DNA AND CELL BIOLOGY Volume 9, Number 3, 1990 Mary Ann Liebert, Inc., Publishers
Pp. 205-212
Structure, Sequence, Chromosomal Location, and Evolution of the Human Ferredoxin Gene CHI-YAO
CHANG,*'t
DU-AN
WU,* T.K.
MOHANDAS,Í
Family
and BON-CHU CHUNG*
ABSTRACT Ferredoxin is an iron-sulfur protein that serves as an electron transport intermediate for mitochondrial cytochromes P450 involved in steroid, vitamin D, and bile acid metabolism. We cloned and characterized the human ferredoxin gene family, which includes two expressed genes and two pseudogenes. Sequence analysis of this gene family revealed that it encodes only one protein product. The expressed genes were assigned to chromosome 11 and pseudogenes to chromosomes 20 and 21 by identifying single-copy probes from each gene segment and hybridizing them to DNA from rodent-human hybrid cells. The pseudogenes lacked ¡litrons and contained numerous mutations, including insertion, deletion, and substitution which rendered them inactive. They were 96% and 85% homologous to the expressed gene, yet they were only 78% homologous with each other. The intronless nature, higher diversity among themselves, and distinct chromosomal location of the pseudogenes suggests that they arose by independent, retroposon-mediated events.
INTRODUCTION
protein small, acidic, Ferredoxin transport intermediate mitochondrial cytochromes P450. Electrons is
functions
iron-sulfur
a
as an
electron
that for
are trans-
ferred from NADPH
flavin-containing protein
through (ferredoxin oxidoreductase) and ferredoxin to the terminal cytochrome P450 for oxidation/reduction reactions. Mitoa
chondrial P450s and their ferredoxin are found mainly in the steroidogenic tissues, including adrenal, ovary, testis, and placenta (Jefcoate, 1986). Small amounts of them are also found in the liver and kidney for bile acid and vitamin D synthesis (Bjorkhem et ai, 1980; Driscoll and Omdahl, 1986; Jefcoate, 1986). Because of its relative abundance, the adrenal ferredoxin, designated adrenodoxin, has been characterized in most detail (Kimura et ai, 1969; Tanaka et ai, 1973). It is synthesized as a precursor in which 60 amino acids of the signal peptide is later cleaved upon transport into the mitochondrial inner matrix to form a mature protein of 124 amino acids (Okamura et ai, 1985). A single [2Fe-2S] center is incorporated into the apoprotein to form the activity center (Coghlan and Vickery,
1989).
Human adrenodoxin and placental ferredoxin cDNAs share identical sequence, suggesting that they are the same (Mittal et ai, 1988). At least three polyadenylation sites in the adrenodoxin cDNA have been identified; all three are functional and result in mRNAs of 1.75, 1.4, and 0.9 kb (Okamura et ai, 1985; Picado-Leonard et ai, 1988). Bovine adrenodoxin mRNA has multiple precursor sequences associated with the same mature sequence (Okamura et al., 1987). The ferredoxin RNA is detected in many tissues and its production is stimulated by various tropic hormones via cyclic AMP as an intracellular intermediate in the steroidogenic tissues (Kramer et ai, 1984; Anderson and Mendelson, 1985; Okamura et ai, 1985; Voutilainen et ai, 1988). The effect of cyclic AMP stimulation on the adrenodoxin and cytochrome P450 gene expression is at the transcriptional level (John et ai, 1986). To delineate the transcriptional mechanism of the adrenodoxin gene, we cloned the human gene (Chang et ai, 1988). It contains four exons and three introns, spanning more than 20 kb. In this report, we characterized the entire human ferredoxin gene family, including two expressed genes and two processed pseudogenes. We also localized these genes to human chromosomes 11, 20, and 21.
"Institute of Molecular Biology, Academia Sinica, Nankang, Taipei, Taiwan, 11529 Republic of China. t Institute of Life Sciences, Tsing-Hwa University, Hsin-Chu, Taiwan, Republic of China. ÎDepartment of Pediatrics and Division of Medical Genetics, Harbor/UCLA Medical Center, Torrance, CA 90509.
205
CHANG ET AL.
206 The structure, divergence, and chromosomal location of the pseudogenes suggest that they might arise by indepen-
dent, retroposon-mediated
events.
MATERIALS AND METHODS
General procedures DNA
dures
purification, sequencing, and hybridization procedescribed previously (Sanger et al, 1977;
were
Wyman and White, 1980; Maniatis, 1982; Mohandas et al, 1986), except that for the hybridization of the humanrodent hybrid DNA, the blot was washed additionally at 65°C for 10 min. Primers were synthesized by Gene Assembler (Pharmacia), and purified by a mono Q column
(Pharmacia).
Cloning ferredoxin
genes
DNA digested separately with five different restriction enzymes shown in Fig. 1. In each digest there were more than three hybridizing bands. The Eco RI digest showed seven bands with sizes of 11, 9.2, 8.3, 5.5, 3.2, 2.9, and 0.86 kb (the 0.86-kb band could be seen in other gels not shown here). We have previously demonstrated that the 8.3-, 5.5-, 3.2-, and 2.9-kb bands belong to an expressed adrenodoxin gene (Chang et al., 1988). The rest of the bands could represent other ferredoxin genes in the same family. To clone the human ferredoxin gene family, we screened three genomic DNA libraries (Maniatis et al, 1978; Bookstein et al, 1988) constructed in two different vectors to ascertain all the genes in the family were isolated. Two probes corresponding to the 5' region and the entire length of the cDNA were used to isolate five X clones hi to h5. Clones hi to h5 each contained three to eight Eco RI fragments in addition to both arms, representing 10 to 18 kb of inserts (Fig. 2, A and C). Among them one or two fragments in each clone showed hybridization to the cDNA probe (Fig. 2, B and D). They were 8.3 and 2.9 kb in hi (lane 2), 0.86 kb in h2 (lane 1), 9.2 kb in h3 (lane 3), 11 kb in h4 (lane 5), and 5.5 and 3.2 kb in h5 (lane 4). These bands accounted for all seven Eco RI bands seen in the genomic blot (Fig. 1); therefore, our collection of the human ferredoxin genomic clones appeared to be com-
Three human genomic DNA libraries were used in this study. The first library contained human genomic DNA partially digested with Eco RI and cloned into vector Charon 4A. The second library contained DNA digested partially by Alu l-Hae III in the same vector (Maniatis et al., 1978). Using a 1.3-kb cDNA probe, clones hi and h3 were isolated from the first library, while clones h2 and h4 plete. were isolated from the second library. Clone h5 was isoThe coding regions in these clones were located and their lated from the third library constructed by ligating DNA were determined and compared with that of the sequences partially digested with Mbo I and cloned into vector X dash cDNA. Clones hi and h4 were highly homologous and (Bookstein et al., 1988). This library was screened with a their restriction maps were also very similar (Fig. 3). One 400-bp Eco Rl-Xba I fragment from the 5' end of the cDNA (Picado-Leonard et al, 1988) to obtain h5. Cloned DNAs were digested with Eco RI and hybridized with cDNA probe to identify coding regions which were then subcloned into pUC18 and analyzed further by digestion and hybridization to locate exons. The exon-containS m i a m ¡é S ing subfragments were subcloned into M13 for sequence determination.
23. 1
Identification of single-copy probes were hybridized with 32P-labeled total hugenomic DNA to locate unique sequences which do not show hybridization upon short exposure. These DNAs were labeled and hybridized to a genomic Southern blot to ascertain that only one band appeared. A 500-bp Pst l-Eco RI fragment upstream from h2, a 180-bp Hinc II fragment located in a 6-kb Ssp I fragment downstream from the pseudogene h3, and a 414-bp Eco Rl-Pvu II fragment (PR410, covering -333/+81) in h5 appeared to be unique in this test (shown as black bars in Fig. 3). The single-copy probe in hi was previously identified (Morel et al., 1988).
Cloned DNAs
man
RESULTS
Isolation of the human ferredoxin genes A 1.3-kb human adrenodoxin cDNA probe Leonard et al., 1988) was hybridized to human
(Picadogenomic
6. 7
•" 30% of analyzed cells; (+) not 11%-30%; (-) 3%-10%; t(9;x) present in 90%; # t(x;10). The two lines at the bottom designate the number of cell lines discordant for each chromosome. Note that h2 probe shows no discordances for chromosome 20 and that h3 probe shows no discordances for chromosome 21.
Note:
+
=
detected;
Presence of
=
*
=
=
-
=
=
211
FERREDOXIN GENE FAMILY
Table 2. Single Base Substitutions
within the
Coding Regions
of
Ferredoxin Pseudogenes
No. substitutions in codon
Pseudogene designation
Chromosome location
position -
1st
2nd
3rd
Total no. of substitutions*
Time of integration (Myr)b
Percent
divergence
h2
20
8
6
10
19
3
5
h3
21
18
12
18
39
10
15
¡»Values were calculated assuming half of the substitutions in the third codon position were accounted for active gene, half in the pseudogene. bCalculated assuming the substitution rate is 7 x 10"' substitutions per nucleotide site per year.
ai, 1988). Renal ferredoxin has similar optical, magnetic, and immunochemical properties to adrenodoxin, although one report suggested that they had minor differences (Maruya et ai, 1983). Now that we have identified only one protein sequence, there is no need to designate ferredoxin according to their tissue origins. The same ferredoxin exists in the adrenal, placenta, kidney, and liver for steroid, vitamin D3, and bile acid metabolism. Therefore, ferredoxin is the more appropriate name than adrenodoxin, renodoxin, or hepatoredoxin. Both pseudogenes h2 and h3 lack introns and possess the properties of fully processed mRNA (Vanin, 1985). Although h3 does not have poly(A) tracks at the 3' end, its sequence ends at the polyadenylation site (17 nucleotides
downstream from AATAAA and coincides with the 3' end of cDNA hAdx-7) (Picado-Leonard et ai, 1988). H3 is flanked by 7-bp direct repeats TCAGTGA (underlined in Fig. 5), which usually result from réintégration of the cDNA into the chromosome (Vanin, 1985; Weiner et ai, 1986). These suggest that ferredoxin psdueogenes might arise by a retroposon-like mechanism having mRNA as an intermediate before integration (Vanin, 1985; Weiner et
ai, 1986).
The number of substitution within the coding region of h2 and h3 is tabulated in Table 2. Since neither h2 nor h3 carries a promoter, it is assumed that upon insertion both genes immediately lost activity and started to diverge from the active gene by neutral mutation. The number of substitutions was counted with the assumption that half of the changes in the third codon position were due to evolutionary drift in the functional gene, and half to changes in the pseudogene following integration (Lee et ai, 1983; Scarpulla and Wu, 1983). For neutral mutations, the rate of divergence is about 0.7% per Myr (million years) (Perler et ai, 1980). Thus, h2 apparently diverged from the expressed gene about 5 Myr ago. H3 diverged even more; it started to lose function about 15 Myr ago. Both pseudogenes appeared very recently, long after man and cattle diverged, about 85 Myr ago (Miller et ai, 1981). We located pseudogene h2 on chromosome 20 and h3 on chromosome 21. Different chromosomal locations of these genes support the notion that réintégration of the pro-
by evolutionary drift in the
cessed gene is a random event; therefore, they landed at different chromosomes by chance. These two pseudogenes are more homologous to the expressed gene (96% and 85%) than between themselves (only 78% homology). It is expected that they should maintain higher homology to their parent gene, which evolved more slowly due to positive selection pressure on an active gene (Miyata et ai, 1980). Homology between the pseudogenes, however, was lower, possibly because they evolved independently and there was no selective pressure to maintain their homology. Thus, the degree of divergence and chromosomal locations of h2 and h3 indicate that they might have arisen by inde-
pendent transposition
events.
ACKNOWLEDGMENTS We would like to thank Drs. Wen-Hwa Lee and James Shen for providing the human genomic DNA libraries and helpful discussion. D.-A.W. is a predoctoral student from the National Defense Medical Center, Republic of China. This work was supported by National Science Council, #NSC-78-0203-B001-ll, and Academia Sinica, Republic of China.
REFERENCES ANDERSON, CM., and MENDELSON, C.R. (1985). Regulation of steroidogenesis in rat Leydig cells in culture: Effect of human chorionic gonadotropin and dibutyryl cyclic AMP on the synthesis of cholesterol side chain cleavage cytochrome P-450 and adrenodoxin. Arch. Biochem. Biophys. 238, 378387. BJORKHEM, I., HOLMBERG, I., OFTEBRO, H., and PEDERSEN, J.I. (1980). Properties of a reconstituted vitamin D3 25-hydroxylase from rat liver mitochondria. J. Biol. Chem. 255, 5244-5249.
BOOKSTEIN, R., LEE, E.Y.-H.P., TO, H., YOUNG, L.-J., SERY, T.W., HAYES, R.C., FRIEDMANN, T., and LEE, W.-H. (1988). Human retinoblastoma susceptibility gene: Ge-
212
CHANG ET AL.
nomic organization and analysis of heterozygous intragenic deletion mutants. Proc. Nati. Acad. Sei. USA 85, 2210-2214. CHANG, C.-Y., WU, D.-A., LAI, C.-C, MILLER, W.L., and CHUNG, B.-C. (1988). Cloning and structure of the human adrenodoxin gene. DNA 7, 609-615.
CHASHCHIN, V.L., LAPKO, V.N., ADAMOVICH, T.B., KIRILLOVA, N.M., LAPKO, A.G., and AKHREM, A.A. (1986). The primary structure of hepatoredoxin from bovine liver mitochondria. Bioorg. Khim. 12, 1286-1289. COGHLAN, V.M., and VICKERY, L.E. (1989). Expression of human ferredoxin and assembly of the [2Fe-2S] center in Escherichia coli. Proc. Nati. Acad. Sei. USA 86, 835-839. DRISCOLL, W.J., and OMDAHL, J.L. (1986). Kidney and adrenal mitochondria contain two forms of NADPH-adrenodoxin reductase-dependent iron-sulfur proteins. Isolation of the two porcine renal ferredoxins. J. Biol. Chem. 261, 4122-4125. JEFCOATE, C.R. (1986). In Cytochrome P-450. P.R. Ortiz de Montellano, Ed. (Plenum Press, New York) pp. 387-428.
JOHN, M.E., JOHN, M.C., BOGGARAM, V., SIMPSON, E.R., and WATERMAN, M.R. (1986). Transcriptional regulation of steroid hydroxylase genes by corticotropin. Proc. Nati. Acad. Sei. USA
83, 4715-4719. KIMURA, T., SUZUKI, K., PADMANABHAN, R., SAMEJIMA, T., TARUTANI, O., and UI, N. (1969). Studies on steroid hydroxylase. Molecular properties of adrenal iron-sulfur protein. Biochemistry 8, 4027-4031. KRAMER, R.E., RAINEY, W.E., FUNKENSTEIN, B., DEE, A., SIMPSON, E.R., and WATERMAN, M.R. (1984). Induction of synthesis of mitochondrial steroidogenic enzymes of bovine adrenocortical cells by analogs of cyclic AMP. J. Biol. Chem. 259, 707-713. LEE, M.G.-S., LEWIS, S.A., WILDE, CD., and COWAN, N.J. (1983). Evolutionary history of a multigene family: An expressed human 0-tubulin gene and three processed pseudogenes. Cell 33, 477-487. MANIATIS, T., HARDISON, R.C., LACY, E., LAUER, J., O'CONNELL, C, QUON, D., SIM, G.K., and EFSTRATIADIS, A. (1978). The isolation of structural genes from libraries of eucaryotic DNA. Cell 15, 687-701. MANIATIS, T., FRITSCH, E.F., and SAMBROOK, J. (1982). In Molecular Cloning: A Laboratory Manual. (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY). MARUYA, N., HIWATASHI, A., ICHIKAWA, Y„ and YAMANO, T. (1983). Purification and characterization of renal ferredoxin from bovine renal mitochondria. J. Biochem. 93, 1239-1247. MILLER, W.L., COIT, D., BAXTER, J.D., and MARTIAL, J.A. (1981). Cloning of bovine prolactin cDNA and evolutionary implication of its sequence. DNA 1, 37-50. MITTAL, S., ZHU, Y.-Z., asnd VICKERY, L.E. (1988). Molecular cloning and sequence analysis of human placental ferredoxin. Arch. Biochem. Biophys. 264, 383-391. MIYATA, T., YASUNAGA, T., and NISHIDA, T. (1980). Nucleotide sequence divergence and functional constraint in mRNA evolution. Proc. Nati. Acad. Sei. USA 77, 7328-7332.
MOHANDAS, T.K., HEINZMANN, C, SPARKES, R.S., WASMUTH, J., EDWARDS, P., and LUSIS, A.J. (1986). Assignment of human 3-hydroxy-3-methylglutaryl coenzyme A reducíase gene to the ql3 q23 region of chromosome 5. Somat. Cell Mol. Genet. 12, 89-94. —
MOREL, Y, PICADO-LEONARD, J., WU, D.-A., CHANG, C.-Y., MOHANDAS, T.K., CHUNG, B.-C, and MILLER, W.L. (1988). Assignment of the functional gene for human adrenodoxin to chromosome llql3 —qter and of adrenodoxin
pseudogenes to chromosome 20cen ql3.1. Am. J. Hum. Genet. 43, 52-59. OKAMURA, T., JOHN, M.E., ZUBER, M.X., SIMPSON, E.R., and WATERMAN, M.R. (1985). Molecular cloning and —
amino acid sequence of the precursor form of bovine adrenodoxin: Evidence for a previously unidentified COOH-terminal peptide. Proc. Nati. Acad. Sei. USA 82, 5705-5709. OKAMURA, T., KAGIMOTO, M., SIMPSON, E.R., and WATERMAN, M.R. (1987). Multiple species of bovine adrenodoxin mRNA. Occurrence of two different mitochondrial precursor sequences associated with the same mature sequence. J. Biol. Chem. 262, 10335-10338. PERLER, F., EFSTRATIADIS, A., LOMEDICO, P., GILBERT, W., KOLODNER, R., and DODGSON, J. (1980). The evolution of genes: The chicken preproinsulin gene. Cell 20, 555-566.
PICADO-LEONARD, J., VOUTILAINEN, R., KAO, L.-C, CHUNG, B.-C, STRAUSS III, J.F., and MILLER, W.L. (1988). Human adrenodoxin: Cloning of three cDNAs and cycloheximide enhancement in JEG-3 cells. J. Biol. Chem. 263, 3240-3244, corrected
p. 11016.
SANGER, F., NICKLEN, S., and COULSON, A.R. (1977). DNA sequencing with chain-terminating inhibitors. Proc. Nati. Acad. Sei. USA 74, 5463-5467. SCARPULLA, R.C, and WU, R. (1983). Nonallelic members of the cytochrome c multigene family of the rat may arise through
different messenger RNAs. Cell 32, 473-482. TANAKA, M., HANIU, M., YASUNOBU, K.T., and KIMURA, T. (1973). The amino acid sequence of bovine adrenodoxin. J. Biol. Chem. 248, 1141-1157. VANIN, E.F. (1985). Processed pseudogenes: Characteristics and evolution. Annu. Rev. Genet. 19, 253-272. VOUTILAINEN, R., PICADO-LEONARD, J., DIBLASIO, A.M., and MILLER, W.L. (1988). Hormonal and developmental regulation of adrenodoxin messenger ribonucleic acid in steroidogenic tissues. J. Clin. Endocrinol. Metab. 66, 383-388. WEINER, A.M., DEININGER, P.L., and EFSTRATIADIS, A. (1986). Nonviral retroposons: Genes, pseudogenes, and transposable elements generated by the reverse flow of genetic information. Annu. Rev. Biochem. 55, 631-661. WYMAN, A.R., and WHITE, R. (1980). A highly polymorphic locus in human DNA. Proc. Nati. Acad. Sei. USA 77, 67546758.
Address
reprint requests to: Dr. Bon-chu Chung
Institute
of Molecular Biology Academia Sínica
Nankang, Taipei, Taiwan, 11529 Republic of China Received for
publication December 13, 1989,
January 29,
1990.
and in revised form
Pp. 205-212
Structure, Sequence, Chromosomal Location, and Evolution of the Human Ferredoxin Gene CHI-YAO
CHANG,*'t
DU-AN
WU,* T.K.
MOHANDAS,Í
Family
and BON-CHU CHUNG*
ABSTRACT Ferredoxin is an iron-sulfur protein that serves as an electron transport intermediate for mitochondrial cytochromes P450 involved in steroid, vitamin D, and bile acid metabolism. We cloned and characterized the human ferredoxin gene family, which includes two expressed genes and two pseudogenes. Sequence analysis of this gene family revealed that it encodes only one protein product. The expressed genes were assigned to chromosome 11 and pseudogenes to chromosomes 20 and 21 by identifying single-copy probes from each gene segment and hybridizing them to DNA from rodent-human hybrid cells. The pseudogenes lacked ¡litrons and contained numerous mutations, including insertion, deletion, and substitution which rendered them inactive. They were 96% and 85% homologous to the expressed gene, yet they were only 78% homologous with each other. The intronless nature, higher diversity among themselves, and distinct chromosomal location of the pseudogenes suggests that they arose by independent, retroposon-mediated events.
INTRODUCTION
protein small, acidic, Ferredoxin transport intermediate mitochondrial cytochromes P450. Electrons is
functions
iron-sulfur
a
as an
electron
that for
are trans-
ferred from NADPH
flavin-containing protein
through (ferredoxin oxidoreductase) and ferredoxin to the terminal cytochrome P450 for oxidation/reduction reactions. Mitoa
chondrial P450s and their ferredoxin are found mainly in the steroidogenic tissues, including adrenal, ovary, testis, and placenta (Jefcoate, 1986). Small amounts of them are also found in the liver and kidney for bile acid and vitamin D synthesis (Bjorkhem et ai, 1980; Driscoll and Omdahl, 1986; Jefcoate, 1986). Because of its relative abundance, the adrenal ferredoxin, designated adrenodoxin, has been characterized in most detail (Kimura et ai, 1969; Tanaka et ai, 1973). It is synthesized as a precursor in which 60 amino acids of the signal peptide is later cleaved upon transport into the mitochondrial inner matrix to form a mature protein of 124 amino acids (Okamura et ai, 1985). A single [2Fe-2S] center is incorporated into the apoprotein to form the activity center (Coghlan and Vickery,
1989).
Human adrenodoxin and placental ferredoxin cDNAs share identical sequence, suggesting that they are the same (Mittal et ai, 1988). At least three polyadenylation sites in the adrenodoxin cDNA have been identified; all three are functional and result in mRNAs of 1.75, 1.4, and 0.9 kb (Okamura et ai, 1985; Picado-Leonard et ai, 1988). Bovine adrenodoxin mRNA has multiple precursor sequences associated with the same mature sequence (Okamura et al., 1987). The ferredoxin RNA is detected in many tissues and its production is stimulated by various tropic hormones via cyclic AMP as an intracellular intermediate in the steroidogenic tissues (Kramer et ai, 1984; Anderson and Mendelson, 1985; Okamura et ai, 1985; Voutilainen et ai, 1988). The effect of cyclic AMP stimulation on the adrenodoxin and cytochrome P450 gene expression is at the transcriptional level (John et ai, 1986). To delineate the transcriptional mechanism of the adrenodoxin gene, we cloned the human gene (Chang et ai, 1988). It contains four exons and three introns, spanning more than 20 kb. In this report, we characterized the entire human ferredoxin gene family, including two expressed genes and two processed pseudogenes. We also localized these genes to human chromosomes 11, 20, and 21.
"Institute of Molecular Biology, Academia Sinica, Nankang, Taipei, Taiwan, 11529 Republic of China. t Institute of Life Sciences, Tsing-Hwa University, Hsin-Chu, Taiwan, Republic of China. ÎDepartment of Pediatrics and Division of Medical Genetics, Harbor/UCLA Medical Center, Torrance, CA 90509.
205
CHANG ET AL.
206 The structure, divergence, and chromosomal location of the pseudogenes suggest that they might arise by indepen-
dent, retroposon-mediated
events.
MATERIALS AND METHODS
General procedures DNA
dures
purification, sequencing, and hybridization procedescribed previously (Sanger et al, 1977;
were
Wyman and White, 1980; Maniatis, 1982; Mohandas et al, 1986), except that for the hybridization of the humanrodent hybrid DNA, the blot was washed additionally at 65°C for 10 min. Primers were synthesized by Gene Assembler (Pharmacia), and purified by a mono Q column
(Pharmacia).
Cloning ferredoxin
genes
DNA digested separately with five different restriction enzymes shown in Fig. 1. In each digest there were more than three hybridizing bands. The Eco RI digest showed seven bands with sizes of 11, 9.2, 8.3, 5.5, 3.2, 2.9, and 0.86 kb (the 0.86-kb band could be seen in other gels not shown here). We have previously demonstrated that the 8.3-, 5.5-, 3.2-, and 2.9-kb bands belong to an expressed adrenodoxin gene (Chang et al., 1988). The rest of the bands could represent other ferredoxin genes in the same family. To clone the human ferredoxin gene family, we screened three genomic DNA libraries (Maniatis et al, 1978; Bookstein et al, 1988) constructed in two different vectors to ascertain all the genes in the family were isolated. Two probes corresponding to the 5' region and the entire length of the cDNA were used to isolate five X clones hi to h5. Clones hi to h5 each contained three to eight Eco RI fragments in addition to both arms, representing 10 to 18 kb of inserts (Fig. 2, A and C). Among them one or two fragments in each clone showed hybridization to the cDNA probe (Fig. 2, B and D). They were 8.3 and 2.9 kb in hi (lane 2), 0.86 kb in h2 (lane 1), 9.2 kb in h3 (lane 3), 11 kb in h4 (lane 5), and 5.5 and 3.2 kb in h5 (lane 4). These bands accounted for all seven Eco RI bands seen in the genomic blot (Fig. 1); therefore, our collection of the human ferredoxin genomic clones appeared to be com-
Three human genomic DNA libraries were used in this study. The first library contained human genomic DNA partially digested with Eco RI and cloned into vector Charon 4A. The second library contained DNA digested partially by Alu l-Hae III in the same vector (Maniatis et al., 1978). Using a 1.3-kb cDNA probe, clones hi and h3 were isolated from the first library, while clones h2 and h4 plete. were isolated from the second library. Clone h5 was isoThe coding regions in these clones were located and their lated from the third library constructed by ligating DNA were determined and compared with that of the sequences partially digested with Mbo I and cloned into vector X dash cDNA. Clones hi and h4 were highly homologous and (Bookstein et al., 1988). This library was screened with a their restriction maps were also very similar (Fig. 3). One 400-bp Eco Rl-Xba I fragment from the 5' end of the cDNA (Picado-Leonard et al, 1988) to obtain h5. Cloned DNAs were digested with Eco RI and hybridized with cDNA probe to identify coding regions which were then subcloned into pUC18 and analyzed further by digestion and hybridization to locate exons. The exon-containS m i a m ¡é S ing subfragments were subcloned into M13 for sequence determination.
23. 1
Identification of single-copy probes were hybridized with 32P-labeled total hugenomic DNA to locate unique sequences which do not show hybridization upon short exposure. These DNAs were labeled and hybridized to a genomic Southern blot to ascertain that only one band appeared. A 500-bp Pst l-Eco RI fragment upstream from h2, a 180-bp Hinc II fragment located in a 6-kb Ssp I fragment downstream from the pseudogene h3, and a 414-bp Eco Rl-Pvu II fragment (PR410, covering -333/+81) in h5 appeared to be unique in this test (shown as black bars in Fig. 3). The single-copy probe in hi was previously identified (Morel et al., 1988).
Cloned DNAs
man
RESULTS
Isolation of the human ferredoxin genes A 1.3-kb human adrenodoxin cDNA probe Leonard et al., 1988) was hybridized to human
(Picadogenomic
6. 7
•" 30% of analyzed cells; (+) not 11%-30%; (-) 3%-10%; t(9;x) present in 90%; # t(x;10). The two lines at the bottom designate the number of cell lines discordant for each chromosome. Note that h2 probe shows no discordances for chromosome 20 and that h3 probe shows no discordances for chromosome 21.
Note:
+
=
detected;
Presence of
=
*
=
=
-
=
=
211
FERREDOXIN GENE FAMILY
Table 2. Single Base Substitutions
within the
Coding Regions
of
Ferredoxin Pseudogenes
No. substitutions in codon
Pseudogene designation
Chromosome location
position -
1st
2nd
3rd
Total no. of substitutions*
Time of integration (Myr)b
Percent
divergence
h2
20
8
6
10
19
3
5
h3
21
18
12
18
39
10
15
¡»Values were calculated assuming half of the substitutions in the third codon position were accounted for active gene, half in the pseudogene. bCalculated assuming the substitution rate is 7 x 10"' substitutions per nucleotide site per year.
ai, 1988). Renal ferredoxin has similar optical, magnetic, and immunochemical properties to adrenodoxin, although one report suggested that they had minor differences (Maruya et ai, 1983). Now that we have identified only one protein sequence, there is no need to designate ferredoxin according to their tissue origins. The same ferredoxin exists in the adrenal, placenta, kidney, and liver for steroid, vitamin D3, and bile acid metabolism. Therefore, ferredoxin is the more appropriate name than adrenodoxin, renodoxin, or hepatoredoxin. Both pseudogenes h2 and h3 lack introns and possess the properties of fully processed mRNA (Vanin, 1985). Although h3 does not have poly(A) tracks at the 3' end, its sequence ends at the polyadenylation site (17 nucleotides
downstream from AATAAA and coincides with the 3' end of cDNA hAdx-7) (Picado-Leonard et ai, 1988). H3 is flanked by 7-bp direct repeats TCAGTGA (underlined in Fig. 5), which usually result from réintégration of the cDNA into the chromosome (Vanin, 1985; Weiner et ai, 1986). These suggest that ferredoxin psdueogenes might arise by a retroposon-like mechanism having mRNA as an intermediate before integration (Vanin, 1985; Weiner et
ai, 1986).
The number of substitution within the coding region of h2 and h3 is tabulated in Table 2. Since neither h2 nor h3 carries a promoter, it is assumed that upon insertion both genes immediately lost activity and started to diverge from the active gene by neutral mutation. The number of substitutions was counted with the assumption that half of the changes in the third codon position were due to evolutionary drift in the functional gene, and half to changes in the pseudogene following integration (Lee et ai, 1983; Scarpulla and Wu, 1983). For neutral mutations, the rate of divergence is about 0.7% per Myr (million years) (Perler et ai, 1980). Thus, h2 apparently diverged from the expressed gene about 5 Myr ago. H3 diverged even more; it started to lose function about 15 Myr ago. Both pseudogenes appeared very recently, long after man and cattle diverged, about 85 Myr ago (Miller et ai, 1981). We located pseudogene h2 on chromosome 20 and h3 on chromosome 21. Different chromosomal locations of these genes support the notion that réintégration of the pro-
by evolutionary drift in the
cessed gene is a random event; therefore, they landed at different chromosomes by chance. These two pseudogenes are more homologous to the expressed gene (96% and 85%) than between themselves (only 78% homology). It is expected that they should maintain higher homology to their parent gene, which evolved more slowly due to positive selection pressure on an active gene (Miyata et ai, 1980). Homology between the pseudogenes, however, was lower, possibly because they evolved independently and there was no selective pressure to maintain their homology. Thus, the degree of divergence and chromosomal locations of h2 and h3 indicate that they might have arisen by inde-
pendent transposition
events.
ACKNOWLEDGMENTS We would like to thank Drs. Wen-Hwa Lee and James Shen for providing the human genomic DNA libraries and helpful discussion. D.-A.W. is a predoctoral student from the National Defense Medical Center, Republic of China. This work was supported by National Science Council, #NSC-78-0203-B001-ll, and Academia Sinica, Republic of China.
REFERENCES ANDERSON, CM., and MENDELSON, C.R. (1985). Regulation of steroidogenesis in rat Leydig cells in culture: Effect of human chorionic gonadotropin and dibutyryl cyclic AMP on the synthesis of cholesterol side chain cleavage cytochrome P-450 and adrenodoxin. Arch. Biochem. Biophys. 238, 378387. BJORKHEM, I., HOLMBERG, I., OFTEBRO, H., and PEDERSEN, J.I. (1980). Properties of a reconstituted vitamin D3 25-hydroxylase from rat liver mitochondria. J. Biol. Chem. 255, 5244-5249.
BOOKSTEIN, R., LEE, E.Y.-H.P., TO, H., YOUNG, L.-J., SERY, T.W., HAYES, R.C., FRIEDMANN, T., and LEE, W.-H. (1988). Human retinoblastoma susceptibility gene: Ge-
212
CHANG ET AL.
nomic organization and analysis of heterozygous intragenic deletion mutants. Proc. Nati. Acad. Sei. USA 85, 2210-2214. CHANG, C.-Y., WU, D.-A., LAI, C.-C, MILLER, W.L., and CHUNG, B.-C. (1988). Cloning and structure of the human adrenodoxin gene. DNA 7, 609-615.
CHASHCHIN, V.L., LAPKO, V.N., ADAMOVICH, T.B., KIRILLOVA, N.M., LAPKO, A.G., and AKHREM, A.A. (1986). The primary structure of hepatoredoxin from bovine liver mitochondria. Bioorg. Khim. 12, 1286-1289. COGHLAN, V.M., and VICKERY, L.E. (1989). Expression of human ferredoxin and assembly of the [2Fe-2S] center in Escherichia coli. Proc. Nati. Acad. Sei. USA 86, 835-839. DRISCOLL, W.J., and OMDAHL, J.L. (1986). Kidney and adrenal mitochondria contain two forms of NADPH-adrenodoxin reductase-dependent iron-sulfur proteins. Isolation of the two porcine renal ferredoxins. J. Biol. Chem. 261, 4122-4125. JEFCOATE, C.R. (1986). In Cytochrome P-450. P.R. Ortiz de Montellano, Ed. (Plenum Press, New York) pp. 387-428.
JOHN, M.E., JOHN, M.C., BOGGARAM, V., SIMPSON, E.R., and WATERMAN, M.R. (1986). Transcriptional regulation of steroid hydroxylase genes by corticotropin. Proc. Nati. Acad. Sei. USA
83, 4715-4719. KIMURA, T., SUZUKI, K., PADMANABHAN, R., SAMEJIMA, T., TARUTANI, O., and UI, N. (1969). Studies on steroid hydroxylase. Molecular properties of adrenal iron-sulfur protein. Biochemistry 8, 4027-4031. KRAMER, R.E., RAINEY, W.E., FUNKENSTEIN, B., DEE, A., SIMPSON, E.R., and WATERMAN, M.R. (1984). Induction of synthesis of mitochondrial steroidogenic enzymes of bovine adrenocortical cells by analogs of cyclic AMP. J. Biol. Chem. 259, 707-713. LEE, M.G.-S., LEWIS, S.A., WILDE, CD., and COWAN, N.J. (1983). Evolutionary history of a multigene family: An expressed human 0-tubulin gene and three processed pseudogenes. Cell 33, 477-487. MANIATIS, T., HARDISON, R.C., LACY, E., LAUER, J., O'CONNELL, C, QUON, D., SIM, G.K., and EFSTRATIADIS, A. (1978). The isolation of structural genes from libraries of eucaryotic DNA. Cell 15, 687-701. MANIATIS, T., FRITSCH, E.F., and SAMBROOK, J. (1982). In Molecular Cloning: A Laboratory Manual. (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY). MARUYA, N., HIWATASHI, A., ICHIKAWA, Y„ and YAMANO, T. (1983). Purification and characterization of renal ferredoxin from bovine renal mitochondria. J. Biochem. 93, 1239-1247. MILLER, W.L., COIT, D., BAXTER, J.D., and MARTIAL, J.A. (1981). Cloning of bovine prolactin cDNA and evolutionary implication of its sequence. DNA 1, 37-50. MITTAL, S., ZHU, Y.-Z., asnd VICKERY, L.E. (1988). Molecular cloning and sequence analysis of human placental ferredoxin. Arch. Biochem. Biophys. 264, 383-391. MIYATA, T., YASUNAGA, T., and NISHIDA, T. (1980). Nucleotide sequence divergence and functional constraint in mRNA evolution. Proc. Nati. Acad. Sei. USA 77, 7328-7332.
MOHANDAS, T.K., HEINZMANN, C, SPARKES, R.S., WASMUTH, J., EDWARDS, P., and LUSIS, A.J. (1986). Assignment of human 3-hydroxy-3-methylglutaryl coenzyme A reducíase gene to the ql3 q23 region of chromosome 5. Somat. Cell Mol. Genet. 12, 89-94. —
MOREL, Y, PICADO-LEONARD, J., WU, D.-A., CHANG, C.-Y., MOHANDAS, T.K., CHUNG, B.-C, and MILLER, W.L. (1988). Assignment of the functional gene for human adrenodoxin to chromosome llql3 —qter and of adrenodoxin
pseudogenes to chromosome 20cen ql3.1. Am. J. Hum. Genet. 43, 52-59. OKAMURA, T., JOHN, M.E., ZUBER, M.X., SIMPSON, E.R., and WATERMAN, M.R. (1985). Molecular cloning and —
amino acid sequence of the precursor form of bovine adrenodoxin: Evidence for a previously unidentified COOH-terminal peptide. Proc. Nati. Acad. Sei. USA 82, 5705-5709. OKAMURA, T., KAGIMOTO, M., SIMPSON, E.R., and WATERMAN, M.R. (1987). Multiple species of bovine adrenodoxin mRNA. Occurrence of two different mitochondrial precursor sequences associated with the same mature sequence. J. Biol. Chem. 262, 10335-10338. PERLER, F., EFSTRATIADIS, A., LOMEDICO, P., GILBERT, W., KOLODNER, R., and DODGSON, J. (1980). The evolution of genes: The chicken preproinsulin gene. Cell 20, 555-566.
PICADO-LEONARD, J., VOUTILAINEN, R., KAO, L.-C, CHUNG, B.-C, STRAUSS III, J.F., and MILLER, W.L. (1988). Human adrenodoxin: Cloning of three cDNAs and cycloheximide enhancement in JEG-3 cells. J. Biol. Chem. 263, 3240-3244, corrected
p. 11016.
SANGER, F., NICKLEN, S., and COULSON, A.R. (1977). DNA sequencing with chain-terminating inhibitors. Proc. Nati. Acad. Sei. USA 74, 5463-5467. SCARPULLA, R.C, and WU, R. (1983). Nonallelic members of the cytochrome c multigene family of the rat may arise through
different messenger RNAs. Cell 32, 473-482. TANAKA, M., HANIU, M., YASUNOBU, K.T., and KIMURA, T. (1973). The amino acid sequence of bovine adrenodoxin. J. Biol. Chem. 248, 1141-1157. VANIN, E.F. (1985). Processed pseudogenes: Characteristics and evolution. Annu. Rev. Genet. 19, 253-272. VOUTILAINEN, R., PICADO-LEONARD, J., DIBLASIO, A.M., and MILLER, W.L. (1988). Hormonal and developmental regulation of adrenodoxin messenger ribonucleic acid in steroidogenic tissues. J. Clin. Endocrinol. Metab. 66, 383-388. WEINER, A.M., DEININGER, P.L., and EFSTRATIADIS, A. (1986). Nonviral retroposons: Genes, pseudogenes, and transposable elements generated by the reverse flow of genetic information. Annu. Rev. Biochem. 55, 631-661. WYMAN, A.R., and WHITE, R. (1980). A highly polymorphic locus in human DNA. Proc. Nati. Acad. Sei. USA 77, 67546758.
Address
reprint requests to: Dr. Bon-chu Chung
Institute
of Molecular Biology Academia Sínica
Nankang, Taipei, Taiwan, 11529 Republic of China Received for
publication December 13, 1989,
January 29,
1990.
and in revised form
