- Email: [email protected]
Isolation of a Cytochrome P450 3A cDNA Sequence (CYP3A30) from the Marine TeleostFundulus heteroclitusand Phylogenetic Analyses ofCYP3AGenes
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.
236, 306–312 (1997)
RC976956
Isolation of a Cytochrome P450 3A cDNA Sequence (CYP3A30) from the Marine Teleost Fundulus heteroclitus and Phylogenetic Analyses of CYP3A Genes Malin Celander1 and John J. Stegeman2 Biology Department, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543
Received June 11, 1997
A reverse transcriptase polymerase chain reaction (RT-PCR) protocol, using degenerate PCR-primers specific to highly conserved regions of mammalian CYP3A genes, was employed to amplify a 400 base pair cDNA fragment from Fundulus heteroclitus liver RNA. The 124 amino acid sequence deduced from this cDNA sequence was aligned with corresponding sequences from representative members from the CYP1, 2, 3, and 4 gene families retrieved from the GenBank database. Phylogenetic trees were constructed using distancematrix and maximum parsimony methods. The F. heteroclitus sequence and all mammalian CYP3A sequences cluster together when compared to sequences of members of CYP gene families 1, 2, and 4. This fish sequence was 57 to 70 % identical to the corresponding region of mammalian CYP3A genes. These data indicate that the sequence obtained from F. heteroclitus represents a teleost fish CYP3A gene and it has been designated CYP3A30. q 1997 Academic Press
Members of the cytochrome P450 3A (CYP3A) subfamily are major constitutively expressed CYP forms in the liver and in the gastro-intestinal tract of mammals (1). CYP3A enzymes metabolize steroids and a wide variety of non-steroidal lipophilic organic molecules, many of which commonly are used as therapeutic drugs (2, 3). CYP3A enzymes also can activate several pro-carcinogens and are therefore believed to be involved in chemical carcinogenesis (4). Many substrates of CYP3A enzymes, including glucocorticoids and macrolide antibiotics, induce the expression of CYP3A genes (5, 6). Although the mechanisms for induction by these various inducers are not yet understood, some 1 Permanent address: Department of Zoophysiology, University of Go¨teborg, Medicinaregatan 18, S-413 90 Go¨teborg, Sweden. 2 To whom correspondence should be addressed. Fax: (508) 457 2169. E-mail: [email protected].
progress has been made in understanding drug metabolism and drug-drug interactions as a result of CYP3A activities (7, 8, 9). The CYP3A subfamily is one of the largest in the P450 superfamily and multiple CYP3A genes have been identified in several species (10). At present there is insufficient information on the properties of most CYP3A forms to identify orthologous relationships among the multiple genes within this subfamily. Most studies of structure, function and regulation of CYP3As have been in mammalian systems, whereas relatively little is known about CYP3A in other vertebrate groups. Catalytic and immunochemical analyses have pointed to the existence of a CYP3A-like protein in fish (11, 12). However, sequence data are required before these proteins can be assigned to a subfamily in the P450 superfamily. In this paper we present evidence for the presence of a CYP3A gene in a marine teleost fish species, Fundulus heteroclitus, and we also provide a phylogenetic analysis of CYP3A genes. Fish represent the earliest diverging vertebrates in evolution and it is therefore important to include fish gene sequences in phylogenetic analyses if we are to understand the evolution of vertebrate CYP3A genes. MATERIALS AND METHODS Animals and RNA isolation. Killifish, Fundulus heteroclitus, were captured by minnow traps in salt marshes on Cape Cod, MA, USA. The fish were killed by a cervical transection and the liver was dissected out and immediately placed in liquid nitrogen. Frozen livers from 8 untreated adult males were pulverized under liquid nitrogen and the total RNA fraction was directly isolated using the ‘‘Perfect RNA Total RNA Isolation Kit (Maxi Scale)’’ purchased from 5 Primer3 Prime Inc., Boulder, CO, USA. The quality of the preparation was investigated on an ethidium bromide-stained 1 % (w/v) agarose gel in 2.2 M formaldehyde, 1 1 MOPS/acetate-buffer pH 7.0 (20 mM MOPS/8 mM Na-acetate/1 mM EDTA) and the concentration and the purity of the RNA fraction was determined spectrophotometrically at A260 and A280 (13). Oligonucleotides. Oligonucleotides were designed based on conserved regions of 17 mammalian CYP3A deduced amino acid (a.a.)
306
0006-291X/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.
AID
BBRC 6956
/
6932$$$521
06-30-97 19:03:58
bbrcgs
AP: BBRC
Vol. 236, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
sequences as aligned by Dr. David Nelson (www.drnelson.utmem.edu/nelsonhomepage.html). The sequence of the N-terminal primer targeted to a.a. 123-128 of CYP3As is [5*-GAYGARGARTGGAARMG-3*] having a degeneracy of 32. The sequence of the C-terminal primer targeted to a.a. 434-444 in the heme binding site of CYP3As is [5*-CCDATRCARTTICKIGGICC-3*] having a degeneracy of 24. These oligonucleotides were synthesized by National Biosciences (Plymouth, MN, USA). Degenerate primers targeted to the N-terminal a.a. 304-311 of CYP3As [5*-TTTGCKGGNTATGARACMACNAGCAG-3*] and to the C-terminal a.a. 441-446 in the heme binding site of CYP3As [5*-CCTCATKCCAAKGCARTT-3*] were gifts from David J. Fraser and Dr. James R. Halpert, University of Arizona, AZ, USA. RT-PCR and DNA sequencing. First-strand cDNA synthesis and subsequent amplification were performed using the ‘‘GeneAmp RNAPCR kit’’ by Perkin Elmer, manufactured by Roche Molecular Systems, Inc., Branchburg, NJ, USA. Total RNA (1 mg) was reverse transcribed with priming by random hexamers. Amplification was performed in the same tube with N-terminal primers targeted to a.a. 123-128 or a.a. 434-444 and C-terminal primers targeted to a.a. 434444 or a.a. 441-446, each at 1 mM. The PCR profile recommended by Perkin Elmer was used with the annealing temperature at 45 7C. The PCR products were detected after separation on an ethidium bromide-stained 1 % (w/v) agarose gel in 1 1 TAE buffer pH 8.5 (40 mM Tris/acetate/1 mM EDTA) (13). The PCR products obtained from 2 separate PCR reactions were excised from the gel using a clean razor blade and was purified using glassmilk (Geneclean II; Bio 101, La Jolla, CA, USA). The purified PCR fragments were directly ligated into the pT7Blue(R) vector and the ligation products were used to transform Novablue competent cells (Novagen, Madison, WI, USA). Blue-white screening was employed to identify transformed cells. White colonies were picked using a sterile toothpick and plasmid DNA was prepared using the ‘‘maxi-prep’’ boiling method (13). Clones containing the insert were identified by restriction digest of the purified plasmid DNA using EcoRI and PstI (Promega, Madison, WI, USA) and were selected for DNA sequencing. Plasmid DNA from 2 separate clones were sequenced in duplicate samples using an automated sequencer (LI-COR, Inc., Lincoln, NE, USA) and the ‘‘SequiTherm EXCEL Long-Read DNA-Sequencing Kit-LC’’ protocol (Epicentre Technologies, Madison, WI, USA). Phylogenetic analyses of CYP genes. The rat (Rattus norvegicus) CYP1A1, 1A2, 1B1, 2B1, 2C6, 2D1, 2E1, 2G2, 3A1, 3A2, 3A9, 3A18, 3A23, 4A1; human (Homo sapiens) 3A3, 3A4, 3A5, 3A7; monkey (Macaca fascicularis) 3A8; rabbit (Oryctolagus cuniculus) 3A6; hamster (Mesocricetus auratus) 3A10; mouse (Mus musculus) 3a11, 3a13, 3a16, 3a25; dog (Canis familiaris) 3A12; guinea pig (Cavia porcellus) 3A14, 3A15, 3A17, 3A20; goat (Capra hircus) 3A19; sheep (Ovis aries) 3A24, bovine (Bos taurus) 3A28 and pig (Sus scrofa) 3A29 nucleotide
and deduced a.a. sequences were retrieved from the GenBank database. The accession numbers of each gene or partial gene sequence are listed in TABLE 1. The F. heteroclitus cDNA sequence was aligned with these mammalian CYP sequences using the Clustal W 1.4 program. Phylogenetic trees were constructed with the distancematrix method (14) using the Clustal W 1.4 program and the maximum parsimony method (15) using the PAUP 3.1.1 program.
RESULTS A cDNA fragment was amplified from Fundulus heteroclitus liver RNA by RT-PCR using degenerate PCR primers targeted to a.a. 304-311 and each of the two different PCR primers targeted to the heme binding site of CYP3As. No visible PCR product was obtained using the degenerate PCR primer targeted to a.a. 123128. The PCR products obtained from two different PCR reactions were cloned and sequenced. The inserts were identical, and were 430 base pairs each, including both the 5*-end and the 3*-end PCR primer regions. The nucleotide and the deduced a.a. sequences of these cDNA fragments between the two primer sites are shown in FIG. 1. The deduced 124 long a.a. sequence was compared to other sequences in a BLAST search. The results showed that the F. heteroclitus sequence was 57 to 70 % identical to mammalian members of the CYP3A subfamily and the p-value for each alignment with these sequences ranged from 1.8 1 10059 to 7.6 1 10040. This F. heteroclitus sequence has been classified as CYP3A30 (Nelson, personal communication). Figure 2 shows the alignment of F. heteroclitus CYP3A30 with 25 mammalian CYP3A sequences obtained from GenBank. Figure 2 also shows the a.a. in this region that are conserved among all these mammalian CYP3A sequences, among mammalian and F. heteroclitus, and the a.a. residues in CYP3A30 not seen in mammalian CYP3As. The F. heteroclitus CYP3A30 a.a. sequence was used to infer phylogenetic relationships between this fish cDNA sequence and the corresponding region of repre-
FIG. 1. Nucleotide and deduced amino acid sequence of the CYP3A30 cDNA fragment from Fundulus heteroclitus (killifish).
307
AID
BBRC 6956
/
6932$$$521
06-30-97 19:03:58
bbrcgs
AP: BBRC
Vol. 236, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. (A) Alignment of the F. heteroclitus CYP3A30 sequence and mammalian CYP3A sequences from amino acid (a.a.) 313 to 436 using the MacVector 6.0 program. Boxes represent a.a. where more than 75 % of the a.a. are identical (shaded) or similar (unshaded) in all the 26 CYP3A sequences analyzed. The conserved consensus sequence is shown at the bottom of the alignment and each letter represent identical a.a. and : represents similar a.a. in more than 75 % of the sequenced analyzed. The mammalian CYP3A sequences were retrieved from GenBank and the accession numbers are listed in TABLE 1. (B) Alignment of the F. heteroclitus CYP3A30 sequence and human CYP3A4 sequence from amino acid (a.a.) 313 to 436. The conserved sequence in all CYP3A, in all mammalian CYP3As and residues unique to F. heteroclitus CYP3A30 are shown.
308
AID
BBRC 6956
/
6932$$6956
06-30-97 19:03:58
bbrcgs
AP: BBRC
Vol. 236, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
formed a cluster and finally the hamster 3A10, the rat 3A18, the mouse 3a25 and the F. heteroclitus 3A30 sequences formed a cluster (FIG. 4A, 4B and 4C).
TABLE 1
Accession Numbers of Mammalian CYP Genes Gene
Species
Accession number
CYP1A1 CYP1A2 CYP1B1 CYP2B1 CYP2C6 CYP2E1 CYP3A1 CYP3A2 CYP3A3 CYP3A4 CYP3A5 CYP3A6 CYP3A7 CYP3A8 CYP3A9 CYP3A10 Cyp3a11 CYP3A12 Cyp3a13 CYP3A14 CYP3A15 Cyp3a16 CYP3A17 CYP3A18 CYP3A19 CYP3A20 CYP3A24 CYP3a25 CYP3A28 CYP3A29 CYP4A1
rat rat rat rat rat rat rat rat human human human rabbit human monkey rat hamster mouse dog mouse guinea pig guinea pig mouse guinea pig rat goat guinea pig sheep mouse bovine pig rat
K02246 K02422 X83867 J00719 M18335 J02627 M10161 M13646 D00003 M18907 J04813 J05034 D00408 S53047 U46118 M73992 X60452 X54915 X63023 D16363 D26487 D26137 D28515 X79991 X76503 D49731 U59378 Y11995 Y10214 Z93099 M14972
DISCUSSION The CYP3A subfamily in mammals has received considerable attention over the past ten years, because of its clinical significance as a result of metabolism of therapeutic drugs (3, 4). Expression of CYP3A also has been shown to be induced by therapeutic drugs and other compounds, including natural products, PCBs and pesticides (5, 16-18). Few reports have addressed questions concerning the CYP3A subfamily in non-mammalian species, although many non-mammalian wildlife populations are continuously exposed to CYP3A inducers/substrates in their natural environment as a result of food preferences and human activities. Earlier reports have suggested that CYP3A-like proteins are prominent CYP forms in untreated fish, and there are marked similarities between these CYP3Alike proteins in different fish species and mammalian CYP3A proteins (11, 12, 19, 20). In this report, the cDNA fragment isolated from F. heteroclitus showed 57 to 70 % identity with the same region of mammalian CYP3A genes retrieved from GenBank, and was termed CYP3A30. Another teleost fish CYP3A gene, CYP3A27, has been cloned from liver of rainbow trout (Oncorhynchus mykiss) (21). The F. heteroclitus CYP3A30 is 77
senting mammalian members of gene families CYP1, 2, 3 and 4. A phylogenetic tree was generated using the distance-matrix method and it showed that the F. heteroclitus CYP3A30 and the mammalian CYP3A sequences were assembled in a major branch that was distinct from the CYP1, CYP2 and CYP4 branches (FIG. 3). Phylogenetic trees showing the cluster of the CYP3A a.a. sequences were constructed using the maximum parsimony method (FIG. 4A) and the distance-matrix method (FIG. 4B). A phylogenetic tree of the nucleotide sequences of the same region of these CYP3A genes also was constructed using the distance-matrix method (FIG. 4C). The topologies of all three trees were very similar, showing six major branches within the CYP3A cluster (FIG. 4A, 4B and 4C). All the primate sequences (3A3, 3A4, 3A5, 3A7 and 3A8) clustered together and all the guinea pig sequences (3A14, 3A15, 3A17 and 3A20) clustered together. The dog 3A12 sequence and the ungulate sequences, goat 3A19, sheep 3A24, bovine 3A28 and pig 3A29, formed a cluster. The rat 3A1, 3A2, 3A23 and the mouse 3a11 and 3a16 sequences formed a cluster. The mouse 3a13 and the rat 3A9 sequences
FIG. 3. Phylogeny of CYP amino acid sequences. The tree inferred by the distance-matrix method using the Clustal W 1.4 program. Distances between genes are represented by the sum of their horizontal separation. The scale bar under the tree indicates the distance corresponding to 11 differences in 100 positions. Numbers above branches represent bootstrap values based on 100 samplings and are a measure of relative confidence of a grouping.
309
AID
BBRC 6956
/
6932$$$521
06-30-97 19:03:58
bbrcgs
AP: BBRC
Vol. 236, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. Phylogeny of CYP3A sequences. The tree inferred by the maximum parsimony method using a heuristic procedure and the PAUP 3.1.1 program for amino acid sequences is shown in (A). The tree inferred by the distance-matrix method using the Clustal W 1.4 program is shown for amino acid sequences in (B) and for nucleotide sequences in (C). Distances between genes are represented by the sum of their horizontal separation. The scale bar under the tree indicates the distance corresponding to 4 differences in 100 positions. Numbers above branches represent bootstrap values based on 100 samplings and are a measure of relative confidence of a grouping of genes.
% identical to the corresponding region of rainbow trout CYP3A27 (drnelson.utmem.edu/bibloB.html). Phylogenetic analysis of CYP3A genes disclosed six major branches within this subfamily. The topology of this clustering seems to be determined partly by the taxa from which the various genes were isolated. The primate sequences formed one cluster, the guinea pig sequences formed another cluster and the dog and the ungulate sequences formed a third cluster. Furthermore, five different rat CYP3A sequences were organized into three separate clusters, two of which also contained other rodent CYP3A sequences. Thus, the brain specific rat CYP3A9 clustered together with the
aflatoxin B1-metabolizing mouse Cyp3a13 and the female specific rat CYP3A18 clustered together with the male specific hamster CYP3A10 and with mouse Cyp3a25. The rifampicin-inducible rabbit CYP3A6 sequence did not cluster with any of these groups, but seems to represent an additional branch. This suggests that there has been independent diversification of CYP3A genes in different mammalian groups or possibly allelic variation of these genes. The topology of the CYP3A clusters generated with these partial sequences (Fig. 4) was the same as that of the distance-matrix phylogenetic tree with full length mammalian CYP3A sequences (not shown). This indi-
310
AID
BBRC 6956
/
6932$$$521
06-30-97 19:03:58
bbrcgs
AP: BBRC
Vol. 236, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
cates that the analysis based on the region of the CYP3A genes presented here is representative of the entire coding sequence of these genes. The scattered distribution of the rat and mouse genes is intriguing, and it is possible that this grouping of genes reflects functional similarities between different forms, and may indicate orthologous relationships. This is best illustrated for rat where the steroid-inducible forms, CYP3A1, CYP3A2 and CYP3A23, all cluster together, whereas the brain specific CYP3A9 and the female specific CYP3A18 are found in two different clusters. The CYP3A subfamily has been most studied in rat and mouse and it is conceivable that as additional genes from other species are sequenced, some will group with these various rodent-clusters. The F. heteroclitus CYP3A30 sequence obtained here clusters with the sex-specific hamster CYP3A10 and rat CYP3A18 sequences. Interestingly, sex-specific differences have been reported in CYP3A-like protein expression in salmonoid fish species during their reproductive cycle (19, 22) and there appear to be differences between piscine CYP3A and mammalian CYP3A in terms of response to traditional CYP3A inducers (12, 19, 23). However, whether the grouping of the F. heteroclitus CYP3A30 with CYP3A10, CYP3A18 and mouse Cyp3a25 reflects functional- and/or regulatory similarities between these enzymes requires further investigations. Moreover, immunochemical analyses have suggested the existence of multiple CYP3A-like proteins in liver from several teleost species including F. heteroclitus (12) and it is possible that these proteins represent different teleost CYP3A genes. With the current nomenclature system, each new gene in a CYP subfamily, with the exception of the CYP1 genes, is given a new number. This system does not reveal orthologous relationships between genes from different species and as more CYP genes are sequenced, comparison among genes within a subfamily becomes increasingly complicated. The CYP3As and the CYP2C and CYP2D subfamilies, in particular, illustrate these difficulties and suggest the need for a change in the nomenclature system to include functional and regulatory relationships, where possible. The overall high degree of similarity between the fish CYP3A30 cDNA sequence and mammalian CYP3A forms indicates that this part of the gene has been well conserved during vertebrate evolution. Evidence has been presented for the existence of CYP3A-like proteins in invertebrates (24, 25). A new CYP gene recently has been isolated from the clam, Mercenaria mercenaria, and it has been classified as CYP30 (Brown and Van Beneden, personal communication). This gene shows greater similarity to the CYP3A genes than to other CYP genes and it may represent an early divergence in the line leading to CYP3A. Incorporating invertebrate sequences into phylogenetic analyses could
indicate regions of these genes conserved through evolution. Additional sequences from fish and other phyla will be important to understand the evolution of these genes and may disclose orthologous relationships among the CYP3As. ACKNOWLEDGMENTS This study was supported by grants from the Swedish Council for Forest and Agricultural Research (to M. Celander) and Sea Grant NA46RG 0470 R/P60 (to J. J. Stegeman). We gratefully acknowledge the generous gift of PCR-primers from David Fraser and Dr. James Halpert, University of Arizona, Tuscon. We also thank Rachel Cox and Dr. Mark Hahn for valuable discussions and comments on the manuscript and Drs. Rebecca Van Beneden and David Brown for sharing the information on CYP30 with us. This is contribution number 9447 from the Woods Hole Oceanographic Institution.
REFERENCES 1. De Waziers, I., Cugnenc, P. H., Yang, C. S., Leroux, J.-P., and Beaune, P. H. (1990) J. Pharmacol. Exper. Ther. 253, 387–394. 2. Kerr, B. M., Thummel, K. E., Wurden, C. J., Klein, S. M., Kroetz, D. L., Gonzalez, F. J., and Levy, R. H. (1994) Biochem. Pharmacol. 47, 1969–1979. 3. Perotti, B. Y. T., Okudaira, N., Prueksaritanto, T., and Benet, L. Z. (1994) Drug Metab. Dispos. 22, 85–89. 4. Guengerich, F. P., and Shimada, T. (1991) Chem. Res. Toxicol. 4, 391–407. 5. Daujat, M., Pichard, L., Fabre, I., Pineau, T., Fabre, G., Bonfils, C., and Maurel, P. (1991) Met. Enzymol. 206, 345–353. 6. Schuetz, E. G., Schuetz, J. D., Strom, S. C., Thompson, M. T., Fisher, R. A., Molowa, D. T., Li, D., and Guzelian, P. S. (1993) Hepatology 18, 1254–1262. 7. Pichard, L., Fabre, I., Daujat, M., Domergue, J., Joyeux, H., and Maurel, P. (1992) Mol. Pharmacol. 41, 1047–1055. 8. Halpert, J. R., Guengerich, F. P., Bend, J. R., and Correia, M. A. (1994) Toxicol. Appl. Pharmacol. 125, 163–175. 9. Kostrubsky, V. E., Wood, S. G., Bush, M. D., Szakacs, J., Bement, W. J., Sinclair, P. R., Jeffery, E. H., and Sinclair, J. F. (1995) Biochem. Pharmacol. 50, 1743–1748. 10. Nelson, D. R., Koymans, L., Kamataki, T., Stegeman, J. J., Feyereisen, R., Waxman, D. J., Waterman, M. R., Gotoh, O., Coon, M. J., Estabrook, R. W., Gunsalus, I. C., and Nebert, D. W. (1996) Pharmacogenetics 6, 1–42. 11. Miranda, C. L., Wang, J.-L., Henderson, M. C., Zhao, X., Guengerich, F. P., and Buhler, D. R. (1991) Biochem. Biophys. Res. Commun. 176, 558–563. 12. Celander, M., Buhler, D. R., Fo¨rlin, L., Goksøyr, A., Miranda, C. L., Woodin, B. R., and Stegeman, J. J. (1996) Fish Physiol. Biochem. 15, 323–332. 13. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 14. Olsen, G. J. (1988) Methods Enzymol. 164, 793–812. 15. Swofford, D. L. (1990) PAUP: Phylogenetic Analyses Using Parsimony, Illinois Natural History Survey, Champaign, IL. 16. Curi-Pedrosa, R., Daujat, M., Pichard, L., Ourlin, J. C., Clair, P., Gervot, L., Lesca, P., Domergue, J., Joyeux, H., Fourtanier, G., and Maurel, P. (1994) J. Pharmacol. Exper. Ther. 269, 384– 269.
311
AID
BBRC 6956
/
6932$$$521
06-30-97 19:03:58
bbrcgs
AP: BBRC
Vol. 236, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
17. Dubois, M., De Waziers, I., Thome, J. P., and Kremers, P. (1996) Comp. Biochem. Physiol. 113C, 51–59. 18. Paolini, M., Pozzetti, L., Mesirca, R., Sapone, A., and CantelliForti, G. (1996) Arch. Toxicol. 70, 451–456. 19. Celander, M., Ronis, M., and Fo¨rlin, L. (1989) Mar. Environ. Res. 28, 9–13. 20. Husøy, A. M., Myers, M. S., Willis, M. J., Collier, T. K., Celander, M., and Goksøyr, A. (1994) Toxicol. Appl. Pharmacol. 129, 294– 308.
21. Lee, S-J., Yang, Y.-H., Wang, J.-L., Miranda, C. V., Lech, J. J., and Buhler, D. R. [In preparation] 22. Larsen, H. E., Celander, M., and Goksøyr, A. (1992) Fish Physiol. Biochem. 10, 291–301. 23. Celander, M., Hahn, M. E., and Stegeman, J. J. (1996) Arch. Biochem. Physiol. 329, 113–122. 24. Wootton, A. N., Goldfarb, P. S., Lemaire, P., O’Hara, S. C. M., and Livingstone, D. R. Mar. Environ. Res. 42, 297–301. 25. Oberdo¨rster, E., Rittschof, D., McClelland-Green, P. (1997) Fund. Appl. Toxicol. Supplement The Toxicologist 36, 136.
312
AID
BBRC 6956
/
6932$$$521
06-30-97 19:03:58
bbrcgs
AP: BBRC
236, 306–312 (1997)
RC976956
Isolation of a Cytochrome P450 3A cDNA Sequence (CYP3A30) from the Marine Teleost Fundulus heteroclitus and Phylogenetic Analyses of CYP3A Genes Malin Celander1 and John J. Stegeman2 Biology Department, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543
Received June 11, 1997
A reverse transcriptase polymerase chain reaction (RT-PCR) protocol, using degenerate PCR-primers specific to highly conserved regions of mammalian CYP3A genes, was employed to amplify a 400 base pair cDNA fragment from Fundulus heteroclitus liver RNA. The 124 amino acid sequence deduced from this cDNA sequence was aligned with corresponding sequences from representative members from the CYP1, 2, 3, and 4 gene families retrieved from the GenBank database. Phylogenetic trees were constructed using distancematrix and maximum parsimony methods. The F. heteroclitus sequence and all mammalian CYP3A sequences cluster together when compared to sequences of members of CYP gene families 1, 2, and 4. This fish sequence was 57 to 70 % identical to the corresponding region of mammalian CYP3A genes. These data indicate that the sequence obtained from F. heteroclitus represents a teleost fish CYP3A gene and it has been designated CYP3A30. q 1997 Academic Press
Members of the cytochrome P450 3A (CYP3A) subfamily are major constitutively expressed CYP forms in the liver and in the gastro-intestinal tract of mammals (1). CYP3A enzymes metabolize steroids and a wide variety of non-steroidal lipophilic organic molecules, many of which commonly are used as therapeutic drugs (2, 3). CYP3A enzymes also can activate several pro-carcinogens and are therefore believed to be involved in chemical carcinogenesis (4). Many substrates of CYP3A enzymes, including glucocorticoids and macrolide antibiotics, induce the expression of CYP3A genes (5, 6). Although the mechanisms for induction by these various inducers are not yet understood, some 1 Permanent address: Department of Zoophysiology, University of Go¨teborg, Medicinaregatan 18, S-413 90 Go¨teborg, Sweden. 2 To whom correspondence should be addressed. Fax: (508) 457 2169. E-mail: [email protected].
progress has been made in understanding drug metabolism and drug-drug interactions as a result of CYP3A activities (7, 8, 9). The CYP3A subfamily is one of the largest in the P450 superfamily and multiple CYP3A genes have been identified in several species (10). At present there is insufficient information on the properties of most CYP3A forms to identify orthologous relationships among the multiple genes within this subfamily. Most studies of structure, function and regulation of CYP3As have been in mammalian systems, whereas relatively little is known about CYP3A in other vertebrate groups. Catalytic and immunochemical analyses have pointed to the existence of a CYP3A-like protein in fish (11, 12). However, sequence data are required before these proteins can be assigned to a subfamily in the P450 superfamily. In this paper we present evidence for the presence of a CYP3A gene in a marine teleost fish species, Fundulus heteroclitus, and we also provide a phylogenetic analysis of CYP3A genes. Fish represent the earliest diverging vertebrates in evolution and it is therefore important to include fish gene sequences in phylogenetic analyses if we are to understand the evolution of vertebrate CYP3A genes. MATERIALS AND METHODS Animals and RNA isolation. Killifish, Fundulus heteroclitus, were captured by minnow traps in salt marshes on Cape Cod, MA, USA. The fish were killed by a cervical transection and the liver was dissected out and immediately placed in liquid nitrogen. Frozen livers from 8 untreated adult males were pulverized under liquid nitrogen and the total RNA fraction was directly isolated using the ‘‘Perfect RNA Total RNA Isolation Kit (Maxi Scale)’’ purchased from 5 Primer3 Prime Inc., Boulder, CO, USA. The quality of the preparation was investigated on an ethidium bromide-stained 1 % (w/v) agarose gel in 2.2 M formaldehyde, 1 1 MOPS/acetate-buffer pH 7.0 (20 mM MOPS/8 mM Na-acetate/1 mM EDTA) and the concentration and the purity of the RNA fraction was determined spectrophotometrically at A260 and A280 (13). Oligonucleotides. Oligonucleotides were designed based on conserved regions of 17 mammalian CYP3A deduced amino acid (a.a.)
306
0006-291X/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.
AID
BBRC 6956
/
6932$$$521
06-30-97 19:03:58
bbrcgs
AP: BBRC
Vol. 236, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
sequences as aligned by Dr. David Nelson (www.drnelson.utmem.edu/nelsonhomepage.html). The sequence of the N-terminal primer targeted to a.a. 123-128 of CYP3As is [5*-GAYGARGARTGGAARMG-3*] having a degeneracy of 32. The sequence of the C-terminal primer targeted to a.a. 434-444 in the heme binding site of CYP3As is [5*-CCDATRCARTTICKIGGICC-3*] having a degeneracy of 24. These oligonucleotides were synthesized by National Biosciences (Plymouth, MN, USA). Degenerate primers targeted to the N-terminal a.a. 304-311 of CYP3As [5*-TTTGCKGGNTATGARACMACNAGCAG-3*] and to the C-terminal a.a. 441-446 in the heme binding site of CYP3As [5*-CCTCATKCCAAKGCARTT-3*] were gifts from David J. Fraser and Dr. James R. Halpert, University of Arizona, AZ, USA. RT-PCR and DNA sequencing. First-strand cDNA synthesis and subsequent amplification were performed using the ‘‘GeneAmp RNAPCR kit’’ by Perkin Elmer, manufactured by Roche Molecular Systems, Inc., Branchburg, NJ, USA. Total RNA (1 mg) was reverse transcribed with priming by random hexamers. Amplification was performed in the same tube with N-terminal primers targeted to a.a. 123-128 or a.a. 434-444 and C-terminal primers targeted to a.a. 434444 or a.a. 441-446, each at 1 mM. The PCR profile recommended by Perkin Elmer was used with the annealing temperature at 45 7C. The PCR products were detected after separation on an ethidium bromide-stained 1 % (w/v) agarose gel in 1 1 TAE buffer pH 8.5 (40 mM Tris/acetate/1 mM EDTA) (13). The PCR products obtained from 2 separate PCR reactions were excised from the gel using a clean razor blade and was purified using glassmilk (Geneclean II; Bio 101, La Jolla, CA, USA). The purified PCR fragments were directly ligated into the pT7Blue(R) vector and the ligation products were used to transform Novablue competent cells (Novagen, Madison, WI, USA). Blue-white screening was employed to identify transformed cells. White colonies were picked using a sterile toothpick and plasmid DNA was prepared using the ‘‘maxi-prep’’ boiling method (13). Clones containing the insert were identified by restriction digest of the purified plasmid DNA using EcoRI and PstI (Promega, Madison, WI, USA) and were selected for DNA sequencing. Plasmid DNA from 2 separate clones were sequenced in duplicate samples using an automated sequencer (LI-COR, Inc., Lincoln, NE, USA) and the ‘‘SequiTherm EXCEL Long-Read DNA-Sequencing Kit-LC’’ protocol (Epicentre Technologies, Madison, WI, USA). Phylogenetic analyses of CYP genes. The rat (Rattus norvegicus) CYP1A1, 1A2, 1B1, 2B1, 2C6, 2D1, 2E1, 2G2, 3A1, 3A2, 3A9, 3A18, 3A23, 4A1; human (Homo sapiens) 3A3, 3A4, 3A5, 3A7; monkey (Macaca fascicularis) 3A8; rabbit (Oryctolagus cuniculus) 3A6; hamster (Mesocricetus auratus) 3A10; mouse (Mus musculus) 3a11, 3a13, 3a16, 3a25; dog (Canis familiaris) 3A12; guinea pig (Cavia porcellus) 3A14, 3A15, 3A17, 3A20; goat (Capra hircus) 3A19; sheep (Ovis aries) 3A24, bovine (Bos taurus) 3A28 and pig (Sus scrofa) 3A29 nucleotide
and deduced a.a. sequences were retrieved from the GenBank database. The accession numbers of each gene or partial gene sequence are listed in TABLE 1. The F. heteroclitus cDNA sequence was aligned with these mammalian CYP sequences using the Clustal W 1.4 program. Phylogenetic trees were constructed with the distancematrix method (14) using the Clustal W 1.4 program and the maximum parsimony method (15) using the PAUP 3.1.1 program.
RESULTS A cDNA fragment was amplified from Fundulus heteroclitus liver RNA by RT-PCR using degenerate PCR primers targeted to a.a. 304-311 and each of the two different PCR primers targeted to the heme binding site of CYP3As. No visible PCR product was obtained using the degenerate PCR primer targeted to a.a. 123128. The PCR products obtained from two different PCR reactions were cloned and sequenced. The inserts were identical, and were 430 base pairs each, including both the 5*-end and the 3*-end PCR primer regions. The nucleotide and the deduced a.a. sequences of these cDNA fragments between the two primer sites are shown in FIG. 1. The deduced 124 long a.a. sequence was compared to other sequences in a BLAST search. The results showed that the F. heteroclitus sequence was 57 to 70 % identical to mammalian members of the CYP3A subfamily and the p-value for each alignment with these sequences ranged from 1.8 1 10059 to 7.6 1 10040. This F. heteroclitus sequence has been classified as CYP3A30 (Nelson, personal communication). Figure 2 shows the alignment of F. heteroclitus CYP3A30 with 25 mammalian CYP3A sequences obtained from GenBank. Figure 2 also shows the a.a. in this region that are conserved among all these mammalian CYP3A sequences, among mammalian and F. heteroclitus, and the a.a. residues in CYP3A30 not seen in mammalian CYP3As. The F. heteroclitus CYP3A30 a.a. sequence was used to infer phylogenetic relationships between this fish cDNA sequence and the corresponding region of repre-
FIG. 1. Nucleotide and deduced amino acid sequence of the CYP3A30 cDNA fragment from Fundulus heteroclitus (killifish).
307
AID
BBRC 6956
/
6932$$$521
06-30-97 19:03:58
bbrcgs
AP: BBRC
Vol. 236, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. (A) Alignment of the F. heteroclitus CYP3A30 sequence and mammalian CYP3A sequences from amino acid (a.a.) 313 to 436 using the MacVector 6.0 program. Boxes represent a.a. where more than 75 % of the a.a. are identical (shaded) or similar (unshaded) in all the 26 CYP3A sequences analyzed. The conserved consensus sequence is shown at the bottom of the alignment and each letter represent identical a.a. and : represents similar a.a. in more than 75 % of the sequenced analyzed. The mammalian CYP3A sequences were retrieved from GenBank and the accession numbers are listed in TABLE 1. (B) Alignment of the F. heteroclitus CYP3A30 sequence and human CYP3A4 sequence from amino acid (a.a.) 313 to 436. The conserved sequence in all CYP3A, in all mammalian CYP3As and residues unique to F. heteroclitus CYP3A30 are shown.
308
AID
BBRC 6956
/
6932$$6956
06-30-97 19:03:58
bbrcgs
AP: BBRC
Vol. 236, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
formed a cluster and finally the hamster 3A10, the rat 3A18, the mouse 3a25 and the F. heteroclitus 3A30 sequences formed a cluster (FIG. 4A, 4B and 4C).
TABLE 1
Accession Numbers of Mammalian CYP Genes Gene
Species
Accession number
CYP1A1 CYP1A2 CYP1B1 CYP2B1 CYP2C6 CYP2E1 CYP3A1 CYP3A2 CYP3A3 CYP3A4 CYP3A5 CYP3A6 CYP3A7 CYP3A8 CYP3A9 CYP3A10 Cyp3a11 CYP3A12 Cyp3a13 CYP3A14 CYP3A15 Cyp3a16 CYP3A17 CYP3A18 CYP3A19 CYP3A20 CYP3A24 CYP3a25 CYP3A28 CYP3A29 CYP4A1
rat rat rat rat rat rat rat rat human human human rabbit human monkey rat hamster mouse dog mouse guinea pig guinea pig mouse guinea pig rat goat guinea pig sheep mouse bovine pig rat
K02246 K02422 X83867 J00719 M18335 J02627 M10161 M13646 D00003 M18907 J04813 J05034 D00408 S53047 U46118 M73992 X60452 X54915 X63023 D16363 D26487 D26137 D28515 X79991 X76503 D49731 U59378 Y11995 Y10214 Z93099 M14972
DISCUSSION The CYP3A subfamily in mammals has received considerable attention over the past ten years, because of its clinical significance as a result of metabolism of therapeutic drugs (3, 4). Expression of CYP3A also has been shown to be induced by therapeutic drugs and other compounds, including natural products, PCBs and pesticides (5, 16-18). Few reports have addressed questions concerning the CYP3A subfamily in non-mammalian species, although many non-mammalian wildlife populations are continuously exposed to CYP3A inducers/substrates in their natural environment as a result of food preferences and human activities. Earlier reports have suggested that CYP3A-like proteins are prominent CYP forms in untreated fish, and there are marked similarities between these CYP3Alike proteins in different fish species and mammalian CYP3A proteins (11, 12, 19, 20). In this report, the cDNA fragment isolated from F. heteroclitus showed 57 to 70 % identity with the same region of mammalian CYP3A genes retrieved from GenBank, and was termed CYP3A30. Another teleost fish CYP3A gene, CYP3A27, has been cloned from liver of rainbow trout (Oncorhynchus mykiss) (21). The F. heteroclitus CYP3A30 is 77
senting mammalian members of gene families CYP1, 2, 3 and 4. A phylogenetic tree was generated using the distance-matrix method and it showed that the F. heteroclitus CYP3A30 and the mammalian CYP3A sequences were assembled in a major branch that was distinct from the CYP1, CYP2 and CYP4 branches (FIG. 3). Phylogenetic trees showing the cluster of the CYP3A a.a. sequences were constructed using the maximum parsimony method (FIG. 4A) and the distance-matrix method (FIG. 4B). A phylogenetic tree of the nucleotide sequences of the same region of these CYP3A genes also was constructed using the distance-matrix method (FIG. 4C). The topologies of all three trees were very similar, showing six major branches within the CYP3A cluster (FIG. 4A, 4B and 4C). All the primate sequences (3A3, 3A4, 3A5, 3A7 and 3A8) clustered together and all the guinea pig sequences (3A14, 3A15, 3A17 and 3A20) clustered together. The dog 3A12 sequence and the ungulate sequences, goat 3A19, sheep 3A24, bovine 3A28 and pig 3A29, formed a cluster. The rat 3A1, 3A2, 3A23 and the mouse 3a11 and 3a16 sequences formed a cluster. The mouse 3a13 and the rat 3A9 sequences
FIG. 3. Phylogeny of CYP amino acid sequences. The tree inferred by the distance-matrix method using the Clustal W 1.4 program. Distances between genes are represented by the sum of their horizontal separation. The scale bar under the tree indicates the distance corresponding to 11 differences in 100 positions. Numbers above branches represent bootstrap values based on 100 samplings and are a measure of relative confidence of a grouping.
309
AID
BBRC 6956
/
6932$$$521
06-30-97 19:03:58
bbrcgs
AP: BBRC
Vol. 236, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. Phylogeny of CYP3A sequences. The tree inferred by the maximum parsimony method using a heuristic procedure and the PAUP 3.1.1 program for amino acid sequences is shown in (A). The tree inferred by the distance-matrix method using the Clustal W 1.4 program is shown for amino acid sequences in (B) and for nucleotide sequences in (C). Distances between genes are represented by the sum of their horizontal separation. The scale bar under the tree indicates the distance corresponding to 4 differences in 100 positions. Numbers above branches represent bootstrap values based on 100 samplings and are a measure of relative confidence of a grouping of genes.
% identical to the corresponding region of rainbow trout CYP3A27 (drnelson.utmem.edu/bibloB.html). Phylogenetic analysis of CYP3A genes disclosed six major branches within this subfamily. The topology of this clustering seems to be determined partly by the taxa from which the various genes were isolated. The primate sequences formed one cluster, the guinea pig sequences formed another cluster and the dog and the ungulate sequences formed a third cluster. Furthermore, five different rat CYP3A sequences were organized into three separate clusters, two of which also contained other rodent CYP3A sequences. Thus, the brain specific rat CYP3A9 clustered together with the
aflatoxin B1-metabolizing mouse Cyp3a13 and the female specific rat CYP3A18 clustered together with the male specific hamster CYP3A10 and with mouse Cyp3a25. The rifampicin-inducible rabbit CYP3A6 sequence did not cluster with any of these groups, but seems to represent an additional branch. This suggests that there has been independent diversification of CYP3A genes in different mammalian groups or possibly allelic variation of these genes. The topology of the CYP3A clusters generated with these partial sequences (Fig. 4) was the same as that of the distance-matrix phylogenetic tree with full length mammalian CYP3A sequences (not shown). This indi-
310
AID
BBRC 6956
/
6932$$$521
06-30-97 19:03:58
bbrcgs
AP: BBRC
Vol. 236, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
cates that the analysis based on the region of the CYP3A genes presented here is representative of the entire coding sequence of these genes. The scattered distribution of the rat and mouse genes is intriguing, and it is possible that this grouping of genes reflects functional similarities between different forms, and may indicate orthologous relationships. This is best illustrated for rat where the steroid-inducible forms, CYP3A1, CYP3A2 and CYP3A23, all cluster together, whereas the brain specific CYP3A9 and the female specific CYP3A18 are found in two different clusters. The CYP3A subfamily has been most studied in rat and mouse and it is conceivable that as additional genes from other species are sequenced, some will group with these various rodent-clusters. The F. heteroclitus CYP3A30 sequence obtained here clusters with the sex-specific hamster CYP3A10 and rat CYP3A18 sequences. Interestingly, sex-specific differences have been reported in CYP3A-like protein expression in salmonoid fish species during their reproductive cycle (19, 22) and there appear to be differences between piscine CYP3A and mammalian CYP3A in terms of response to traditional CYP3A inducers (12, 19, 23). However, whether the grouping of the F. heteroclitus CYP3A30 with CYP3A10, CYP3A18 and mouse Cyp3a25 reflects functional- and/or regulatory similarities between these enzymes requires further investigations. Moreover, immunochemical analyses have suggested the existence of multiple CYP3A-like proteins in liver from several teleost species including F. heteroclitus (12) and it is possible that these proteins represent different teleost CYP3A genes. With the current nomenclature system, each new gene in a CYP subfamily, with the exception of the CYP1 genes, is given a new number. This system does not reveal orthologous relationships between genes from different species and as more CYP genes are sequenced, comparison among genes within a subfamily becomes increasingly complicated. The CYP3As and the CYP2C and CYP2D subfamilies, in particular, illustrate these difficulties and suggest the need for a change in the nomenclature system to include functional and regulatory relationships, where possible. The overall high degree of similarity between the fish CYP3A30 cDNA sequence and mammalian CYP3A forms indicates that this part of the gene has been well conserved during vertebrate evolution. Evidence has been presented for the existence of CYP3A-like proteins in invertebrates (24, 25). A new CYP gene recently has been isolated from the clam, Mercenaria mercenaria, and it has been classified as CYP30 (Brown and Van Beneden, personal communication). This gene shows greater similarity to the CYP3A genes than to other CYP genes and it may represent an early divergence in the line leading to CYP3A. Incorporating invertebrate sequences into phylogenetic analyses could
indicate regions of these genes conserved through evolution. Additional sequences from fish and other phyla will be important to understand the evolution of these genes and may disclose orthologous relationships among the CYP3As. ACKNOWLEDGMENTS This study was supported by grants from the Swedish Council for Forest and Agricultural Research (to M. Celander) and Sea Grant NA46RG 0470 R/P60 (to J. J. Stegeman). We gratefully acknowledge the generous gift of PCR-primers from David Fraser and Dr. James Halpert, University of Arizona, Tuscon. We also thank Rachel Cox and Dr. Mark Hahn for valuable discussions and comments on the manuscript and Drs. Rebecca Van Beneden and David Brown for sharing the information on CYP30 with us. This is contribution number 9447 from the Woods Hole Oceanographic Institution.
REFERENCES 1. De Waziers, I., Cugnenc, P. H., Yang, C. S., Leroux, J.-P., and Beaune, P. H. (1990) J. Pharmacol. Exper. Ther. 253, 387–394. 2. Kerr, B. M., Thummel, K. E., Wurden, C. J., Klein, S. M., Kroetz, D. L., Gonzalez, F. J., and Levy, R. H. (1994) Biochem. Pharmacol. 47, 1969–1979. 3. Perotti, B. Y. T., Okudaira, N., Prueksaritanto, T., and Benet, L. Z. (1994) Drug Metab. Dispos. 22, 85–89. 4. Guengerich, F. P., and Shimada, T. (1991) Chem. Res. Toxicol. 4, 391–407. 5. Daujat, M., Pichard, L., Fabre, I., Pineau, T., Fabre, G., Bonfils, C., and Maurel, P. (1991) Met. Enzymol. 206, 345–353. 6. Schuetz, E. G., Schuetz, J. D., Strom, S. C., Thompson, M. T., Fisher, R. A., Molowa, D. T., Li, D., and Guzelian, P. S. (1993) Hepatology 18, 1254–1262. 7. Pichard, L., Fabre, I., Daujat, M., Domergue, J., Joyeux, H., and Maurel, P. (1992) Mol. Pharmacol. 41, 1047–1055. 8. Halpert, J. R., Guengerich, F. P., Bend, J. R., and Correia, M. A. (1994) Toxicol. Appl. Pharmacol. 125, 163–175. 9. Kostrubsky, V. E., Wood, S. G., Bush, M. D., Szakacs, J., Bement, W. J., Sinclair, P. R., Jeffery, E. H., and Sinclair, J. F. (1995) Biochem. Pharmacol. 50, 1743–1748. 10. Nelson, D. R., Koymans, L., Kamataki, T., Stegeman, J. J., Feyereisen, R., Waxman, D. J., Waterman, M. R., Gotoh, O., Coon, M. J., Estabrook, R. W., Gunsalus, I. C., and Nebert, D. W. (1996) Pharmacogenetics 6, 1–42. 11. Miranda, C. L., Wang, J.-L., Henderson, M. C., Zhao, X., Guengerich, F. P., and Buhler, D. R. (1991) Biochem. Biophys. Res. Commun. 176, 558–563. 12. Celander, M., Buhler, D. R., Fo¨rlin, L., Goksøyr, A., Miranda, C. L., Woodin, B. R., and Stegeman, J. J. (1996) Fish Physiol. Biochem. 15, 323–332. 13. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 14. Olsen, G. J. (1988) Methods Enzymol. 164, 793–812. 15. Swofford, D. L. (1990) PAUP: Phylogenetic Analyses Using Parsimony, Illinois Natural History Survey, Champaign, IL. 16. Curi-Pedrosa, R., Daujat, M., Pichard, L., Ourlin, J. C., Clair, P., Gervot, L., Lesca, P., Domergue, J., Joyeux, H., Fourtanier, G., and Maurel, P. (1994) J. Pharmacol. Exper. Ther. 269, 384– 269.
311
AID
BBRC 6956
/
6932$$$521
06-30-97 19:03:58
bbrcgs
AP: BBRC
Vol. 236, No. 2, 1997
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
17. Dubois, M., De Waziers, I., Thome, J. P., and Kremers, P. (1996) Comp. Biochem. Physiol. 113C, 51–59. 18. Paolini, M., Pozzetti, L., Mesirca, R., Sapone, A., and CantelliForti, G. (1996) Arch. Toxicol. 70, 451–456. 19. Celander, M., Ronis, M., and Fo¨rlin, L. (1989) Mar. Environ. Res. 28, 9–13. 20. Husøy, A. M., Myers, M. S., Willis, M. J., Collier, T. K., Celander, M., and Goksøyr, A. (1994) Toxicol. Appl. Pharmacol. 129, 294– 308.
21. Lee, S-J., Yang, Y.-H., Wang, J.-L., Miranda, C. V., Lech, J. J., and Buhler, D. R. [In preparation] 22. Larsen, H. E., Celander, M., and Goksøyr, A. (1992) Fish Physiol. Biochem. 10, 291–301. 23. Celander, M., Hahn, M. E., and Stegeman, J. J. (1996) Arch. Biochem. Physiol. 329, 113–122. 24. Wootton, A. N., Goldfarb, P. S., Lemaire, P., O’Hara, S. C. M., and Livingstone, D. R. Mar. Environ. Res. 42, 297–301. 25. Oberdo¨rster, E., Rittschof, D., McClelland-Green, P. (1997) Fund. Appl. Toxicol. Supplement The Toxicologist 36, 136.
312
AID
BBRC 6956
/
6932$$$521
06-30-97 19:03:58
bbrcgs
AP: BBRC