Regional distributions of UDP-glucuronosyltransferase activities toward estradiol and serotonin in the liver and small intestine of cynomolgus macaques

The cynomolgus macaque is a nonhuman primate species that is often used in drug metabolism studies during drug development. However, the localization of UDP-glucuronosyltransferases (UGTs), essential drug-metabolizing enzymes, has not been fully investigated in the liver and small intestine of cynomolgus macaques. In this study, UGT activities were analyzed in liver (five lobes) and small intestine (the duodenum and six sections from the proximal jejunum to the distal ileum) using typical probe substrates of human UGTs: 7-hydroxycoumarin, estradiol, serotonin, propofol, and zidovudine. In liver, UGT activities with respect to all substrates were detected, and the activity levels were similar in all liver lobes of the cynomolgus macaques tested. In contrast, in the small intestine, UGT activities toward all substrates were detected, but their levels generally decreased from jejunum to ileum in cynomolgus macaques. The localization of estradiol 3-O-glucuronosyltransferases and serotonin O-glucuronosyltransferases (which are mainly UGT1A enzymes) appear to be different in liver and small intestine. These results collectively suggest that, in cynomolgus macaques, UGT1As are differentially localized in the small intestine but are relatively homogeneously distributed in the liver.     

Trimethylamine N-oxygenation in cynomolgus macaques genotyped for flavin-containing monooxygenase 3 (FMO3)

Polymorphic human and cynomolgus macaque flavin-containing monooxygenases (FMO) 3 are important oxygenation enzymes for nitrogen-containing drugs. Inter-animal variability of FMO3-dependent drug oxygenations in vivo is suspected in cynomolgus macaques because such variability is evident in humans. Therefore, this follow-up study was performed to investigate the pharmacokinetics of orally administered deuterium-labeled trimethylamine in three cynomolgus macaques genotyped for FMO3. Trimethylamine-d₉ was rapidly absorbed and attained plasma concentrations greater than the background levels of non-labeled trimethylamine. Trimethylamine-d₉ was then converted to trimethylamine-d₉ N-oxide. The half-lives, maximum plasma concentrations, and areas under the curve for trimethylamine-d₉ and its N-oxygenated metabolite and the total clearance for orally administered trimethylamine-d₉ were not different among the heterozygote for Q506K FMO3, the heterozygote for V325I FMO3, and the heterozygote for both S99N and F510S FMO3. Trimethylamine N-oxygenation activities mediated by liver microsomes prepared from the same three animals were not substantially different. However, recombinant proteins of the corresponding cynomolgus FMO3 variants showed apparent reduced trimethylamine N-oxygenation activities compared with the wild-type proteins. This study suggests only limited polymorphic effects on the in vivo catalytic function of cynomolgus FMO3. These findings yield important insights in terms of both quantitative and qualitative variations of polymorphic FMO3 in cynomolgus liver.     

Non-synonymous genetic variants of flavin-containing monooxygenase 3 (FMO3) in cynomolgus macaques

Polymorphic human flavin-containing monooxygenase (FMO) 3 is an important drug-metabolizing enzyme for nitrogen- or sulfur-containing compounds. Cynomolgus macaques, a non-human primate species widely used in drug metabolism studies, have corresponding FMO3 molecular and enzymatic similarities to humans; however, genetic polymorphisms have not been investigated in macaques. In this study, re-sequencing of FMO3 in 64 cynomolgus and 32 rhesus macaques found a total of 18 non-synonymous variants. Nine variants were unique to cynomolgus macaques, of which 4 (including Q506K) were found only in Indochinese, 4 (including V299I, E348H, and G530A) only in Indonesian lineages, and one was common. Other five variants (including S504T at >10% allele frequencies) were unique to rhesus macaques. By functional characterization using cynomolgus FMO3 proteins heterologously expressed in Escherichia coli, FMO3 R509H variant appeared to suppress methimazole and benzydamine S- or N-oxygenations. Seven variants showed substantially lower benzydamine N-oxygenation as compared with wild-type FMO3 protein. Further analysis indicated that two of these variants, FMO3 G530A and R417H, showed significantly lower benzydamine N-oxygenation in liver microsomes of the homozygotes as compared with wild-type animals. Therefore, inter-animal variability of FMO3-dependent drug metabolism is at least partly accounted for by genetic polymorphisms in cynomolgus and rhesus macaques, similar to humans.     

Functionally relevant genetic variants of glutathione S-transferase GSTM5 in cynomolgus and rhesus macaques

1. Glutathione S-transferase (GST) is a family of enzymes important for conjugation with glutathione of endogenous and exogenous compounds. Human GSTM1 null allele is associated with toxicity and cancers. Cynomolgus and rhesus macaques have molecular and enzymatic similarities of GSTs to humans; however, genetic variants have not been investigated. In macaques, instead of pseudogenized GSTM1, GSTM5 is a predominant GSTM isoform.

2. In this study, re-sequencing of GSTM5 in 64 cynomolgus and 31 rhesus macaques found 6 non-synonymous variants, and 1 variant (IVS5?+?1) causing exon skip. Of these 7 variants, 3 and 1 were found only in Indochinese and Indonesian cynomolgus macaques, respectively.

3. Cynomolgus GSTM5-mediated styrene 7,8-oxide and trans-stilbene oxide conjugation activities correlated with GSTM protein levels immunochemically quantified in cynomolgus liver samples. Using recombinant GSTM5 proteins, 4 of the 6 non-synonymous variants including E29Q, L96R, M166V and S201N showed substantially lower metabolic activities. Moreover, a homozygote for E29Q and heterozygotes for S201N or IVS5?+?1 showed significantly lower conjugation activities in liver cytosolic fractions as compared with wild-type animals.

4. Therefore, the present results suggest that inter-animal variability of GST-dependent drug metabolism is at least partly accounted for by GSTM5 variants in cynomolgus and rhesus macaques as pre-clinical animal models.

     

mRNA levels of drug-metabolizing enzymes in 11 brain regions of cynomolgus macaques

The cynomolgus macaque is an important nonhuman primate species in drug metabolism studies, in part because of its evolutionary closeness to humans. Cytochromes P450 (P450s) have been investigated in the major drug-metabolizing organs, i.e., the liver and small intestine, but have not been fully investigated in the brain. However, recent investigations have indicated possible important roles for P450s in the brain. In this study, by using the quantitative polymerase chain reaction, we measured the mRNA levels of 38 cynomolgus drug-metabolizing enzymes, including 19 P450s, 10 UDP-glycosyltransferases, and 9 other enzymes, in 11 brain regions. Among these drug-metabolizing enzymes, expression of 32 enzyme mRNAs were detected in one or more brain regions, indicating their possible roles in the brain. Further investigation of metabolic activities would facilitate better understanding of the importance of these enzymes in the brain.     

Quantitative regioselectivity of glucuronidation of quercetin by recombinant UDP-glucuronosyltransferases 1A9 and 1A3 using enzymatic kinetic parameters

Quercetin has been suggested to exert its pharmacological effects, at least in part, via its metabolites, such as glucuronides. Quantitative regioselectivity analyses are important to understand the contributions of UDP-glucuronosyltransferases (UGTs) to the pharmacological activity of quercetin. The present work obtained active UGT1A9 and UGT1A3 enzymes with a Bac-to-Bac expression system, and quercetin was metabolized by each of them to four monoglucuronides (7-, 3-, 4′- and 3′-glucuronide). Enzymatic kinetic parameters of each glucuronide were calculated to elucidate quantitatively UGT1A9's and UGT1A3's regioselectivities for quercetin. UGT1A3's highest glucuronidation efficiency was observed for the 3′-glucuronide, then the 3-, 4′- and 7-glucuronide. The catalytic efficiency order for UGT1A9 was 3-?>?7-?>?3′-?>?4′-glucuronide.     

CYP2C19 polymorphisms account for inter-individual variability of drug metabolism in cynomolgus macaques

CYP2C19 (formerly known as CYP2C75), highly homologous to human CYP2C19, has been identified in cynomolgus and rhesus macaques, non-human primate species widely used in drug metabolism studies. CYP2C19 is predominantly expressed in liver and encodes a functional drug-metabolizing enzyme. Genetic variants in human CYP2C genes account for the inter-individual variability in drug metabolism; however, genetic variants have not been investigated in macaque CYP2C19. In the present study, re-sequencing of CYP2C19 in 78 cynomolgus and 36 rhesus macaques identified 34 non-synonymous variants. Among these, 6 were located in substrate recognition sites, the domains important for protein function. Eighteen and 6 variants were unique to cynomolgus and rhesus macaques, respectively. Four variants were characterized by site-directed mutagenesis and metabolic assays, and 3 variants (p.Phe100Asn, p.Ala103Val, and p.Ile112Leu) showed substantially reduced activity as compared with wild type in flurbiprofen 4′-hydroxylation, omeprazole 5-hydroxylation, and R-/S-warfarin 7-hydroxylation. These variants, co-segregating in the animals analyzed, influenced metabolic activities because the homozygotes and/or heterozygotes showed significantly reduced catalytic activities in liver toward flurbiprofen 4′-hydroxylation and omeprazole 5-hydroxylation as compared with wild type. Kinetic analysis for R-warfarin 7-hydroxylation and docking simulation indicated that CYP2C19 Ala103Val would change the function and conformation of this enzyme. Ala103Val variation diminished homotropic cooperativity of CYP2C19 with R-warfarin yielding low metabolic capacity. These results indicated that the interindividual variability of CYP2C-dependent drug metabolism is at least partly accounted for by CYP2C19 variants in cynomolgus macaques.     

Comprehensive evaluation of liver microsomal cytochrome P450 3A (CYP3A) inhibition: comparison of cynomolgus monkey and human

1.?Members of the cytochrome P450 3A (CYP3A) subfamily metabolize numerous compounds and serve as the loci of drug–drug interactions (DDIs). Because of high amino acid sequence identity with human CYP3A, the cynomolgus monkey has been proposed as a model species to support DDI risk assessment.

2.?Therefore, the objective of this study was to evaluate 35 known inhibitors of human CYP3A using human (HLM) and cynomolgus monkey (CLM) liver microsomes. Midazolam was employed as substrate to generate IC₅₀ values (concentration of inhibitor rendering 50% inhibition) in the absence and presence of a preincubation (30 mins) with NADPH.

3.?In the absence of preincubation, the IC₅₀ values generated with CLM were similar to those obtained with HLM (86% within 2-fold; 100% within 3-fold difference). However, significant differences (up to 48-fold) in preincubation IC₅₀ were observed with 17% of the compounds (raloxifene, bergamottin, nicardipine, mibefradil, ritonavir, and diltiazem).

4.?Our results indicate that in most cases the cynomolgus monkey can be a viable DDI model. However, significant species differences in time-dependent CYP3A inhibition can be observed for some compounds. In the case of raloxifene, such a difference can be ascribed to a specific CYP3A4 amino acid residue.     

A multiple-dose toxicity study of tanezumab in cynomolgus monkeys

Nerve growth factor (NGF) is an important mediator of pain and hyperalgesia and has become a target of novel analgesic therapeutics. Tanezumab is a humanized IgG(2) antibody that binds NGF with high affinity and specificity. In a study to assess the toxicity and pharmacokinetic properties of tanezumab in adult, male and female, cynomolgus monkeys following weekly intravenous administration of 1, 10, or 30 mg/kg for up to 26 weeks (followed by an 8-week recovery period), tanezumab was well tolerated with no macroscopic or microscopic effects on those brain, spinal cord, nerve, or ganglia sections evaluated. One fifth of tanezumab-treated monkeys developed an antibody response to tanezumab that prevented maintenance of tanezumab exposure between dosing. In the antibody-negative animals, accumulation of tanezumab was observed; steady state was achieved approximately 8 weeks after the first dose of study drug, and exposure to tanezumab was approximately dose proportional with no observed difference between male and female animals. One monkey died during the study; this monkey had findings suggestive of hypersensitivity reaction. The favorable toxicity and pharmacokinetic profile of tanezumab seen in this study supports its further evaluation for the treatment of pain in clinical practice.     

Characterization of CYP1A in hepatocytes of cynomolgus monkeys (Macaca fascicularis) and induction by different substituted polychlorinated biphenyls (PCBs)

Cynomolgus monkeys (Macaca fascicularis) have been used previously as a model to study effects on cytochrome P450 (CYP) regulation. Until now it has not been elucidated which CYP1A proteins are present in this primate species. The aim of this study was to characterize CYP1A in untreated hepatocytes of cynomolgus monkey using two specific CYP1A inhibitors (α -naphthoflavone and furafylline). The effect of different substituted polychlorinated biphenyls (PCBs) on CYP1A regulation was also studied in these hepatocytes. Small quantities of CYP1A2 have been identified in untreated hepatocytes. Northern blots showed the presence of a CYP1A mRNA in untreated hepatocytes, when hybridizations where performed with human CYP1A2 cDNA. Inhibitions with furafylline and α -naphthoflavone also suggested the presence of CYP1A2 properties. After induction with different PCBs, (probably) CYP1A1 mRNA and enzyme activity were induced in cynomolgus monkey hepatocytes. As expected, 2,3′,4,4′,5-PeCB (PCB no. 118), a mono-ortho substituted congener, was a potent CYP1A inducer but 2,2′,3,4,4′,5′,5′-HpCB (PCB no. 180), a di-ortho and 2,2′,3,4′,5,5′,6-HpCB (PCB no. 187), a tri-ortho substituted PCB, could induce CYP1A mRNA and enzyme activity in cynomolgus monkey hepatocytes as well. Received: 20 April 1998 / Accepted: 1 July 1998     

Use of mixed-effect models and tolerance limits to evaluate control cynomolgus monkey body weight change and variability during preclinical toxicology studies

Cynomolgus monkeys are an important and widely used species in preclinical toxicology studies. During the in-life phase of study, body weight effects may be indicative of toxicity; however, trends in body weight and body weight variability are often difficult to interpret due to small sample size and/or inter- and intra-animal variability. The present analysis utilizes mixed-effect modelling, which incorporates random and fixed effects into linear regression models, to evaluate control monkey body weight trends and variability relative to baseline (initial) weight and study duration. The primary aim of this analysis is to evaluate whether mixed-effect model based tolerance limits can aide in determining whether apparent test article-related changes in body weight deviate more than the predicted variability defined by the model tolerance limits. The models for this study are based on vehicle control animal body weight data from the following studies: 1-month (20 studies, 198 animals), 3-month (19 studies, 180 animals), and 9-month (17 studies, 182 animals). The analysis presented herein provides the framework for evaluating control monkey body weight change in studies with small sample size, and anticipated control monkey body weight change relative to gender and study duration.     

Glucuronidation of flavonoids by recombinant UGT1A3 and UGT1A9

Flavonoids are highlighted for their potential roles in the prevention of oxidative stress-associated diseases. Their metabolisms in vivo, such as glucuronidation, are the key points to determine their health beneficial properties. In this paper, we tested the glucuronidation of nineteen flavonoids by both recombinant human UGT1A3 and UGT1A9. Eleven compounds could be catalyzed by both enzymes. In general, both enzymes showed moderate to high catalyzing activity to most flavonoid aglycones, while the catalyzing efficiency changed with structures. Each flavonoid produced more than one monoglucuronide with no diglucuronide detected by liquid chromatography-mass spectrometry (LC-MS). Enzymatic kinetic analysis indicated that the catalyzing efficiency (Vmax/Km) of UGT1A9 was higher than that of UGT1A3, suggesting its important role in flavonoid glucuronidation. Both human UGT1A3 and UGT1A9 preferred flavonoid aglycone to flavonoid glycoside, and their metabolism to arabinoside was stronger than to other glycosides. Of the flavonoids studied, it is the first time to report isorhamnetin, morin, silybin, kaempferol, daidzein, quercetin-3',4'-OCHO-, quercetin xylopyranoside and avicularin as substrates of UGT1A3. Apigenin, morin, daidzein, quercetin-3',4'-OCHO-, quercetin xylopyranoside and avicularin were the newly reported substrates of UGT1A9.     

Contribution of UDP-glucuronosyltransferases 1A9 and 2B7 to the glucuronidation of indomethacin in the human liver

Objective We characterized the kinetics of indomethacin glucuronidation by recombinant UDP-glucuronosyltransferase (UGT) isozymes and human liver microsomes (HLM) and identified the human UGT isozymes involved. Methods Indomethacin glucuronidation was investigated using HLM and recombinant human UGT isozymes. Human UGTs involved in indomethacin glucuronidation were assessed in kinetic studies, chemical inhibition studies, and correlation studies. Results Among the UGT isozymes investigated, UGT1A1, 1A3, 1A9, and 2B7 showed glucuronidation activity for indomethacin, with UGT1A9 possessing the highest activity, followed by UGT2B7. Glucuronidation of indomethacin by recombinant UGT1A9 and 2B7 showed substrate inhibition kinetics with K _m values of 35 and 32 μM, respectively. The glucuronidation of indomethacin was significantly correlated with morphine 3OH-glucuronidation (r = 0.69, p < 0.05) and 3′-azido-3′-deoxythymidine glucuronidation (r = 0.82, p < 0.05), a reaction mainly catalyzed by UGT2B7. Propofol inhibited indomethacin glucuronidation in HLM with an IC₅₀ value of 248 μM, which is between the IC₅₀ value in recombinant UGT1A9 (106 μM) and UGT2B7 (> 400 μM). Conclusions These findings suggest that UGT2B7 plays a predominant role in indomethacin glucuronidation in the human liver and that UGT1A9 is partially involved.     

人骨髓间充质干细胞脑内注射给予食蟹猴的长期毒性研究

目的研究人骨髓间充质干细胞(BMSC)脑内注射给予食蟹猴的长期毒性,为临床试验提供依据。方法24只食蟹猴随机分为对照组(生理盐水)、低剂量组(3×105个细胞/kg)和高剂量组(2.5×106个细胞/kg),每只动物脑内基底节外囊区域定向注射人BMSC2次,间隔3周。注射后观察动物的一般毒性反应,并进行体重、体温、神经功能、心电图、血液学和血液生化学、体液免疫(IgG、IgM)、细胞免疫(CD3、CD4、CD8)、尿液、脑脊液(一般性状、细胞计数、葡萄糖、蛋白质和氯化物含量)及组织病理学等指标测定。结果病理结果显示食蟹猴脑内基底节外囊区域注射人BMSC后,注射部位大脑局部出现组织坏死及炎症反应,对照组、低剂量组和高剂量组的损伤程度和范围呈递增趋势,随着时间推移,各组脑组织病变基本修复,残存少量瘢痕组织。其他各项检测指标均未见有生物学意义的显著改变。结论在本试验条件下,2.5×106/kg的剂量范围内,人BMSC食蟹猴脑内基底节外囊区域注射仅对注射局部大脑组织产生一定的病理损伤,损伤可随着时间推移逐渐修复。     

Influence of UGT1A8 and UGT2B7 genetic polymorphisms on mycophenolic acid pharmacokinetics in Japanese renal transplant recipients

Objective UGT1A8 and UGT2B7 are important uridine diphosphate-glucuronosyltransferase isoforms for the glucuronidation of mycophenolic acid (MPA). The aim of this investigation was to elucidate MPA pharmacokinetics in UGT1A8 and UGT2B7 genotypes in Japanese renal transplant recipients. Methods Seventy-two recipients received repeated doses of mycophenolate mofetil and tacrolimus. On day 28 after renal transplantation, plasma MPA concentrations were measured for the next 24 h using high-performance liquid chromatography. UGT1A8*2 (A¹⁷³G) and UGT2B7*2 (Y²⁶⁸) were detected using a PCR-RFLP-based procedure. Results There were no significant differences in daytime and nighttime pharmacokinetics of MPA between UGT1A8 or UGT2B7 genotypes. The mean daytime dose-adjusted AUC_0–12 of MPA in UGT1A8*1/*1, *1/*2 and *2/*2 were 2.47, 2.33 and 2.57 ng·h/ml/mg/kg (P = 0.7711), and the mean nighttime AUC_0–12 were 2.15, 2.00 and 2.08 ng·h/ml/mg/kg (P = 0.4656). The mean daytime and nighttime dose-adjusted AUC_0–12 of MPA in UGT2B7*1/*1, *1/*2 and *2/*2 were 2.61, 2.24 and 2.03 ng·h/ml/mg/kg and 2.18, 1.94, and 1.45 ng·h/ml/mg/kg, respectively (P = 0.3475 and 0.2575). The mean nighttime C_max, t_max, and AUC_6–12/AUC_0–12 ratio (enterohepatic circulation and recirculation ratio) of MPA in all UGT1A8 and UGT2B7 genotypes were lower, longer, and higher, respectively, than the daytime values. Conclusions Both UGT1A8 and UGT2B7 allelic variants seem not to affect Japanese interindividual variability for plasma MPA concentration. Regardless of UGT1A8 and UGT2B7 genetic polymorphisms, the absorption of MPA through enterohepatic recirculation is higher at night.     

Cloning and tissue expression of ATP-binding cassette transporters in cynomolgus macaques

Cynomolgus macaques are used in preclinical studies in part because of their evolutionary closeness to humans. However, drug transporters, including ATP-binding cassette (ABC) transporters, which are essential for the absorption and excretion of drugs, have not been fully investigated at the molecular level in cynomolgus macaques. In this study, ABCB4, ABCC3, ABCC4, and ABCG2 cDNAs were newly identified and characterized, along with ABCB1, ABCB11, and ABCC2 cDNAs previously identified, in cynomolgus macaques. All seven cynomolgus ABC transporters had high sequence identities (96–98%) with their human orthologs in terms of amino acid sequences and were also most closely clustered with their human orthologs by phylogenetic analysis. Furthermore, the gene structures and genomic organization were similar in cynomolgus macaques and humans. The mRNAs of these cynomolgus ABC transporters, as analyzed using the quantitative polymerase chain reaction, showed distinct tissue expression patterns. Among the ten tissues, ABCB1, ABCC2, ABCC3, and ABCG2 mRNAs were most abundantly expressed in jejunum; ABCB4 and ABCB11 in liver; and ABCC4 in kidney, which are similar to the expression patterns of human ABC transporters. These results suggest molecular similarities of the ABC transporters in cynomolgus macaques and humans.     

A comparative study for detecting CYP3A induction by CYP3A probe drugs and endogenous markers in cynomolgus monkeys

CYP3A probe drugs such as midazolam and endogenous markers, and plasma 4β‐hydroxycholesterol (4β‐OHC) and urinary 6β‐hydroxycortisol‐to‐cortisol ratios (6β‐OHC/C) have been used as markers of CYP3A induction in cynomolgus monkeys, as with humans. However, there is limited information on their sensitivity and ability to detect CYP3A induction, as most studies were evaluated only at a high dose of the inducer, rifampicin (RIF; 20 mg/kg). In the present study, the CYP3A induction by RIF over a range doses of 0.2, 2 and 20 mg/kg (n = 4) was examined using CYP3A probe drugs (midazolam, triazolam and alprazolam) and the plasma and urinary endogenous CYP3A markers (4β‐OHC and 6β‐OHC/C). The sensitivity and relationship for detecting CYP3A induction was compared among the markers. Four days repeated oral administration of rifampicin to cynomolgus monkeys reduced the area under the plasma concentration–time curve of all CYP3A probe drugs in a rifampicin dose‐dependent manner. Although the endogenous CYP3A markers (4β‐OHC and 6β‐OHC/C) were also changed for the middle (2 mg/kg) and high (20 mg/kg) doses of rifampicin, the fold‐changes were relatively small, and CYP3A induction could not be detected at the lowest dose of rifampicin (0.2 mg/kg). In conclusion, CYP3A probe drugs are more sensitive for detecting CYP3A induction than endogenous CYP3A markers in cynomolgus monkeys, even for a short experimental period.