Abstract
Rat CYP2D1 has been considered as an ortholog of human CYP2D6. To assess the role of CYP2D1 in physiologic processes and drug metabolism, a CYP2D1-null rat model was generated with a CRISPR/Cas9 method. Seven base pairs were deleted from exon 4 of CYP2D1 of Sprague-Dawley wild-type (WT) rats. The CYP2D1-null rats were viable and showed no abnormalities in general appearance and behavior. The metabolism of venlafaxine (VLF) was further studied in CYP2D1-null rats. The Vmax and intrinsic clearance of the liver microsomes in vitro from CYP2D1-null rats were decreased (by ∼46% and ∼57% in males and ∼47% and ∼58% in females, respectively), while the Michaelis constant was increased (by ∼24% in males and ∼25% in females) compared with WT rats. In the pharmacokinetic studies, compared with WT rats, VLF in CYP2D1-null rats had significantly lower apparent total clearance and apparent volume of distribution (decreased by ∼36% and ∼48% in males and ∼23% and ∼25% in females, respectively), significantly increased area under the curve (AUC) from the time of administration to the last time point, AUC from the start of administration to the theoretical extrapolation, and Cmax (increased by ∼64%, ∼59%, and ∼26% in males and ∼43%, ∼35%, and ∼15% in females, respectively). In addition, O-desmethyl venlafaxine formation was reduced as well in CYP2D1-null rats compared with that in WT rats. Rat depression models were developed with CYP2D1-null and WT rats by feeding them separately and exposing them to chronic mild stimulation. VLF showed better efficacy in the WT depression rats compared with that in the CYP2D1-null rats. In conclusion, a CYP2D1-null rat model was successfully generated, and CYP2D1 was found to play a certain role in the metabolism and efficacy of venlafaxine.
SIGNIFICANCE STATEMENT A novel CYP2D1-null rat model was generated using CRISPR/Cas9 technology, and it was found to be a valuable tool in the study of the in vivo function of human CYP2D6. Moreover, our data suggest that the reduced O-desmethyl venlafaxine formation was associated with a lower VLF efficacy in rats.
Introduction
Cytochrome P450 enzymes are part of an extended family of hemoproteins and are an important family of oxidases in the microsomal mixed-function oxidase system (Yamazaki, 2014). To date, a total of 57 cytochrome P450 isoforms have been identified in humans, which are grouped into 18 families and 44 subfamilies by their sequence similarity (Zanger et al., 2008). Among them, the CYP1, CYP2, and CYP3 subfamilies are responsible for the biotransformation of most foreign substances including 70%−80% of clinically used drugs (Guengerich, 2008; Zhou et al., 2012).
Although being only 4% of the total cytochrome P450 enzyme in the human liver, CYP2D6 is associated with biotransformation of approximately 30% of commonly used drugs (Teh and Bertilsson, 2012). Substrates of CYP2D6 include a variety of antiarrhythmics, antipsychotics, and second-generation and tricyclic antidepressants (Zanger and Schwab, 2013). It has been reported that the frequency of the CYP2D6 allele varies by racial/ethnic groups; about 6%−10% of Caucasians and only 2% of Asians are CYP2D6 poor metabolizers (PMs) (Meyer et al., 1990; Bradford, 2002). If the amount of drug administered to CYP2D6 PMs is the same as that administered to normal individuals, serious adverse reactions are more likely to occur in PMs because the metabolism of the drug is often unpredictable in PMs (de Leon et al., 2005; Hertz et al., 2015). Further research is needed to improve the efficacy of drugs, reduce the incidence of adverse reactions, and achieve individualized drug therapies.
Venlafaxine (VLF) is an antidepressant of selective serotonin and norepinephrine (NE) reuptake inhibitors (Bymaster et al., 2001). In humans, VLF was metabolized to O-desmethyl venlafaxine (ODV) by CYP2D6 (∼89%), CYP2C19 (∼10%), and CYP2C9 (∼1%), and venlafaxine is also metabolized by CYP3A4 to N-desmethyl venlafaxine (Magalhães et al., 2014). ODV is the main metabolite of VLF to exert pharmacological activity (Wellington and Perry, 2001). For the rat CYP2D subfamily, it was reported that six members (CYP2D1, -2D2, -2D3, -2D4, -2D5, and -2D18) have been identified by genomic analysis and CYP2D1 was considered as an ortholog of human CYP2D6 (∼83% homology) (Nelson et al., 1996; Martignoni et al., 2006).
A variety of cytochrome P450 gene knockout mouse models and cytochrome P450–humanized mouse models have been reported to illustrate the function of cytochrome P450 isoforms on drug metabolism, toxicity, and carcinogenicity (Scheer et al., 2012, 2014; Bissig et al., 2018). However, the mouse model may have some limitations. For example, the small volume of blood samples collected from mice increases the difficulty of pharmacokinetic studies. Moreover, the mouse model is poorly tolerated in some experiments, such as in bacterial infection and chronic heart failure experiments (Robinson et al., 1968; Li et al., 2004; Handa et al., 2009). Compared with the mouse model, the rat model could be more suitable for certain experiments.
In 2013, a knockout rat model was generated using the CRISPR/Cas9 method (Li et al., 2013), and since then knockout rats have become an accessible tool for studying human cytochrome P450 functions. It has been reported that the metabolism of chlorzoxazone is significantly reduced in CYP2E1 knockout rats (Wang et al., 2016), while the metabolism of nifedipine and midazolam were reduced as well in CYP3A1/2 double knockout rats (Lu et al., 2017). A CYP2C11 knockout rat model was successfully generated by our group and was used to study the metabolism of tolbutamide and warfarin (Wei et al., 2018; Ye et al., 2019).
To better understand the role of human CYP2D6 in drug metabolism, in this study, a unique CYP2D1-null rat model was generated using the CRISPR/Cas9 technology. The general phenotypes were characterized and the hepatic expression of major cytochrome P450s was determined in the knockout rat model. Venlafaxine, an inhibitor of serotonin-norepinephrine reuptake and a substrate of human CYP2D6, was used as a probe drug to study the metabolism and efficacy function of CYP2D1 in rats.
Materials and Methods
Chemicals and Reagents.
NADPH (>98% purity) was purchased from Aladdin Industrial Co. Ltd. (Shanghai, China). Carvedilol (>99% purity) was purchased from Aladdin Bio-Chem Technology Co. Ltd. (Shanghai, China). Venlafaxine hydrochloride and O-desmethyl venlafaxine (>98% purity) were purchased from Sam Chemical Technology Co., Ltd. (Shanghai, China). Acetonitrile and methanol, as well as all other reagents, were purchased from Sinopharm Chemical Reagent Co. Ltd. (Shanghai, China).
Generation of CYP2D1-Null Rats by CRISPR/Cas9.
Three specific single guide RNA (sgRNA) target sequences were selected from the target site in exon 4 of the rat CYP2D1 gene (Gene ID: 266684) using Gene Knock-Out with Cas9 software from Xunqi Biotechnology Co. Ltd. (Nanjing, China). Three pairs of guiding primers were designed: sgRNA1 (forward primer: 5′-GCAGCATGGCCTTGGGATTGA-3′, reverse primer: 5′-TCAATCCCAAGGCCATGCTG-3′), sgRNA2 (forward primer: 5′-AGACCCTTACCTCATCAGGA-3′, reverse primer: 5′-TCCTGATGAGGTAAGGGTCT-3′), and sgRNA3 (forward primer: 5′-CTAGTTTCACCATCCTGATG-3′, reverse primer: 5′-CATCAGGATGGTGAAACTAG-3′). The sgRNAs were transcribed in vitro using the T7 Quick High Yield RNA Synthesis Kit, recovered by phenol-chloroform extraction and alcohol precipitation, and finally dissolved in nuclease-free water. The pST1374-cas9 vector was linearized by Agel (New England Bioscience, Ipswich, MA), transcribed into Cas9 mRNA, and purified with the RNeasy Mini Kit (Qiagen, Valencia, CA). The constructed Cas9 mRNA was coinjected with sgRNA1, sgRNA2, and sgRNA3 (Cas9 mRNA: 50 ng/μl; sgRNA1: 100 ng/μl; sgRNA2: 100 ng/μl; and sgRNA3: 100 ng/μl) into zebrafish zygotes to determine their activities, respectively.
The sgRNA3 set primer (30 ng/μl) was injected into wild-type (WT) Sprague-Dawley rat monocytic embryos with Cas9 mRNA (50 ng/μl) since sgRNA3 showed the highest activity in zebrafish (Supplemental Table 1). The obtained CYP2D1+/− rat (F0, 1 female) was hybridized with a WT rat (1 male) and the pups were subjected to polymerase chain reaction (PCR) amplification and genomic DNA sequencing to select the heterozygous CYP2D1+/− rats (F1). The F1 generation rats were inbred (1 male: 2 females) to produce F2 generation rats, and then the homozygous CYP2D−/− rats were screened out from the F2 generation rats by the same method. The homozygous CYP2D−/− rats were transferred from the Model Animal Research Center at Nanjing University to the Laboratory Animal Center at Jiangsu University for further breeding and animal experiments. The animal facilities at Nanjing University and Jiangsu University have passed the certification of the Association for Assessment and Accreditation of Laboratory Animal Care. All animal breeding and experimental procedures were performed in accordance with the Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committees at Nanjing University and Jiangsu University.
Off-Target Analysis.
The sgRNA3 sequence (CTAGTTTCACCATCCTGATG) was submitted to TagScan (http://ccg.vital-it.ch/tagger) to predict the possible off-target sites. The detailed sequences around the potential off-target sites were obtained from the University of California at Santa Cruz database (http://genome.ucsc.edu/). The corresponding primers were designed for the potential off-target sites and PCR was performed. The PCR products were then sequenced by Sangon Biotech Co. Ltd. (Shanghai, China) and compared with those from the University of California at Santa Cruz database.
General Characterization of CYP2D1-Null Rats.
The appearance, fertility, and general behavior of CYP2D1-null rats were observed. CYP2D1-null rats at 12 weeks of age (six rats per group, male and female) were sacrificed by exposure to an ascending concentration of CO2 and dissected for further study. The viscera indices (defined as the ratio of tissue weight to body weight) were calculated and compared with those of WT rats.
Analysis of the Expression Levels of Major Cytochrome P450s.
The rat liver microsomes were prepared by differential centrifugation using a previously described method (Egorin et al., 1997). The concentration of liver microsomes was determined using the Enhanced Bicinchoninic Acid Protein Assay Kit (Beyotime, Nantong, China) and the expression of major subtypes of cytochrome P450 in the CYP2D1-null and WT rats were determined by western blotting. The common loading controls such as glyceraldehyde-3-phosphate dehydrogenase and β-actin could not be used in the experiment since most of them had been removed during microsomal preparation. Therefore, we used Ponceau S stain on the duplicate gel to guarantee that the loading amount of protein per lane (5 µg/lane) was consistent (Gu et al., 2003). The blots were incubated with CYP1A2 (sc-53241, mouse monoclonal IgG1; Santa Cruz Biotechnology, CA), CYP2A (PAB19502, rabbit polyclonal; Abnova, Taipei, Taiwan, China), CYP3A (sc-25845, rabbit polyclonal; Santa Cruz Biotechnology), CYP2B (sc-73546, mouse monoclonal IgG1; Santa Cruz Biotechnology), CYP2C11 (PA3-034, rabbit polyclonal; Thermo Fisher Scientific, Waltham, MA), CYP2D1 (AB1271, rabbit polyclonal; EMD Millipore Corporation, Temecula, CA), CYP2E1 (BML-CR3271-0100, rabbit polyclonal; Enzo, NY), and NADPH-cytochrome P450 oxidoreductase (sc-55477, mouse monoclonal; Santa Cruz Biotechnology) antibodies, respectively, at room temperature for 2 hours, followed by incubation with the corresponding secondary antibodies for 1 hour. Chemiluminescence solution was added to the blot and the ChemiDoc XRS imaging system (BioRad, Hercules, CA) was used to visualize the images.
mRNA Analysis of CYP2D Genes.
Total RNA was isolated from livers of WT and CYP2D1-null rats using Trizol reagent (CW0580S; CWBIO, Beijing, China) according to the manufacturer’s instructions. To detect selected cytochrome P450 expression in mRNA level, cDNA was prepared from total RNA using the RevertAid First Strand cDNA Synthesis Kit (K1622; Thermo Fisher Scientific). β-Actin was set as the internal reference. Selected genes for quantification in livers of WT and CYP2D1-null rats were CYP2D1, 2D2, 2D3, 2D4/18, and 2D5, and the primers designed for quantification are listed in Supplemental Table 2.
VLF Metabolism in CYP2D1-Null Rats.
Liver microsomes (1 mg/ml) and various concentrations of VLF (7.5, 10, 15, 20, or 25 μM) were mixed and preincubated at 37°C in a water bath for 5 minutes, followed by adding β-NADPH (1 mM) and MgCl2 (5 mM) to initiate the reaction in a total volume of 200 μl, and then incubated at 37°C in a water bath for 50 minutes. The reaction was then stopped with 800 μl of ice-cold ethyl acetate containing 50 μg/ml carvedilol, which was used as an internal standard solution. The upper organic phase was collected by vigorous vortexing for 3 minutes and centrifugation at 9600g for 10 minutes at 4°C. Then, 400 μl of ethyl acetate was used to extract the precipitate for the second time. The combined organic phase was dried with nitrogen and reconstituted with 100 μl acetonitrile for high-performance liquid chromatography analysis.
For determination of the pharmacokinetic parameters, CYP2D1-null and WT rats were divided into four groups (six rats per group, 12 week olds, male and female). The rats were fasted for 12 hours before the experiment but were provided water. After administration of a single dose of VLF (20 mg/kg, intragastric administration), approximately 0.3 ml blood was collected from the orbital vein at different time points (0.25, 0.5, 1, 2, 4, 8, 12, 24, and 36 hours) in both CYP2D1-null and WT rats. The blood samples were centrifuged at 1500g for 10 minutes to obtain plasma. Next, 50 μl of 1.0 M hydrochloric acid solution and 10 μl of 50 μg/ml carvedilol (internal standard) were added into 100 μl plasma and thoroughly mixed, and then 200 μl of ethyl acetate was added and vortexed for 3 minutes. The supernatant was collected by centrifugation at 9600g for 10 minutes. The precipitate was re-extracted with 200 μl ethyl acetate. The combined supernatant was dried with nitrogen and redissolved in 100 μl acetonitrile for high-performance liquid chromatography analysis.
An Agilent Eclipse Plus C18 (5 μm, 4.6 × 250 mm; Agilent Technologies, Santa Clara, CA) column was used at room temperature. The flow rate was 0.8 ml/min and the wavelength for detection was set at 228 nm. The mobile phase consisted of 60% acetonitrile and 40% 0.01 mol/l ammonium acetate buffer (pH 4.8), and the injection volume was 20 μl.
Development of the Rat Model of Depression.
The development of the depression model was based on previous literature and in accordance with our laboratory conditions (Pucilowski et al., 1993; Zhao et al., 2008). Briefly, the depression models in WT and CYP2D1-null rats were developed by feeding them separately and exposing them to chronic mild stimulation (CMS). In the first and second week, the CMS groups were given eight kinds of stimulations, including tail suspension for 5 minutes, rat cage tilting at 45°, humid padding, overcrowding, electric shock (36 V, 2 times/min), forced swimming, alternated light and dark, and fasting and water deprivation for 24 hours. From day 15 to day 20, ice water (4°C, 5 minutes) was added in the bath coupled with heat stress (45°C, 5 minutes). The same kind of stimulation was intermittently used to prevent the rats from anticipating the upcoming stimulation (Supplemental Table 3). During the development of the rat model of depression, all CMS rats were fed with one rat per cage except in the overcrowding stimulation experiment. The control groups were bred normally with five rats per cage.
Body Weight Test.
The effects of depression on body weight in rats were investigated with reference to previous literature (Liu et al., 2011). Rats in each group were weighed at 9:00 AM on day 21 of the depression modeling.
Locomotor Activity Test.
The experiment was performed following the method used in a previous report (Zhao et al., 2008). On day 21 of depression modeling, rats were individually placed in an open box of 80 × 80 × 50 cm. A TSE Systems multifunction conditioning system (Thuringia, Germany) was used to automatically record the trajectory of the movement within 8 minutes and calculate the total distance of locomotor activity.
The 1% Sucrose Water Consumption Test.
The experiment was performed following a previous study (Kim et al., 2003). After 4 hours of locomotor activity test, the rats were fasted with no food or water for 24 hours, and then 1% sucrose water was provided to rats for another 24 hours. The 1% sucrose water consumption for 24 hours was determined with the ratio of intake volume (milliliters) to rat body weight (100 g).
Efficacy of Venlafaxine Hydrochloride in the Rat Model of Depression.
After CMS modeling, the rats (five rats per group, 12 weeks old, male and female) were administered with venlafaxine hydrochloride (CMS + VLF groups, 20 mg/kg, intragastrical administration) or 2 ml 0.9% saline (CMS + saline groups) every day for 14 days. Then, body weight, locomotor activity, and 1% sucrose water consumption were also evaluated. After these analyses, approximately 1.0 ml of blood was taken from the orbital vein of each rat, and plasma was prepared by 1500g centrifugation for 20 minutes. The levels of NE and 5-hydroxytryptamine in plasma were determined by an ELISA kit (SenBeiJia, Nanjing, China). The rats were sacrificed by CO2 after blood sampling.
Statistical Analyses.
The experimental data were analyzed using Prism 5.0 statistical software (GraphPad, La Jolla, CA). The experimental data are presented as mean ± S.D. One-way ANOVA was performed and a value of P < 0.05 was considered statistically significant.
Results
Generation of CYP2D1-Null Rats.
Two rat founders with different genotypes, 7-base pair (bp) (ACCTCAT) deletion, or 1-bp (A) insertion in exon 4 were generated. Due to the infertility of the other one (with 1-bp insertion), only the founder with 7-bp deletion was used for further studies. The generation scheme of the CYP2D1-null rat model is shown in Fig. 1A. The PCR amplification results of the CYP2D1-null and WT rats are shown in Fig. 1B. The DNA fragments show bright bands around 586 bp. The homozygous CYP2D1-null rats lacked the 7 bp (ACCTCAT) at 270–280 bp compared with the WT rats, and the heterozygous CYP2D1-null rats presented mixed peaks in the corresponding area (Fig. 1C). Based on the genotyping results, CYP2D1-null rats were successfully generated and the genotype-stable offspring were obtained.
Off-Target Analysis.
Nine potential off-target sites of sgRNA3 (Supplemental Table 4) were predicted using the off-target detection software and the corresponding primers (Supplemental Table 5). The sequencing results showed that these potential off-target sites were not mutated in the CYP2D1-null rats.
General Characterization of CYP2D1-Null Rats.
There were no significant differences in the appearance, fertility, and general behavior between the CYP2D1-null and WT rats, in either males or females (Supplemental Tables 6 and 7). The data obtained for viscera indices are given in Table 1. The viscera indices of CYP2D1-null rats also showed no significant difference compared with those of WT rats.
Hepatic Expression Levels of Major Cytochrome P450s.
The hepatic expression levels of major cytochrome P450s in CYP2D1-null and WT rats are shown as Fig. 2. CYP2D1 protein was only detected in the WT rats, but not in the CYP2D1-null rats. The expression levels of the other cytochrome P450s showed no significant differences, except for the expression of CYP2A, which was significantly higher in the CYP2D1-null rats compared with that in the WT rats, for both males and females. Notably, CYP2C11 was not detected in either the WT females or CYP2D1-null females.
mRNA Expression Levels of CYP2D Genes.
A greatly reduced mRNA level of the targeted CYP2D1 gene was detected in the livers of CYP2D1-null rats (Supplemental Fig. 1). Among the mRNA expression of other CYP2D genes, CYP2D2 and CYP2D3 were upregulated and CYP2D4/18 and CYP2D5 were downregulated (Supplemental Fig. 2).
Venlafaxine Hydrochloride Metabolism in CYP2D1-Null Rats.
The VLF in vitro activities were determined with liver microsomes from both WT and CYP2D1-null rats. The kinetic parameters of the Michaelis constant, maximum velocity (Vmax), and intrinsic clearance were calculated and are given in Table 2. There was no sex difference in the model, and the Vmax and intrinsic clearance of CYP2D1-null males decreased by 46% (P < 0.05) and 57% (P < 0.05), respectively, while the Michaelis constant increased by 24% (P < 0.05) compared with WT males. The Vmax and intrinsic clearance of the CYP2D1-null females decreased by 47% (P < 0.05) and 58% (P < 0.05), respectively, while the Michaelis constant increased by 25% compared with WT females.
In the pharmacokinetic study, the plasma levels of VLF in the CYP2D1-null rats were significantly higher than those in the WT rats (at marked time points, see Fig. 3, A and B), while the levels of ODV were significantly lower than those in the WT rats (at marked time points, see Fig. 3, C and D). Compared with the WT males, VLF in the CYP2D1-null males had significantly lower apparent total clearance (CL/F) (decreased by ∼36%, P < 0.05) and apparent volume of distribution (decreased by ∼48%, P < 0.05), and significantly increased area under the curve from the time of administration to the last time point (AUC0–t) (increased by ∼64%, P < 0.05), area under the curve from the start of administration to the theoretical extrapolation (AUC0–∞) (increased by ∼59%, P < 0.05), and Cmax (increased by ∼26%, P < 0.05). VLF in the CYP2D1-null females also had significantly lower CL/F (decreased by ∼23%, P < 0.05) and apparent volume of distribution (decreased by ∼25%, P < 0.05), and had significantly higher AUC0–t (increased by ∼43%, P < 0.05), AUC0–∞ (increased by ∼35%, P < 0.05), and Cmax (increased by ∼15%, P < 0.05) compared with WT rats (Table 3).
Body Weight, 1% Sucrose Water Consumption, and Locomotor Activity after Depression Modeling.
There were significant differences (P < 0.05) in body weight, 1% sucrose water consumption, and locomotor activity in the CMS groups compared with the control groups (Table 4). These results suggested that the depression modeling in both strains was successful.
Efficacy of Venlafaxine Hydrochloride in the Rat Depression Model.
After 14 days of administration of VLF, three depression indicators (e.g., body weight, 1% sucrose water consumption, and locomotor activity) were recorded and the results are given in Table 5. All of the three indicators in the VLF-treated groups (CMS + VLF groups) were significantly increased compared with the saline-treated groups (CMS + saline groups) for WT rats (P < 0.05). These indicators were slightly improved (P > 0.05) in the VLF-treated CYP2D1-null rats compared with the saline-treated groups, except for the 1% sucrose water consumption in CYP2D1-null females. With VLF treatments, these indicators in WT rats were significantly higher (P > 0.05) than in the CYP2D1-null rats. Furthermore, with VLF treatments, the plasma NE and 5-hydroxytryptamine levels in WT rats were significantly higher than in the CYP2D1-null rats in both males and females (Fig. 4).
Discussion
Rat and human CYP2D isoforms share a good homology (>70%), and rat CYP2D1 (∼83% homology) is known as an ortholog of human CYP2D6 (Venhorst et al., 2003; Martignoni et al., 2006). Rat CYP2D1 shares high homology with other CYP2D genes (e.g., CYP2D2, ∼82%; CYP2D3, ∼85%; CYP2D4, ∼80%; CYP2D5, ∼98%; and CYP2D18, ∼80%), which makes it difficult to develop a CYP2D1 knockout model in rats. In this study, exon 4 was found to be a suitable target for the CRISPR/Cas9 technology. We analyzed the coding sequence of the CYP2D1 gene and found that the deletion of ACCTCAT would lead to truncated protein with only 213 amino acids, while the WT CYP2D1 had 511 amino acids.
Previous studies using CRISPR/Cas9 have reported that off-target effects may occur in the sgRNAs of mice and rats, which could also be inherited by the offspring (Tan et al., 2015; Anderson et al., 2018). We used TagScan to predict the potential off-target sites of sgRNA3 sequences in the CYP2D1-null rats and confirmed there were no mutations in these potential off-target sites. Whether the mutation of CRISPR/Cas9 occurred in nonpredicted sites was necessary for further study due to various factors.
In this study, the expression of CYP2C11 protein was detected only in male CYP2D1-null rats (the same as in WT rats), thus the knockout of CYP2D1 had little effect on the male-specific expression of CYP2C11. There was no significant difference in the expression levels of other cytochrome P450s, except that the expression levels of CYP2As were higher in the CYP2D1-null rats compared with the WT rats. The rat CYP2A family consisted of CYP2A1, 2A2, and 2A3 (Su and Ding, 2004). In the liver of rats, CYP2A1 and CYP2A2 were dominantly expressed in females and males, respectively (Haduch et al., 2005). CYP2A1 is responsible for the 7α-hydroxylation of testosterone, while CYP2A2 catalyzes the 15α- and 7α-hydroxylation of testosterone (Martignoni et al., 2006). The induction mechanism of CYP2A genes is still not very clear. To the best of our knowledge, very few endogenous substances have been identified to regulate the expression of CYP2A genes in rats, except that sexually dimorphic growth hormone has been reported to upregulate the expression of CYP2A1 and CYP2A2 (Waxman et al., 1995; Thangavel et al., 2004). Since the expression pattern of CYP2C11 was not changed in the CYP2D1-null rat model, it was unlikely that the induction of CYP2A was caused by growth hormone. Foreign compounds, including phenobarbital, dexamethasone, and 3-methylcholanthrene, were found to upregulate CYP2A1 (Ryan and Levin, 1990), and pregnane X, constitutive androstane, and aryl hydrocarbon receptors could also be involved in the regulation of CYP2As (Wallace and Redinbo, 2013). The specific roles of these nuclear receptors in CYP2D1-null rats need to be further studied.
In human beings, it has been reported that after administration of 150 mg VLF to 141 healthy subjects, the Cmax, AUC0–t, and AUC0–∞ of VLF in PMs increased by 83%, 189%, and 212%, respectively, compared with extensive metabolizers. The half-life was prolonged by 71%, and the CL/F was decreased by 68% (Kandasamy et al., 2010). In another study, 14 healthy volunteers received 75 mg VLF extended release formulations, and the mean Cmax and area under the curve of VLF were increased by 149% and 331% in the PMs compared extensive metabolizers, while the CL/F was decreased by 83% (Preskorn et al., 2009). However, in our VLF metabolism study with rat models, the Cmax, AUC0–t, and AUC0–∞ of VLF in the CYP2D1-null males were only increased by 24%, 64%, and 59%, respectively, compared with WT males, while the CL/F and apparent volume of distribution were decreased by 36% and 48%, respectively. The difference in VLF metabolism between CYP2D1-null and WT rats seemed much smaller than that between human PMs and extensive metabolizers. This fact suggested that other CYP2D isozymes might also contribute to VLF metabolism, and the broad use of the current knockout model was limited. Therefore, a rat model with multiple knockouts of CYP2D genes would be a better approach to understanding VLF metabolism in humans.
The CRISPR/Cas9 system could be used to knockout multiple genes simultaneously by coexpressing multiple sgRNAs targeting different sites (Cong et al., 2013; Cao et al., 2016; Hashimoto et al., 2016). However, these approaches achieved complete bi-allelic knockout animals with low efficiencies (Zuo et al., 2017). Since there were six members in the CYP2D subfamily, the deletion of these genes simultaneously in the rat genome would require an approach having very high gene-targeting efficiency. Zuo et al. (2017) demonstrated a new C-CRISPR approach, in which three genes (Tet1, Tet2, and Tet3) were completely knocked out in mouse embryos after a one-step injection of sgRNAs in zygotes. Most recently, Campa et al. (2019) demonstrated a multiplexed genome targeting method using Cas12a and CRISPR arrays delivered on a single plasmid. These high efficiency methods would greatly facilitate our future generation of a CYP2D-null rat model with deletion of all six CYP2D genes.
Depression is a serious mental disorder characterized by anhedonia, weight loss, sleeping disturbances, a sense of worthlessness, cognitive decline, or repeated thoughts of death (Xu et al., 2002). In the CMS rat model, 1% sucrose water consumption reflected the euphoria of the rats, and the locomotor activity in open-box was used to detect the state of nervousness in the new environment (Matthews et al., 1995; Willner, 1997). Thus, body weight, 1% sucrose water consumption, and locomotor activity tests are commonly used behavioral indicators in depression research (Bortolato et al., 2007; Dai et al., 2010). After 20 days of depression modeling, the body weight, 1% sucrose water consumption, and locomotor activity of the CMS groups were changed similarly to the results obtained in previous reports (Pucilowski et al., 1993; Orsetti et al., 2008; Zhao et al., 2008; Dai et al., 2010). These results demonstrated the reproducibility in our depression modeling. Notably, sporadic death occurred in both WT and CYP2D1-null female rats during the CMS modeling (data not shown), whereas the rat mortality rate in our experiment was less than that of mice under the same challenge (Goshen et al., 2008; Sun et al., 2011), indicating that rats are more tolerant than mice in depression modeling.
For the depressed WT rats, after 14 days of VLF treatment, all three depression indicators were significantly improved compared with the saline-treated rats. This result suggested that VLF was effective in our depression model. Based on the plasma levels of NE and 5-hydroxytryptamine, it was clear that VLF treatment was more effective in WT depression rats than in CYP2D1-null rats, possibly because of the reduced formation of ODV in CYP2D1-null rats. We also tried to determine ODV concentrations in the brain, but were unable to obtain convincing results due to limitations of the analytical instruments.
In conclusion, we have successfully developed a novel and valuable CYP2D1-null rat model in studying human drug metabolism by CYP2D6 using the CRISPR/Cas9 technology and have demonstrated that the antidepressant effects of VLF were decreased in depressed CYP2D1-null rats. Moreover, our data demonstrated that the reduced ODV formation was associated with lower VLF efficacy in rats, suggesting a potential application of monitoring plasma ODV levels in the clinical optimization of VLF for the treatment of depression.
Authorship Contributions
Participated in research design: Wei, Gu, Zhou.
Conducted experiments: Zhou, Yang, Wang, Li.
Performed data analysis: Zhou, Shan.
Wrote or contributed to the writing of the manuscript: Zhou, Wei, Gu, Ouyang.
Footnotes
- Received June 28, 2019.
- Accepted October 9, 2019.
This study was supported by the National Natural Science Foundation of China [Grants 81102522, 81373480, and 81573529], and the Natural Science Foundation of Jiangsu Province [Grant BK2011473].
↵This article has supplemental material available at dmd.aspetjournals.org.
Abbreviations
- AUC0–t
- area under the curve from the time of administration to the last time point
- AUC0–∞
- area under the curve from the start of administration to the theoretical extrapolation
- bp
- base pair
- CL/F
- apparent total clearance
- CMS
- chronic mild stimulation
- NE
- norepinephrine
- ODV
- O-desmethyl venlafaxine
- PCR
- polymerase chain reaction
- PM
- poor metabolizer
- sgRNA
- single guide RNA
- VLF
- venlafaxine
- WT
- wild type
- Copyright © 2019 by The American Society for Pharmacology and Experimental Therapeutics