrument, and an in-line monitoring system under flow conditions⁹⁾ for liquid samples was introduced to the NMR instrument.

In this study, the pseudopolymorphic transition was continuously monitored under constant humidity by connecting a humidification unit to the low-field NMR instrument, and T₁ relaxation, one of the relaxation phenomena, was measured by TD-NMR. Ezetimibe (EZT), for which the pseudopolymorphic transition from anhydrate to monohydrate has been confirmed, was used.

Experimental

Materials

Anhydrous EZT (Nichi-Iko Pharmaceutical, Toyama, Japan) was used as a test API.

Sample Preparation

Anhydrous EZT powder was stored at 25 °C/84% relative humidity (RH) for 24 h using a saturated salt solution method (potassium chloride) and then used as a moisture-adsorbed EZT.

Water Vapor Adsorption and Desorption Measurement

The water vapor adsorption and desorption were measured using the Q5000SA Sorption Analyzer (TA instruments, New Castle, DE, U.S.A.), and EZT anhydrate (5–10 mg) was placed in the sample cup. The RH was controlled by nitrogen gas. Sample mass was measured every 5 min with a microbalance, and the humidity was automatically regulated from 0 to 95% RH at 25 °C. The mass stability criterion employed for equilibrium was that the maximum mass change (% min⁻¹) for each measurement was less than 0.001%. The sample mass at 0% RH was used as the initial value to remove the influence of a slight amount of water adsorbed on the intact sample powder, and the rate of weight change at each humidity was monitored.

Powder X-Ray Diffraction Measurement

The powder X-ray diffraction (PXRD) patterns of all samples were obtained using the SmartLab high-resolution X-ray diffractometer (Rigaku, Tokyo, Japan) with Cu-Kα radiation (λ = 0.154 nm). The count scanning rate was 12°/min and the 2θ scanning angle was from 5 to 35°.

Thermogravimetric and Differential Thermal Analysis

Thermogravimetric and differential thermal analysis (TG-DTA) of all samples was performed on the EXSTAR TG/DTA 7300 thermogravimetric analyzer (Hitachi High-Tech Science, Tokyo, Japan). The sample was heated from ambient temperature to 220 °C at a heating rate of 10 °C/min under nitrogen flow.

¹H T₁ Measurement

¹H T₁ relaxation behavior of the samples was obtained with the TD-NMR technique using the Bruker NMR Minispec mq20 apparatus (Bruker BioSpin, Billerica, MA, U.S.A.) at a ¹H frequency of 20 MHz at 25 °C. A solid echo sequence was used for the measurement and the following parameters were applied for anhydrous EZT: four scans, recycle delay, 4–400 s; and number of points, 10. For EZT monohydrate: 32 scans, recycle delay, 0.05–5 s; and number of points, 50. The measurement was made in triplicate and the mean ± standard deviation was calculated. ¹H T₁ relaxation behavior of EZT powder was calculated according to Equation 1 using the TD-NMR Analyzer software (Bruker BioSpin).

  

(1)

where M_(t) and M₀ are the signal intensities at time t and equilibrium, respectively; t is the time delay interval used during the NMR relaxation measurement; and T₁ is the T₁ relaxation time, which is a time constant that represents the recovery speed from the M_(t) component to the equilibrium M₀.

Continuous ¹H T₁ Relaxation Monitoring

Continuous ¹H T₁ relaxation monitoring was conducted using the Bruker Minispec mq20 apparatus connected to a humidity controller (Kitz Microfilter, Nagano, Japan), which prepared an 80% RH environment (Supplementary Fig. S1). Then 200 mg anhydrous EZT was weighed in a 10 mm diameter NMR tube and placed in the NMR instrument. The humidity controller was connected to the NMR tube, and a pump sent humidity-controlled air, 80% RH, into the NMR tube. Immediately afterwards, the ¹H T₁ relaxation measurement, which was a 0 min measurement, was started. As soon as one measurement was finished, the subsequent measurement was started. The humidified air was discharged through an outlet hole in the lid of the NMR tube.

Results and Discussion

Characterization of EZT

First, the water vapor adsorption and desorption of EZT were measured (Fig. 1). During the adsorption process, the weight increase was 1.1% at 50% RH, which was slight and almost negligible. Then the weight increase was 4.0, 4.2, 4.3, 4.3, and 4.3% at 60, 70, 80, 90, and 95% RH, respectively, which reached a plateau. During the desorption process, the weight increase retained 4.3% from 95 to 25% RH and significantly decreased at 20% RH. Considering the molar mass of EZT anhydrate and monohydrate, 409.4 and 427.4 g/mol, respectively, the theoretical weight increase of EZT anhydrate was 4.4%. Thus, a pseudopolymorphic transition from EZT anhydrate to monohydrate was observed when EZT was humidified above 60% RH.

Fig. 1. Water Adsorption and Desorption Measurement of Anhydrous Ezetimibe (EZT)

The sample mass at 0% RH was used as the initial value to remove the influence of a slight amount of water adsorbed on the intact sample powder, and the rate of weight change at each humidity was monitored.

PXRD measurement and TG-DTA were also conducted to confirm that the moisture-adsorbed EZT prepared by the saturated salt solution method was an EZT monohydrate. The PXRD pattern of moisture-adsorbed EZT differed from that of EZT anhydrate (Fig. 2). For moisture-adsorbed EZT, the characteristic diffraction peaks were observed at 7.9, 12.0, and 15.8°, corresponding to EZT monohydrate.¹⁰⁾ For the TG-DTA of moisture-adsorbed EZT, the TG curve showed that the mass percentage at 70 °C was 4.5% (Fig. 2). Since the DTA curve at 70 °C showed a slight endothermic peak at 47.9 °C, which differed from the melting point of EZT (163–164 °C),^11,12) the mass percentage at 70 °C was due to adsorbed water. Considering the molar mass of EZT anhydrate and monohydrate, 409.4 and 427.4 g/mol, respectively, the theoretical mass percentage of water in EZT monohydrate was 4.2%, which was consistent with the mass percentage of moisture-adsorbed EZT at 70 °C. Thus, the water in the moisture-adsorbed EZT prepared in this study was hydration water in EZT monohydrate. These results confirmed that moisture-adsorbed EZT was EZT monohydrate, and EZT monohydrate and EZT anhydrate can be distinguished using PXRD measurement and TG-DTA.

Fig. 2. Characterization of Moisture-Adsorbed Ezetimibe (EZT)

(a) Powder X-ray diffraction measurement. (b) Thermogravimetry and differential thermal analysis.

Continuous TD-NMR Measurement

Before the continuous TD-NMR monitoring, T₁ relaxation of EZT anhydrate and EZT monohydrate (i.e., moisture-adsorbed EZT) were measured, and the difference between them was compared. The T₁ relaxation of anhydrous EZT was significantly slower than that of EZT monohydrate (Fig. 3). The T₁ relaxation time of anhydrous EZT was 53.5 ± 3.8 s, which was significantly longer than that of EZT monohydrate, 0.33 ± 0.02 s. This significant difference in T₁ relaxation time may be due to the absence of methyl groups in the chemical structure of EZT. Munson’s group indirectly measured ¹H T₁ relaxation of carbohydrate¹³⁾ and dicumarol,¹⁴⁾ which do not have methyl groups, from well-resolved ¹³C spectra. They mentioned that rigid solids with no relaxation sink (e.g., methyl groups) have long T₁ relaxation times. The authors also noted that rotating methyl groups or other highly mobile regions of sample quickly relax and magnetization from the slow-relaxing domain is transferred to the fast-relaxing domain, which acts as sinks for relaxation of the entire sample. In the present study, anhydrous EZT without methyl groups may be the slow-relaxing domain; by contrast, hydrated water in EZT monohydrate may be the fast-relaxing domain and function as relaxation sinks. From the T₁ relaxation results, EZT anhydrate and monohydrate could be distinguished using T₁ relaxation measurement by TD-NMR. By contrast, the T₂ relaxation did not significantly differ between EZT anhydrate and monohydrate (Supplementary Fig. S2). Presumably, the decay of T₂ relaxation for the solid sample was too fast in principle to detect differences. Thus, we confirmed that the pseudopolymorphic transition from EZT anhydrous to monohydrate can be observed in continuous monitoring using T₁ relaxation.

Fig. 3. Representative ¹H T₁ Relaxation of (a) Anhydrous Ezetimibe (EZT) and (b) Moisture-Adsorbed EZT

Finally, the continuous monitoring of T₁ relaxation of EZT anhydrate under 80% RH was conducted by TD-NMR connected with humidification equipment to monitor the change into its monohydrate. The measurement was repeated under conditions suitable for measuring T₁ relaxation of EZT monohydrate (i.e., recycle delay was from 0.05 to 5 s and approx. 30 min for each measurement) at 80% RH. In this experiment, we expected that the T₁ relaxation of EZT anhydrous, having a significantly slower T₁ relaxation time, would gradually become faster according to the exposure to high humidity and then eventually become consistent with the T₁ relaxation of EZT monohydrate. Regarding the initial phase of T₁ relaxation measurement ranging before humidification to 0 min, the NMR signal was not fully recovered (Fig. 4a). Then a trend of faster T₁ relaxation from 0 to 181 min was observed as the storage time at humidification increased. However, there is a possibility that this trend was simply due to moisture absorption (i.e., an enhancement of the NMR signal or pseudopolymorphism transition). Thus, normalization of the T₁ relaxation was conducted based on the NMR signal at the final recycle delay (t = 5.0 s), and it showed that the normalized T₁ relaxation from 0 to 181 min was obviously different from the latter phase of the measurement (Supplementary Fig. S3). In other words, it was confirmed that T₁ relaxation was gradually faster from 0 to 181 min. However, T₁ relaxation from 211 and 241 min was fully recovered and almost the same. The T₁ relaxation time was 0.40 and 0.38 s for 211 and 241 min, respectively, which gradually approached the T₁ relaxation of EZT monohydrate, 0.33 s. The slight difference in T₁ relaxation time, 0.40, 0.38, and 0.33 s, may be attributed to the amount of adsorbed water on EZT. The T₁ relaxation from 0 to 181 min was considered to be a change from EZT anhydrous to monohydrate, and the T₁ relaxation from 211 and 241 min was that of EZT monohydrate.

Fig. 4. Continuous Monitoring of ¹H T₁ Relaxation of Anhydrous Ezetimibe (EZT)

(a) ¹H T₁ relaxation was measured during humidification at 80% RH. (b) Powder X-ray diffraction. (c) Thermogravimetry and differential thermal analysis was conducted using the EZT powder sampled from the NMR tube after continuous monitoring. T₁ relaxation was repeatedly obtained every 30 min. A recovery curve of T₁ relaxation plots shows the NMR signal’s average during the 30 min: for example, “0 min” is the average of the NMR signal from 0 to 30 min.

To obtain further confirmation of whether the EZT anhydrate turned into monohydrate by storage under humid conditions, the sampled EZT was collected from the NMR tube after the continuous T₁ relaxation measurement for 241 min. Then PXRD and TG-DTA measurements were conducted. The PXRD pattern of the sampled EZT showed characteristic diffraction peaks at 7.8, 12.0, and 15.7°, which was consistent with that of EZT monohydrate (Fig. 4b). For the TG-DTA of the sampled EZT, the TG curve showed that the mass percentage at 70 °C was 4.2%, which was consistent with the theoretical mass percentage of water in EZT monohydrate, 4.2% (Fig. 4c). Therefore, PXRD measurement and TG-DTA strongly supported the EZT powder sampled as an EZT monohydrate after continuous monitoring. In summary, the pseudopolymorphic transition of EZT was successfully monitored by connecting humidification equipment to the TD-NMR technique and measuring T₁ relaxation.

Conclusion

This study successfully monitored the pseudopolymorphic transition from EZT anhydrate to monohydrate using the TD-NMR technique connected with a humidity controller. The biggest advantage of TD-NMR connected to a humidity controller is that a researcher can monitor the transition from the perspective of molecular mobility.

Acknowledgments

This work was supported by the Japan Society for the Promotion of Science (JSPS) KAKENHI (Grant No. 22K15261).

Conflict of Interest

The authors declare no conflict of interest. The Laboratory of Pharmaceutical Technology, University of Toyama, is an endowed department supported by an unrestricted Grant from Nichi-Iko Pharmaceutical Co., Ltd. (Toyama, Japan).

Supplementary Materials

This article contains supplementary materials.

References

• 1) Schumacher S. U., Rothenhausler B., Willmann A., Thun J., Moog R., Kuentz M., J. Pharm. Biomed. Anal., 137, 96–103 (2017).
• 2) Otsuka M., Kanai Y., Hattori Y., J. Pharm. Sci., 103, 2924–2936 (2014).
• 3) Bagheri L., Madadlou A., Yarmand M., Mousavi M. E., Food Res. Int., 62, 1113–1119 (2014).
• 4) Takeshima R., Hattori Y., Managaki S., Otsuka M., Int. J. Pharm., 530, 256–262 (2017).
• 5) Yoshioka S., Aso Y., Osako T., Kawanishi T., J. Pharm. Sci., 97, 4258–4268 (2008).
• 6) Kakiage M., Uehara H., Yamanobe T., Macromol. Rapid Commun., 29, 1571–1576 (2008).
• 7) Räntzsch V., Özen M. B., Ratzsch K. F., Stellamanns E., Sprung M., Guthausen G., Wilhelm M., Macromol. Mater. Eng., 304, 1800586 (2019).
• 8) Fengler C., Keller J., Ratzsch K. F., Wilhelm M., Adv. Sci. (Weinh.), 9, e2104231 (2022).
• 9) Taraban M. B., Briggs K. T., Merkel P., Yu Y. B., Anal. Chem., 91, 13538–13546 (2019).
• 10) Farias M. A., Soares F. L. F., Carneiro R. L., J. Pharm. Biomed. Anal., 121, 209–214 (2016).
• 11) Knapik J., Wojnarowska Z., Grzybowska K., Jurkiewicz K., Stankiewicz A., Paluch M., Mol. Pharm., 13, 1308–1316 (2016).
• 12) Mohara M., Davis M. P., Korter T. M., Shimura K., Ono T., Aiko K., J. Phys. Chem. A, 126, 2879–2888 (2022).
• 13) Lubach J. W., Xu D., Segmuller B. E., Munson E. J., J. Pharm. Sci., 96, 777–787 (2007).
• 14) Dempah K. E., Lubach J. W., Munson E. J., Mol. Pharm., 14, 856–865 (2017).