Stability investigations of cytochrome P450 (CYP) enzymes immediately after death in a pig model support the applicability of postmortem hepatic CYP quantification

Abstract Quantification of drug‐metabolizing cytochrome P450 (CYP) isoforms using LC–MS/MS has been proposed as a potential way of estimating antemortem CYP levels using postmortem tissue, but the postmortem stability of CYP proteins is incompletely investigated. If one can use data obtained from the analysis of postmortem specimens to inform physiologically based pharmacokinetic (PBPK) models this greatly increases the access to rare specimens among special subpopulations. In this study, we developed and validated an LC–MS/MS method for targeted CYP protein quantification in a porcine animal model to study postmortem stability. We measured 19.9–28.3 pmol CYP1A2, 50.3–66.2 pmol CYP2D25, 132.9–142.7 pmol CYP2E1, and 16.8–48 pmol CYP3A29 protein per mg PLM in nondegraded tissue. In tissue stored at 4°C, we found that the CYP protein levels were unaffected by degradation after 72 h. At 21°C CYP1A2, CYP2D25, and CYP2E1 protein levels were nearly unaffected by degradation after 24 h, whereas a loss of approximately 50% was seen after 48 h. At 21°C CYP3A29 had a loss of 50% at 24 h and 70% at 48 h exhibiting less postmortem stability. In vitro enzyme activity measurements in the same tissue stored at 21°C showed a 50% decrease after 24 h and a complete loss of enzyme activity after 48 h. When stored at 4°C, the in vitro enzyme activity decreased to 50% activity after 96 h. In conclusion, measuring CYP levels by an LC–MS/MS approach was clearly less affected by postmortem changes than an activity‐based approach. The found postmortem stability for 24 h at 21°C for 3 out of 4 CYP isoforms supports the use of properly stored postmortem tissue to inform PBPK models.


| INTRODUC TI ON
Physiologically based pharmacokinetic (PBPK) models are increasingly being used to predict human drug disposition across subpopulations. [1][2][3] Hepatic drug metabolism by the cytochrome P450 (CYP) family of enzymes is a major elimination pathway for a large number of drugs and thus affects drug disposition.
Therefore, information regarding CYP levels in subpopulations is Individuals suffering from cirrhosis have markedly reduced activity of CYP1A2 and CYP3A4. 6 A limitation in the aforementioned studies is that in vitro or in vivo CYP-activities have been used to estimate altered CYP protein levels in a given subpopulation. The apparent change in CYP activity in a subpopulation might be a result of different CYP protein levels in combination with e.g., altered drug transporter levels and altered blood flow. Therefore the true altered CYP protein levels in a given subpopulation might not be known. Access to relevant hepatic tissue is scarce. If CYP protein levels could be estimated in postmortem hepatic tissue taken at an autopsy, this would dramatically ease the access to relevant hepatic tissue for special subpopulations, but also raise the question: What is the postmortem timeframe in which CYP protein levels reflect antemortem levels?
Tissue for proteomic research is generally snap-frozen in, for ex- We chose a pig model to study the postmortem stability of CYP proteins. Pigs are highly suitable to study human biotransformation due to the similarities in anatomy and physiology and a good model for studying postmortem protein degradation. 9,12 Many porcine CYP enzymes exhibit more than 80% sequence identity to human drugmetabolizing CYP enzymes, 13,14 and the enzyme activities of key porcine CYP subfamilies have been studied in vitro with probe substrates commonly used for human studies. 15,16 Furthermore, porcine CYP enzymes are evenly expressed in the lobes of the liver and at similar levels as human CYP enzymes. [17][18][19] As a first step to justify using CYP quantification data from forensic postmortem hepatic tissue in PBPK modeling, the aim of the current study was to investigate the postmortem stability of CYP enzymes immediately after death and up to 7 days in a porcine model. We developed a quantitative LC-MS/MS proteomics method based on the AQUA quantification strategy 20 targeting four porcine CYP isoforms, CYP1A2, CYP2D25, CYP2E1, and CYP3A29, which exhibit high sequence identity (>76%) to human CYP1A2, CYP2D6, CYP2E1, and CYP3A4/5, respectively (Table S1). We used the developed method to quantify the CYP proteins in porcine liver microsomes (PLM) from nondegraded hepatic tissue (PMI = 0) and hepatic tissue degraded in vitro at 4°C and 21°C (PMI = 2-168 h).
Finally, we wanted to gain further insight into the stability of CYP enzymes by investigating the in vitro enzymatic activity based on the metabolite formation associated with phenacetin and midazolam metabolism. To our knowledge, this is the first study investigating the postmortem stability of CYP enzymes in hepatic tissue 0-24 h and the first quantification of porcine CYP1A2, CYP2D25, CYP2E1, and CYP3A29 using targeted mass spectrometry.

| Reagents and solutions
Reagents were purchased from Sigma-Aldrichunless otherwise stated. Methyl methanethiosulfonate (MMTS) was dissolved in 2-propanol at a concentration of 80 mM, and TCEP tris(2carboxyethyl)phosphine (TCEP) was prepared as 35 mM in 1.5 M ammonium bicarbonate (Ambic). Paracetamol was purchased from Merck. Midazolam was from Toronto Research Chemicals, and alpha-hydroxymidazolam was from Cerilliant. The NADPH regenerating system (RAPID start K5100) was from XenoTech. The Bradford protein assay for MPPGL determination was from Bio-Rad, and TPCK-treated trypsin for protein digestion was from AB-SCIEX.

| Porcine liver samples and stability assay
Porcine livers from Danish slaughter pigs (Sus scrofa domesticus) (n = 3, female, 6 months old) were placed on ice immediately after sacrifice. The pigs were cared for according to the Danish Animal Welfare Act 2013. Furthermore, the animals were sacrificed as part of meat production before entering our study, and therefore no ethical approval was needed according to the Danish ministerial order on animal experimentation (BEK nr.2028 of 14/12/2020). For each liver, approximately 2 g of tissue was frozen at −80°C, and these samples were annotated as t = 0. Large pieces (>1 kg) from each liver were stored in plastic bags and incubated at 4 or 21°C before the samples (~2 g) were dissected and frozen at t = 2, 4, 6, 20, 24, 48, 72, 96, 120, 144, or 168 h. The liver pieces at 4°C were sampled for up to 7 days (168 h), whereas the liver pieces at 21°C were only sampled up to 72 h due to advanced putrefaction.

| Preparation of porcine liver microsomes
Liver tissue (500 mg) was thawed and minced with a scalpel. The minced tissue was added to 3 ml of cold homogenization buffer (0.1 M potassium phosphate buffer with 0.125 M potassium chloride and 1 mM EDTA, pH 7.5) and homogenized using a Potter-Elvehjem glass homogenizer (VWR) and a Teflon pestle which was driven by a RW16 motor unit (IKA) at speed level 6. The homogenate was centrifuged at 9000g for 20 min at 4°C to generate the S9 frac- as the determined microsomal protein concentration divided by the amount of liver used. MPPGL is not corrected for process loss. 21 The MPPGL measurements were used to adjust the PLM concentrations, ensuring that equal amounts of microsomal protein from each sample preparation were used for CYP quantification and enzyme activity measurements.

| Peptide selection
Tryptic peptides suitable for LC-MS/MS quantification based on the AQUA strategy 20 were selected based on information from PeptideAtlas (ISB), 22 Skyline 19.1 peptide prediction software, 23 and the literature. 24 For each porcine CYP protein, four to six unique synthetic peptides (SpikeTides, JPT) were manually tuned by direct infusion into the mass spectrometer and subsequently evaluated in terms of signal intensity and linearity. The two best-performing peptides for each CYP protein were acquired as quantified synthetic analog stable isotope-labeled internal standard peptides (SIL-IS peptides) containing C-terminally isotopic labeled Arg ( 13 C 6 15 N 4 ) or Lys ( 13 C 6 15 N 2 ) (SpikeTides TQL, JPT) (Tables S1 and S2).

| Trypsin digestion
The isolated PLM were subjected to enzyme digestion following a previously described protocol, with minor deviations. 11 In brief, all pPLM. The precision was determined as within-and between-runs variation calculated using a one-way ANOVA approach.

| Trueness
The trueness of the method was evaluated at the peptide level by

| Calculation of confidence intervals
The 95% confidence interval (CI) of the mean value, of two or three different livers, at each time point was calculated using an assumption of equal variance at different time points. The standard deviation for the 95% CI was calculated as the within groups standard deviation using one-way ANOVA.

| Postmortem stability of microsomal proteins
The yield of total microsomal protein (MPPGL) in liver 1 was stable for 168 h at 4°C (Figure 1). For liver 2 and 3 at 4°C the MPPGL levels were stable for approximately 96 h, after that a decline of approx.
30% was seen. When the tissues were stored at 21°C, the MPPGL levels were stable for at least 24 h. After 48 h of storage at 21°C, over 60% of the t = 0 levels were still measurable, whereas at 72 h, the MPPGL levels were severely affected by putrefaction and less than 25% of the t = 0 levels were detected. The 95% CI in Figure 1 was at 4°C calculated using liver 2 and 3 since liver 1 had a different degradation profile, at 21°C all three livers were used to calculate the 95% CI. The absolute levels of MPPGL are shown in Table 1

| Validation of LC-MS/MS method
To investigate the postmortem stability of CYP proteins specifically, we developed a LC-MS/MS method for quantification of four CYP protein isoforms in PLM based on quantification of unique signature peptides (Table S1). The developed method was validated in a nondegraded matrix (t = 0) in terms of method precision, method recovery, trueness of the method, LOD, and LLOQ. The performances of the quantifying peptides are presented in Table 2.

| Stability of CYP proteins in postmortem hepatic tissue
At 4°C a similar pattern to MPPGL stability was seen for CYP1A2, CYP2D25, CYP2E1, and CYP3A29 ( Figure 2A) Figure 2B). When the tissue was stored at 21°C for 48 h, a significant loss in CYP protein content was observed and approximately 50-60% of the t = 0 h levels were still measurable after 48 h for these three CYP isoforms.
For CYP3A29 at 21°C a different profile was seen. During the first 6 h there was no major decrease, but after 20 h a decline of approximately 40%-50% was seen. The 95% CI at 4°C (Figure 2A) was calculated using liver 2 and 3 since liver 1 had a different degradation profile, at 21°C ( Figure 2B) all three livers were used to calculate the 95% CI. The absolute concentrations of the CYP isoforms are shown in Table 1.

| Enzymatic activity in postmortem hepatic tissue
In vitro enzyme activity rates of CYP1A were estimated in PLM The enzyme activity rates were linear between 0.1 and 1.5 mg/ml PLM for both substrates, and initial reaction rates were observed for reaction times up to 20 min (data not shown). There is a complete chromatographic separation between paracetamol and the internal standard acetaminophenol, assuring no interference of these isobaric compounds.
The in vitro enzymatic activity rates at t = 0 were measured at 1600-6000 pmol/min/mg PLM for the formation of paracetamol and 1000-3800 pmol/min/mg PLM for the formation of alphahydroxymidazolam for the three porcine livers (Table 1). A high degree of variability was seen in the data; therefore, the standard deviation was calculated for each time point for the three livers  (Figure 1). Therefore the rate at which CYP-proteins degrade postmortem seems to vary not only between livers but also between CYP-isoforms. The observation that CYP isoforms have individual degradation profiles at 21°C need further investigation due to the limited data presented here, but this study firmly demonstrates that temperature is an important factor for the degradation rate. Some studies have found proteins to degrade much faster than the CYP proteins investigated in this study whereas others have found certain muscle proteins to be stable for up to 10 days postmortem. [7][8][9][10] In nondegraded tissue, we found an average of 23.3 pmol/mg PLM CYP1A2 and 33.2 pmol/mg PLM CYP3A29 among the livers, suggesting that these CYP isoforms are relatively low abundant.
CYP2D25 and CYP2E1, on the other hand, seem to be expressed at the higher levels of 61.7 pmol/mg PLM and 138.3 pmol/mg PLM, respectively (Table 1). To our knowledge, these porcine CYP proteins have only been quantified by untargeted proteomics until now. 24,27 In agreement with the present study, both papers reported CYP1A2 TA B L E 1 CYP protein expression levels and activity rates in nondegraded porcine liver tissue. The CYP protein levels and the enzyme activity rates are reported as the mean of two PLM (t = 0) preparations from each liver proved the relevance of this study, and we, therefore, found the in vitro degradation setup sufficient for the purpose of this study.
The timespan from death to when an autopsy is performed varies greatly between countries and between different specific cases.
Not only the temperature at the place of death and PMI, but also storage conditions until autopsy will be important for the decay of postmortem tissue. Hepatic CYP quantification will therefore only make sense in selected cases. can further be informed by physiological data noted at autopsy e.g., height, body weight, the weight of the liver-heart-brain-lungskidneys, and diseases and medication history.
In this study, we only investigated the stability of CYP enzymes but it is obvious that information regarding the stability of other proteins involved in drug elimination like UGTs, and transporters would be interesting as well. Such a broad investigation was out of the scope of this work.
In conclusion, we found that postmortem CYP levels quantified by LC-MS/MS for 3 out of 4 CYP isoforms were comparable to antemortem CYP levels for 24 h if the tissues have been stored at 21°C and for 3 days if the tissue has been stored at 4°C for all four CYP isoforms. The use of readily available postmortem specimens for collecting information about protein abundances could increase the availability of data on CYP levels which is important to understanding drug disposition in subpopulations.