Developmental Changes of Aldehyde Oxidase Activity and Protein Expression in Human Liver Cytosol

Yoshitaka TAYAMA1,*,KazumiSUGIHARA1,2,SeigoSANOH2, Katsushi MIYAKE1, Shigeyuki KITAMURA2,3 and Shigeru OHTA2 1Faculty of Pharmaceutical Science, Hiroshima International University, Kure, Japan 2Division of Medicinal Chemistry, Graduate School of Biomedical Sciences, Hiroshima University, Hiroshima, Japan 3Nihon Pharmaceutical University, Saitama, Japan

Full text of this paper is available at http://www.jstage.jst.go.jp/browse/dmpk

Summary: Aldehyde oxidase (AO) plays a role in metabolizing many drugs, such as methotrexate and 6- mercaptopurine. We previously showed that AO activity in rat liver rapidly increases from birth, reaching a plateau within 4 weeks, and is regulated at the protein expression level. However, developmental changes of AO activity and protein expression in human liver have not been reported. Here, we investigated the developmental changes and variability of AO in 16 human livers (13 children ranging from 13 days to 12 years old and 3 adults, 17, 34 and 45 years old). Young children (13 days to 4 months after birth) showed little liver AO activity, evaluated in terms of the activities for oxidation of N-1-methylnicotinamide to N-1- methyl-2-pyridone-5-carboxamide and N-1-methyl-4-pyridone-3-carboxamide in liver cytosol. However, these oxidase activities were markedly increased after 4 months, reaching the adult level by about 2 years of age. The AO band density in immunoblotting analysis waswellcorrelatedwiththeAOactivityamongall subjects (p < 0.01, r2 = 0.771). Therefore, AO activity in the liver of young children is regulated at the AO protein expression level. Thus, as in rats, the AO activity in humans rapidly increases soon after birth, and is regulated at the protein expression level.

Keywords: aldehyde oxidase; developmental change; children; human; liver

P450, flavin-containing monooxygenase¥ activities.16®19¥ In Introduction addition, developmental changes of several phase II drug- Aldehyde oxidase ¤AO, EC 1.2.3.1¥ mediates the oxi- metabolizing enzymes ¤glutathione S-transferase, sulfotrans- dation of aldehydes and nitrogenous heterocyclic xeno- ferase and glucuronyltransferase¥ in young children have biotics.1®5¥ It is involved in the metabolism of many drugs, been reported.20,21¥ This information is important, because including antimalarials ¤quinine¥,6¥ anticancer drugs ¤metho- of the possible influence on drug efficacy and side effects; trexate, 6-mercaptopurine and cyclophosphamide¥,7,8¥ anti- for example, in neonates who have immature glucuronyl- viral drugs ªfamciclovir and zidovudine ¤AZT¥«,9®11¥ and transferase activity, impaired metabolism of chloramphenicol antiepilepsy drugs ¤zonisamide¥.12¥ Therefore, AO activity is can result in so-called grey baby syndrome.15¥ important in clinical treatment, and developmental changes As regards AO, ontogenic changes of its activity in rat of AO in humans should be taken into account in pediatric liver have been reported.22¥ However, there is a marked medicine. species difference of AO activity between rats and humans. Developmental changes of several drug-metabolizing Therefore, in the present study, we examined the develop- enzymes have been examined in mammals,13,14¥ including mental changes of AO activity and their relationship to AO humans.15¥ For example, human liver shows ontogenic protein expression in 16 human livers covering the age range changes of phase I drug-metabolizing enzyme ¤cytochrome of 13 days to 45 years.

Received October 14, 2011; Accepted March 1, 2012 J-STAGE Advance Published Date: March 27, 2012, doi:10.2133/dmpk.DMPK-11-NT-124 *To whom correspondence should be addressed: Yoshitaka TAYAMA, Ph.D., Faculty of Pharmaceutical Science, Hiroshima International University, 5-1-1 Hirokoshingai, Kure 737-0112, Japan. Tel. +81-823-73-8576, Fax. +81-823-73-8981, E-mail: [email protected] This work was supported by a Grant-in-Aid for Scientiﬁc Research (no. 18790132) from the Ministry of Education, Culture, Sports, Science and Technology of Japan.

543 544 Yoshitaka TAYAMA, et al.

means + standard deviation ¤S.D.¥. The statistical signifi- Materials and Methods cance of differences was evaluated by use of the F-test fol- Chemicals and materials: N-1-Methylnicotinamide lowed by Welchös t-test. The correlation between AO activ- ¤NMN¥ and N$-methylnicotinamide were obtained from ity and protein band density was evaluated using Pearsonös Tokyo Chemical Industry Co., Ltd. ¤Tokyo, Japan¥. N-1- correlation. A value of p g 0.05 was considered significant. Methyl-2-pyridone-5-carboxamide ¤2-PY¥ and N-1-methyl- Results 4-pyridone-3-carboxamide ¤4-PY¥ were prepared according to Pullman and Colowick,23¥ and Shibata et al., respec- Developmental changes of liver aldehyde oxidase tively.24¥ Polyvinylidene fluoride ¤PVDF¥ membranes were in young children and adults: The activity for purchased from Millipore Co., Ltd. ¤Billerica, USA¥,10® conversion of NMN to 2-PY and 4-PY was assayed as an 20% polyacrylamide gel for electrophoresis was from Atto indicator of AO activity. As shown in Figure 1, the AO Co., Ltd. ¤Tokyo, Japan¥ and anti-rat AO antibodies were activities of H0354 ¤13 days old¥ and H0282 ¤2 months old¥ from Trans Genic Inc. ¤Kumamoto, Japan¥. Horseradish were low ¤0.44 and 0.49 nmol/30 min/mg protein, respec- peroxidase-conjugated goat anti-rabbit IgG was purchased tively¥. On the other hand, the AO activities of H0057 from Nacalai Tesque Co., Ltd. ¤Kyoto, Japan¥. ¤2 years old¥ were 2.13 nmol/30 min/mg protein, being Human subjects: The S9 fractions of 16 human livers virtually identical to the level in adults. Among samples were purchased from BD-Gentest Co. ¤New Jersey, USA¥. from humans over 2 years old, H0054 ¤12 years old¥ showed The cytosolic fraction was obtained from the S9 fraction by the lowest activity ¤2.31 nmol/30 min/mg protein¥, while centrifugation at 105,000 ' g for 60 min. Protein concen- H0215 ¤6 years old¥ showed the highest activity ¤4.70 nmol/ tration was determined in duplicate by the method of Lowry 30 min/mg protein¥, indicating that inter-individual dif- et al. with bovine serum albumin as the standard protein.25¥ ferences exist. Assay of NMN oxidase activity: NMN oxidase Expression of aldehyde oxidase protein in liver activity was measured according to the method of Ohkubo of young children and adults: SDS-PAGE immuno- et al.26¥ The amounts of 4-PY and 2-PY formed were blot analysis with anti-rat AO antibodies indicated the pres- measured by HPLC. Briefly, the incubation mixture ence of AO ¤150 kDa¥ in all samples from subjects over consisted of 0.2 µmol of NMN and liver cytosol equivalent 4 months old. On the other hand, AO was not detected in to 50®100 mg wet weight of liver in a final volume of 1 ml of H0354 ¤13 days old¥ or H0282 ¤2 months old¥ clearly 0.1 M K, Na-phosphate buffer ¤pH 7.4¥. The incubation was ¤Fig. 1¥. The AO band density was well correlated with AO performed at 37ôC for 30 min. After addition of 10 µg of activity in all subjects ¤p g 0.01, r2 © 0.771¥ and in subjects N$-methylnicotinamide ¤an internal standard¥, the mixture under 2 years old ¤p g 0.05, r2 © 0.631¥¤Fig. 2¥. was extracted with five volumes of ethyl acetate and the Developmental change of aldehyde oxidase extract was evaporated to dryness. The residue was dis- characteristics: We also examined the developmental solved in 0.1 ml of methanol, and an aliquot was subjected change of AO characteristics in terms of the 2-PY/4-PY to analysis by HPLC. HPLC was performed with a Capcell metabolite ratio. The 2-PY/4-PY ratio in the samples from pak C18 UG120 column ¤25 cm ' 4.6 mm, Shiseido Co. 8 subjects under 2 years of age was 8.11 + 0.78, while that Ltd., Tokyo, Japan¥ for the separation of 2-PY and 4-PY. in 6 subjects over 2 years of age was 9.00 + 2.33, and this The mobile phase was acetonitrile-water ¤3:97, v/v¥. The difference was not significant ¤p © 0.33¥. chromatograph was operated at a flow rate of 0.8 ml/min Discussion and with a detection wavelength of 254 nm. The elution times of 4-PY, 2-PY and N$-methylnicotinamide ¤an internal AO plays an important role in metabolizing many drugs standard¥ were 8.0, 9.0 and 19.2 min, respectively. used to treat pediatric patients, so developmental changes Immunoblot analysis of aldehyde oxidase: The of AO activity could be of clinical importance, and could expression levels of AO protein were determined by potentially be a useful parameter to guide dose adjustment in immunoblot analysis of liver cytosol protein. Human cyto- young patients in order to avoid toxicity. solic proteins ¤50 µg¥ were separated by SDS-PAGE on In the present study, we evaluated the ontogenic changes 10®20% polyacrylamide gel ¤Atto Co., Ltd.¥, and trans- of AO in the human liver. Neonates and infants ¤13 days to 4 ferred to PVDF membranes by electroblotting. Membranes months old¥ showed little liver AO activity. However, AO were then blocked with 2% skimmed milk in 25 mM activity subsequently increased, reaching a plateau by 2 years Tris-buffered saline ¤pH 7.6¥®0.1% Tween 20 for 1 h of age ¤Fig. 1¥. Although a spike of AO activity appeared and probed with anti-rat AO antibodies ¤1:1,000¥ for 24 h. to exist at around 6®8 years of age in our samples, we The membranes were washed, and antibody binding was previously observed no spike of in vivo AO activity in this detected with horseradish peroxidase-conjugated goat anti- age range, and large inter-individual differences ¤up to 3.9- rabbit IgG ¤1:1,000¥, followed by development with Chemi- fold¥ were seen in samples from subjects over 2 years.27¥ Lumi One L ¤Nacalai Tesque, Inc., Kyoto, Japan¥. Therefore, the apparent spike at around 6®8 years of age Statistical analysis: Data are presented as the found here is likely to be due to inter-individual difference.

Fig. 1. Developmental changes of AO activity and AO protein expression in human liver cytosol

Fig. 2. Relationship between AO band density in immunoblotting and AO activity (A) All examined human liver cytosol samples, (B) samples of liver cytosol from humans under 2 years of age.

It was reported that the ratio of 2-PY and 4-PY exhibited the other enzymes. However, we have reported that NMN significant differences among species ¤rat, mouse and hu- oxidase activities are inhibited neither by oxypurinol nor by man¥; the metabolic characteristics of AO exhibit significant allopurinol, inhibitors of xanthine oxidase. These activities species differences.28,29¥ So we also examined whether or not were not enhanced by addition of NAD¦/NADP¦, the developmental changes occur in the characteristics of AO, cofactors of aldehyde dehydrogenase. Therefore, oxidase using the the ratio of 2-PY to 4-PY formed from NMN activity of NMN is due mainly to AO, and the ratio of 2-PY as a parameter. The AO-catalyzed oxidation of NMN and 4-PY shows only AO characterization.30,31¥ generates equal amounts of 2-PY and 4-PY in rats, whereas We also evaluated the developmental change of AO the formation of 2-PY is greater than that of 4-PY in protein expression by means of SDS-PAGE immunoblot humans.28,29¥ Here, we found that the 2-PY/4-PY ratio analysis. AO protein was detected in all samples from showed no significant difference between the group under 2 subjects over 4 months of age and the AO band density was years of age and the group over 2 years of age. Thus, it well correlated with AO activity ¤p g 0.01, r2 © 0.771¥. appears that the characteristics of AO activity do not show These results suggest that the developmental change of AO any age-related change in humans. In general, oxidation of activity is regulated at the protein expression level. aromatic aldehydes may be catalyzed by aldehyde dehydro- We had previously used the same method to examine genase and possibly also by xanthine oxidase, in addition to developmental changes of in vivo AO activity in 101 chil- AO. Thus the 2-PY/4-PY ratio by AO may be affected by dren.27¥ However, young children ¤about under 1 years old¥

Copyright © 2012 by the Japanese Society for the Study of Xenobiotics (JSSX) 546 Yoshitaka TAYAMA, et al. are known to show immature renal functions, such as AO activity in 16 human livers. However, a further study excretion, metabolism, and transport, and so the possibility with a larger number of subjects may be worthwhile. remained that their renal excretion of pyridones might not The pattern of developmental change of AO activity in have reflected the in vivo AO activity. Therefore, since the human livers examined here was similar to that in our AO is expressed mainly in the liver,2¥ it was necessary to previous study of rats, as was the finding that the activity examine the developmental changes of liver AO in order to appears to be regulated at the protein expression level.22¥ confirm the applicability of our previously reported method It is interesting that human AO activity rapidly increased to young children. The patterns of the developmental before about 2 years of age, while rat AO activity rapidly changes of liver AO activity and AO protein expression increased from 2®3 weeks of age. These ages seem to corre- found here are consistent with those previously estimated by spond to the weaning periods in the two species, so it seems using our method for evaluating in vivo AO activity from possible that intake of solid foods may trigger an increase measurements of the endogenous substrate NMN and its of AO activity. metabolites. In other words, our findings here demonstrate In conclusion, aldehyde oxidase activity shows a devel- that it is possible to estimate the in vivo AO activity in young opmental increase in young children during growth, and is children, as well as adults, by using our previously reported well correlated with AO protein expression. Since AO plays method. Thus, we have validated our method of in vivo AO an important role in metabolizing many drugs that may be activity estimation as a tool to design appropriate indi- administered to pediatric patients, an understanding of its vidualized dosing regimens for AO substrate drugs in young developmental changes is important for the clinical appli- children. cation of AO-susceptible drugs in infants. The present So far, little is known about ethnic difference in AO results also serve to validate our previously reported method activity. On the other hand, inter-individual variations of of in vivo AO activity estimation as a tool to design appro- AO activities have been reported.32,33¥ Indeed, when the priate individualized dosing regimens for AO substrate drugs activity was assayed with benzaldehyde as a substrate, the in young children. difference between the highest and lowest values was 50.0- fold.33¥ Furthermore, we have reported inter-individual References differences of in vivo AO activity in Japanese young children ¥ and adults.27¥ Thus, although the influence of ethnic differ- 1 Beedham, C.: Molybdenum hydroxylases as drug metabolizing enzyme. Drug Metab. Rev., 16: 119®156 ¤1985¥. ence on developmental changes of AO activity remains to be 2¥ Berger, R., Mezey, E., Clancy, K. P., Harta, G., Wright, R. M., examined, inter-individual differences may be predominant. Repine, J. E., Brown, R. H., Brownstein, M. and Patterson, D.: A sex difference of AO activity has been found in mice: Analysis of aldehyde oxidase and xanthine dehydrogenase/oxidase males exhibited higher AO activity than females.1,34¥ On the as possible candidate genes for autosomal recessive familial fi amyotrophic lateral sclerosis. Somat. Cell Mol. Genet., 21: 121® other hand, Al-Salmy found no signi cant sex difference of ¤ ¥ 34¥ 131 1995 . AO activity in human liver cytosol. In addition, we also ¥ fi 3 Moriwaki, Y., Yamamoto, T. and Higashino, K.: Distribution and observed no signi cant gender difference of AO activity in Histol. ¥ pathophysiologic role of molybdenum-containing enzymes. humans at various growth stages.27 Further, in the present Histopathol., 12: 513®524 ¤1997¥. study, there was no gender difference among subjects over 4¥ Wright, R. M., Vaitaitis, G. M., Weigel, L. K., Repine, T. B., fi 2 years old ¤p © 0.780¥. These results suggest that gender McManaman, J. L. and Repine, J. E.: Identi cation of the candidate ALS2 gene at chromosome 2q33 as a human aldehyde oxidase gene. has little effect on the developmental change of human AO Redox Rep., 1: 313®321 ¤1995¥. activity. The number of subjects examined in this study 5¥ Krenitsky, T. A., Tuttle, J. V., Cattau, E. L. and Wang, P.: A was small. However, the in vivo AO activity can be still comparison of the distribution and electron acceptor specificities estimated noninvasively by measuring the ratio of pyridone of xanthine oxidase and aldehyde oxidase. Comp. Biochem. Physiol. B, ® ¤ ¥ formation ¤2-PY and 4-PY¥ from NMN in the urine. By 49: 687 703 1974 . 6¥ Beedham, C., al-Tayib, Y. and Smith, J. A.: Role of guinea pig and using that method, we evaluated the developmental changes rabbit hepatic aldehyde oxidase in oxidative in vitro metabolism of 27¥ of AO activity in 101 children. The pattern of the develop- cinchona antimalarials. Drug Metab. Dispos., 20: 889®895 ¤1992¥. mental changes of liver AO activity is consistent with that 7¥ Moriyasu, A., Sugihara, K., Nakatani, K., Ohta, S. and Kitamura, previously estimated for in vivo AO activity by using our S.: In vivo-in vitro relationship of methotrexate 7-hydroxylation by method with the endogenous substrate ¤NMN and its aldehyde oxidase in four different strain rats. Drug Metab. ¥ Pharmacokinet., 21: 485®491 ¤2006¥. metabolites . 8¥ Jordan, C. G., Rashidi, M. R., Laljee, H., Clarke, S. E., Brown, Furthermore, the ratio of the AO poor metabolizer to the J. E. and Beedham, C.: Aldehyde oxidase-catalyzed oxidation of normal metabolizer is low. For example, in our previous methotrexate in the liver of guinea-pig, rabbit and man. J. Pharm. study, 1 subject showed low AO activity ¤out of the Pharmacol., 51: 411®418 ¤1999¥. ¥ mean + 2 S.D.¥ in examined 26 adults subjects.27¥ So there 9 Clarke, S. E., Harrell, A. W. and Chenery, R. J.: Role of aldehyde oxidase in the in vitro conversion of famciclovir to penciclovir in is little possibility that the number of subjects had an effect human liver. Drug Metab. Dispos., 23: 251®254 ¤1995¥. on the developmental changes of AO activity on this study. 10¥ Rashidi, M. R., Smith, J. A., Clarke, S. E. and Beedham, C.: Thus, it is enough to discuss the developmental changes of In vitro oxidation of famciclovir and 6-deoxypenciclovir by

aldehyde oxidase from human, guinea pig, rabbit, and rat liver. 23¥ Pullman, M. E. and Colowick, S. P.: Preparation of 2- and 6- Drug Metab. Dispos., 25: 805®813 ¤1997¥. pyridones of N1-methylnicotinamide. J. Biol. Chem., 206: 121®127 11¥ Inaba, T., Fayz, S. and Stewart, D. J.: Aldehyde oxidase inhibitors ¤1954¥. for treatment of AIDS. United States Patent, 6: 400®434 ¤2000¥. 24¥ Shibata, K., Kawada, T. and Iwai, K.: Stimultaneous micro-deter- 12¥ Sugihara, K., Kitamura, S. and Tatsumi, K.: Involvement of mination of nicotinamide and its major metabolites, N1-methyl- mammalian liver cytosols and aldehyde oxidase in reductive 2-pyridone-5-carboxamide and N1-methyl-4-pyridone-3-carbox- metabolism of zonisamide. Drug Metab. Dispos., 24: 199®202 amide, by high-performance liquid chromatography. J. Chromatogr., ¤1996¥. A, 424:23®28 ¤1988¥. 13¥ Lupp, A., Danz, M. and Muller, D.: Histomorphological changes 25¥ Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J.: and cytochrome P450 isoforms expression and activities in Protein measurement with the Folin phenol reagent. J. Biol. Chem., precision-cut liver slices from neonatal rats. Toxicology, 206: 427® 193: 265®275 ¤1951¥. 438 ¤2005¥. 26¥ Ohkubo, M., Sakiyama, S. and Fujimura, S.: Purification and 14¥ Yoon, M., Madden, M. C. and Barton, H. A.: Developmental characterization of N1-methylnicotinamide oxidase I and II sepa- expression of aldehyde dehydrogenase in rat: a comparison of liver rated from rat liver. Arch. Biochem. Biophys., 221: 534®542 ¤1983¥. and lung development. Toxicol. Sci., 89: 386®398 ¤2006¥. 27¥ Tayama, Y., Miyake, K., Sugihara, K., Kitamura, S., Kobayashi, 15¥ Johnson, T. N.: The development of drug metabolising enzymes M., Morita, S., Ohta, S. and Kihira, K.: Developmental changes and their influence on the susceptibility to adverse drug reactions in of aldehyde oxidase activity in young Japanese children. Clin. children. Toxicology, 192:37®48 ¤2003¥. Pharmacol. Ther., 81: 567®572 ¤2007¥. 16¥ Leeder, J. S., Gaedigk, R., Marcucci, K. A., Gaedigk, A., 28¥ Kitamura, S., Sugihara, K. and Ohta, S.: Drug-metabolizing ability Vyhlidal, C. A., Schindel, B. P. and Pearce, R. E.: Variability of of molybdenum hydroxylases. Drug Metab. Pharmacokinet., 21:83® CYP3A7 expression in human fetal liver. J. Pharmacol. Exp. Ther., 98 ¤2006¥. 314: 626®635 ¤2005¥. 29¥ Sugihara, K., Tayama, Y., Shimomiya, K., Yoshimoto, D., Ohta, 17¥ Cotreau, M. M., von Moltke, L. L. and Greenblatt, D. J.: The S. and Kitamura, S.: Estimation of aldehyde oxidase activity in vivo influence of age and sex on the clearance of cytochrome P450 3A from conversion ratio of N1-methylnicotinamide to pyridones, and Substrates. Clin. Pharmacokinet., 44:33®60 ¤2005¥. intraspecies variation of the enzyme activity in rats. Drug Metab. 18¥ Lacroix, D., Sonnier, M., Moncion, A., Cheron, G. and Cresteil, Dispos., 34: 208®212 ¤2006¥. T.: Expression of CYP3A in the human liver¯evidence that the 30¥ Ueda, O., Sugihara, K., Ohta, S. and Kitamura, S.: Involvement shift between CYP3A7 and CYP3A4 occurs immediately after of molybdenum hydroxylases in reductive metabolism of nitro birth. Eur. J. Biochem., 247: 625®634 ¤1997¥. polycyclic aromatic hydrocarbons in mammalian skin. Drug Metab. 19¥ Koukouritaki, S. B., Simpson, P., Yeung, C. K., Rettie, A. E. and Dispos., 33: 1312®1318 ¤2005¥. Hines, R. N.: Human hepatic flavin-containing monooxygenases 1 31¥ Tayama, Y., Sugihara, K., Sanoh, S., Miyake, K., Morita, S., ¤FMO1¥ and 3 ¤FMO3¥ developmental expression. Pediatr. Res., 51: Kitamura, S. and Ohta, S.: Effect of tea beverages on aldehyde 236®243 ¤2002¥. oxidase activity. Drug Metab. Pharmacokinet., 26:94®101 ¤2011¥. 20¥ McCarver, D. G. and Hines, R. N.: The ontogeny of human drug- 32¥ Rodrigues, A. D.: Comparison of levels of aldehyde oxidase with metabolizing enzymes: phase II conjugation enzymes and regulatory cytochrome P450 activities in human liver in vitro. Biochem. mechanisms. J. Pharmacol. Exp. Ther., 300: 361®366 ¤2002¥. Pharmacol., 48: 197®200 ¤1994¥. 21¥ Kawade, N. and Onishi, S.: The prenatal and postnatal develop- 33¥ Sugihara, K., Kitamura, S., Tatsumi, K., Asahara, T. and Dohi, K.: ment of UDP-glucuronyltransferase activity towards bilirubin and Differences in aldehyde oxidase activity in cytosolic preparations of the effect of premature birth on this activity in human liver. human and monkey liver. Biochem. Mol. Biol. Int., 41: 1153®1160 Biochem. J., 196: 257®260 ¤1981¥. ¤1997¥. 22¥ Tayama, Y., Moriyasu, A., Sugihara, K., Ohta, S. and Kitamura, 34¥ Al-Salmy, H. S.: Individual variation in hepatic aldehyde oxidase S.: Developmental changes of aldehyde oxidase in postnatal rat activity. IUBMB Life, 51: 249®253 ¤2001¥. liver. Drug Metab. Pharmacokinet., 22: 119®124 ¤2007¥.