Heather Hochrein (Presenter)
UC San Diego Health System
Bio: Heather Hochrein is a clinical laboratory scientist in the toxicology department for the UC San Diego Center for Advanced Laboratory Medicine.
Authorship: Heather Hochrein, Joshua Akin, Krista Pratico, Judith Stone, and Robert Fitzgerald
UC San Diego Health System
| Short Abstract There are many published studies on the performance of commercially available beta-glucuronidases obtained from different sources used for the hydrolysis of glucuronide-conjugated opioids. We have applied several studies testing three enzymes (beta-glucuronidase from Helix pomatia from Sigma-Aldrich, beta-glucuronidase from Haliotis rufescens from Kura Biotec, and a recombinant beta-glucuronidase from IMCS) for hydrolysis efficiency using spiked standards of parent drug and their corresponding glucuronides. We also performed a selected patient comparison focused on the glucuronide-species of interest, concentrating on parameters that could influence interpretation of results in a clinical setting. We saw up to 12-fold differences in recovery for total codeine, hydromorphone, morphine, and oxymorphone. |
Long Abstract
Introduction
There are many published studies on the performance of commercially available beta-glucuronidases obtained from different sources used for the hydrolysis of glucuronide-conjugated opioids in the clinical setting. It is known that different enzymes can have varying degrees of hydrolysis efficiency. However, what hasn’t been widely appreciated is that different enzymes also have varying degrees of hydrolysis efficiency for different glucuronide-conjugated compounds; and that validation of a hydrolysis method observing only one glucuronide (e.g. morphine-3-glucuronide) can create a false sense of achieving 100 percent recovery of all glucuronide-bound analytes. This plays an important part in a clinical lab that tests patient samples for opiate compliance or abuse. In this study, three beta-glucuronidases (from Helix pomatia, Haliotis rufescens (Kura), and recombinant IMCSzyme (IMCS)) were optimized for concentration, incubation time, and incubation temperature and used to compare hydrolysis efficiency in matrix-matched standards and patient samples for five opiate compounds which are excreted as glucuronide metabolites.
Materials and Methods
Hydrolysis of opiate analytes was tested using beta-glucuronidase from Helix pomatia from Sigma-Aldrich, beta-glucuronidase from Haliotis rufescens from Kura Biotec, and a recombinant beta-glucuronidase from IMCS. A five-point calibration curve was extracted each day of method optimization and patient testing. Two UTAK liquid controls were included in each batch. The Kura and IMCS enzymes were optimized for a single-vial hydrolysis and were compared to our current method of hydrolysis using Helix pomatia that requires a transfer step before LC-MS/MS analysis on a Waters Xevo TQ-S. Below are the optimized conditions that were used for testing.
Helix pomatia (reference method)
Hydrolysis using Helix pomatia was performed in 8 mL glass culture screw-cap tubes. 100 µL of mixed opiate deuterium labeled internal standard (0.2/0.02 ng/µL in methanol) and 100 µL of urine sample were diluted with 2.9 mL of deionized water and hydrolyzed using 300 µL of a 5,000 U/mL Helix pomatia beta-glucuronidase in acetate buffer. The mixture was vortexed and then incubated at 50 degreesC for 90 minutes. After incubation, 4.6 mL of deionized water was added to the tubes followed by vortex mixing. The samples were then centrifuged at 3,000 RPM for 10 minutes. After centrifugation, 800 µL of sample was added to 200 µL of deionized water in autosampler vials. After crimping the autosampler vials, the samples were vortexed for homogeneity and analyzed. An injection volume of 10 µL was injected onto the mass spectrometer.
Haliotis rufescens (optimized protocol)
Hydrolysis using Haliotis rufescens beta-glucuronidase was performed in screw-cap autosampler vials. 75 µL of deionized water, 15 µL of mixed opiate deuterium labeled internal standard (0.2/0.02 ng/µL in methanol), and 15 µL of urine sample were pipetted separately into an autosampler vial. 5 µL of 1 mol/L of acetate buffer and 40 µL of beta-glucuronidase (final concentration of 25,000 U/mL) were added to the vial. The mixture was vortexed and then incubated at 68 degreesC for 90 minutes. After incubation, the vials were uncapped and 1,350 µL of deionized water was added. The vials were capped, vortexed, and centrifuged at 3,000 RPM for 10 minutes. An injection volume of 10 µL was used for analysis on the mass spectrometer.
Recombinant IMCSzyme (optimized protocol)
Hydrolysis using recombinant beta-glucuronidase was performed in screw-cap autosampler vials. 40 µL of deionized water, 15 µL of mixed opiate deuterium labeled internal standard (0.2/0.02 ng/µL in methanol), and 15 µL of urine sample were pipetted separately into an autosampler vial. 40 µL of Rapid Hydrolysis Buffer (proprietary) and 40 µL of beta-glucuronidase (final concentration of 14,667 U/mL) were added to the vial. The mixture was vortexed and then incubated at 55 degrees C for 30 minutes. After incubation, the vials were uncapped and 1,350 µL of deionized water was added. The vials were capped, vortexed, and centrifuged at 3,000 RPM for 10 minutes. An injection volume of 10 µL was used for analysis on the mass spectrometer.
A Waters XSelect HSS C18 2.5 µm, 2.1x150 mm XP column with a Phenomenex C18 SecurityGuard ULTRA Cartridge was used with Mobile Phase A of 5 mM Ammonium formate, pH 3.0, and Mobile Phase B of 0.1% formic acid in acetonitrile.
To optimize hydrolysis efficiency and calculate recovery of free drug, three mixed standards (A, B, and C) in drug-free urine were prepared containing 1,000 ng/mL of parent drug or the molar equivalents of the coinciding glucuronides. Standard A contained morphine, codeine, hydromorphone, and oxymorphone. Standard B contained morphine-6-glucuronide, codeine-6-glucuronide, and hyrdromorphone-3-glucuronide. Standard C contained morphine-3-glucuronide and oxymorphone-3-glucuronide. Because it is often overlooked in lab-developed opiate method validations, we included morphine-6-glucuronide in addition to morphine-3-glucuronide in order to see complete recovery of the parent drug. Of particular interest to us was the inclusion of all glucuronide species for which we report parent drug – as testing only morphine-3-glucuronide may not predict efficacy of hydrolysis of others[1]. Codeine-6-glucuronide and morphine-6-glucuronide have been described in the literature as requiring longer incubation times in order to achieve acceptable parent drug recovery[2,3]. Replicates of each standard were subjected to varying conditions in order to optimize hydrolysis. The conditions consisted of a concentration span ranging from 5,000 U/mL to 25,000 U/mL of beta-glucuronidase; incubation time points of 30, 60, and 90 minutes; and incubation temperatures of 50 degreesC, 55degrees C, and 68 degrees C.
For the patient specimen comparison (N=15), each specimen was tested 3 times, in separate batches over 3 days with extraction of each batch by a different analyst. Because of the small volumes pipetted, we sought to determine between run precision with a robust design in addition to differences in hydrolysis efficiency. Averages and % CVs were calculated between runs for each enzyme and Data Innovations EP Evaluator software was utilized to obtain supporting statistics.
Results
Results obtained from the hydrolysis optimization experiments with parent drug and glucuronide standards reflect those reported in the literature. Of the glucuronides tested, the recovery of parent drug from morphine-6-glucuronide and codeine-6-glucuronide was the most affected by differences in hydrolysis conditions for all enzymes.
For Helix pomatia beta-glucuronidase, recovery of parent drug was most dramatically changed by varying incubation time and amount of enzyme. Recovery of parent drug from morphine-6-glucuronide increased using 20,000 U/mL versus 5,000 U/mL of enzyme (99% versus 75%) and with an incubation time of 90 versus 30 minutes (99% versus 61%). Hydrolysis of codeine-6-glucuronide showed a similar pattern, with an increase in recovery of parent drug from 34% to 56% (concentration of 5,000 U/mL versus 20,000 U/mL) and 27% to 47% (incubation time of 30 versus 90 minutes).
For Haliotis rufescens beta-glucuronidase, incubation time was the variable that had the maximum influence on hydrolysis efficiency, with morphine-6-glucuronide showing an increase from 82% to 103% and codeine-6-glucuronide showing an increase from 66% to 97% recovery of parent drug (30 versus 90 minutes). Incubation temperature and concentration showed less of an influence on morphine-6-glucuronide; however, these parameters had a small effect on codeine-6-glucuronide, showing an increase in recovery from 86% to 98% (55 degreesC versus 68 degrees C) and 98% to 110% (concentration of 5,000 U/mL to 25,000 U/mL).
For recombinant IMCSzyme beta-glucuronidase, incubation time (30, 60, and 90 minutes) and enzyme concentration (11,000 U/mL to 25,667 U/mL) had minimal effect, illustrating less than a 20% difference for all analytes. The higher incubation temperature (68 degrees C versus 55 degrees C) however, showed a dramatic decrease in recovery (-60%) for all analytes except morphine-3-glucuronide.
Patient comparisons were performed using the protocols described above. Fifteen patient specimens (including one negative control patient) were chosen based on the reportable range. The maximum percent CV for all analytes with each enzyme was 11%, showing good reproducibility between runs of each enzyme and between each analyst.
The analytes of interest that showed the largest differences in hydrolysis efficiency were codeine, norcodeine, morphine, hydromorphone, and oxymorphone. Codeine had the most dramatic difference when comparing Helix pomatia (current method) to the Haliotis rufescens and the IMCSzyme enzymes (new enzymes), showing 5-12 fold increases in concentration with the new enzymes (e.g. 97 ng/mL versus 1,207 ng/mL). In contrast, the difference between the two new enzymes was less than 5% for codeine. Norcodeine also showed increased recovery of up to 200% with the Haliotis rufescens and the IMCSzyme enzymes as compared with Helix pomatia, with a less than a 15% bias between the new enzymes. Morphine showed increased recovery of 40% with the new enzymes, with a less than a 2% bias between the new enzymes. Hydromorphone and oxymorphone both showed increased recovery (50%) with the new enzymes compared with Helix pomatia. Recovery for hydromorphone and oxymorphone agreed well between the two new enzymes (<10% bias) but had greater variance (7-53%) versus the Helix pomatia enzyme, unlike morphine and codeine recovery. It is important to note that the other drugs included in this method which are not excreted as a glucuronide (e.g. 6-acetylmorphine, fentanyl, methadone, etc.) did not show significant differences between all the enzymes, signifying we did not compromise the integrity of these parent drugs.
Conclusion
Because 6-acetylmorphine has a short half-life (0.6 hours), morphine-to-codeine ratios have been proposed as a means of distinguishing heroin from codeine exposure[4]. However, Cone et al. states that due to high metabolic conversion variability within individuals, definitive morphine-to-codeine ratios cannot be conclusively made[5,6]. Furthermore, morphine-to-codeine ratios are also affected by the type of beta-glucuronidase used for hydrolysis. We observed two very diverse morphine-to-codeine ratios for the same patient sample using Helix pomatia versus IMCSzyme (1.8 versus 0.24), further evidence that opiate exposure is more complex than simple math.
These experiments illustrate the importance of testing multiple (if not all) glucuronidated analytes within a method during validation in order to determine hydrolysis efficiency. We have also demonstrated the significance of robustly optimizing and validating hydrolysis conditions for any beta-glucuronidase used for patient testing in order to recognize the potential for false-negative interpretations that could arise from incomplete hydrolysis. Conceivably, if proficiency testing material included glucuronides (instead of 100% free drug), clinical laboratories may become more alert to incomplete hydrolysis issues.
References & Acknowledgements:
References
1. P. Wang, J. A. Stone, K. H. Chen, S. F. Gross, C. A. Haller, and A. H. B. Wu. Incomplete Recovery of Prescription Opioids in Urine using Enzymatic Hydrolysis of Glucuronide Metabolites. Journal of Analytical Toxicology. 30: 570-575 (2006).
2. R. W. Romberg and L. Lee. Comparison of the Hydrolysis Rates of Morphine-3-Glucuronide and the Morphine-6-Glucuronide with Acid and β-Glucuronidase. Journal of Analytical Toxicology. 19: 157-162 (1995).
3. L. P. Hackett, L. J. Dusci, K. F. Ilett, and G. M. Chiswell. Optimizing the Hydrolysis of Codeine and Morphine Glucuronides in Urine. Therapeutic Drug Monitoring. 24: 652-657 (2002).
4. J. M. Colby, A. H. B. Wu, and K. L. Lynch. Analysis of Codeine Positivity in Urine of Pain Management Patients. Journal of Analytical Toxicology. 39: 407-410 (2015).
5. E. J. Cone, P. Welch, J. M. Mitchell, and B. D. Paul. Forensic Drug Testing for Opiates: I. Detection of 6-Acetylmorphine in Urine as an Indicator of Recent Heroin Exposure; Drug and Assay Considerations and Detection Times. Journal of Analytical Toxicology. 15: 1-7 (1991).
6. E. J. Cone, P. Welch, B. D. Paul, and J. M. Mitchell. Forensic Drug Testing for Opiates: III. Urinary Excretion Rates of Morphine and Codeine Following Codeine Administration. Journal of Analytical Toxicology. 15: 161-166 (1991).
| Description | Y/N | Source |
| Grants | no | |
| Salary | no | |
| Board Member | no | |
| Stock | no | |
| Expenses | no |
IP Royalty: no
| Planning to mention or discuss specific products or technology of the company(ies) listed above: | no |