- Email: [email protected]
Expression of a Desulfovibrio Tetraheme Cytochrome c in Escherichia coli
Biochemical and Biophysical Research Communications 268, 688 – 691 (2000) doi:10.1006/bbrc.2000.2198, available online at http://www.idealibrary.com on
Expression of a Desulfovibrio Tetraheme Cytochrome c in Escherichia coli Patrı´cia N. da Costa, Cristiano Conte, and Lı´gia M. Saraiva 1 Instituto de Tecnologia Quı´mica e Biolo´gica, Universidade Nova de Lisboa, Rua da Quinta Grande, 6, Apartado 127, 2780-156 Oeiras, Portugal
Received December 20, 1999
A tetraheme cytochrome c was successfully overexpressed for the first time in Escherichia coli. Desulfovibrio desulfuricans ATCC 27774 tetraheme cytochrome c 3 was expressed in aerobically grown Escherichia coli cotransformed with Escherichia coli ccm gene cluster (Arslan et al. (1998) Bioch. Biophys. Res. Commun. 251, 744 –747). The analysis of the produced cytochrome showed that the signal peptide was correctly cleaved, the four heme groups were inserted and the electronic structure around the heme irons was conserved, i.e., the recombinant tetraheme cytochrome was identical to that isolated from the native source. Contradicting previous results which indicated that Escherichia coli was only capable of producing apocytochrome c 3 (Pollock et al. (1989) J. Gen. Microbiol. 135, 2319 –2328), the present work proves unequivocally that the holoform can also be obtained. © 2000 Academic Press
Key Words: multiheme cytochromes; overexpression; Desulfovibrio; E. coli.
The isolation of a diverse range of multiheme cytochromes c, acting either as electron transfer carriers or with enzymatic functions in the metabolic pathways of bacteria, such as reduction of nitrite and oxidation of hydroxylamine, showed that a variable number of heme groups can be present in a single polypeptide chain (1). These multiheme proteins and, in particular, tetraheme cytochromes have been found as part of the reaction center of photosynthetic bacteria (2), in the facultative anaerobe Shewanella frigidimarinus (3), in the microaerotolerant fermenting bacterium Malonomonas rubra (4) and in Nitrosomonas europea, an aerobic bacterium (5). Multiheme cytochromes also play an important role in anaerobic sulfate-reducing bacteria (SRB) since most of their cytochromes c contain more than one heme group (1). Among them, the tet1
To whom correspondence should be addressed. Fax: 351-14428766. E-mail: [email protected]. 0006-291X/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
raheme cytochrome c 3 is particularly well characterized due to several factors: (i) its presence in all Desulfovibrio (D.) species; (ii) its ability to work as a charge separation device which is indicative of a central role in the energy transduction mechanism of SRB (6); and (iii) the proposal that the architecture of its heme cores may constitute a structural model motif for cooperativity mechanisms (6, 7). Furthermore, several threedimensional structures indicate that the tetraheme structural arrangement is the building block of higher heme containing cytochromes c. For example, Desulfomicrobium norvegicum and D. gigas di-tetraheme cytochromes c 3 are dimers of tetraheme modules (8, 9) and D. desulfuricans ATCC 27774 nine-haem cytochrome contains two tetraheme c 3-like domains (10). The analysis of the sequence of the D. vulgaris sixteenheme high molecular weight cytochrome c also suggests that it is formed by three tetraheme domains, one triheme domain and one single heme motif (11). Thus, mechanistic studies of cytochrome c 3 are necessary to understand its functional behavior and that of higher heme content cytochromes c. In order to achieve those objectives, various tetraheme cytochromes c 3 were already cloned (12–15) but previous studies indicated that E. coli could not be used as an expression system. In fact, attempts to overexpress D. vulgaris tetraheme cytochrome c 3 in E. coli, in various conditions, were unsuccessful since only apocytochrome was obtained (16). The 3D-structures of Desulfovibrio tetraheme cytochromes c 3 show that the sixth histidine ligand to heme III is located between the fifth histidine ligand to heme I and its covalent binding motif (CXXCH) (17). This particular structural arrangement suggested the presence of a specific Desulfovibrio heme chaperone thus explaining the failure of heterologous production of holocytochrome c 3 in E. coli (18). In fact, when related hosts such as other Desulfovibrio strains were utilized, after development of technology for conjugative transfer from E. coli to Desulfovibrio strains (19, 20), the functional form of D. vulgaris cytochrome c 3 could be expressed in D. desulfuricans G200 (21). How-
688
Vol. 268, No. 3, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Heme staining after SDS-PAGE separation (A) and dithionite-reduced minus oxidized difference visible spectra (B) of native Dd cytochrome c 3 (1), soluble fraction from E. coli carrying pEC86 alone (2), and soluble fraction from E. coli carrying pEC86 and pT7Ddc 3 (3).
ever, this methodology requires special conditions and the overproduction of the recombinant proteins is not very efficient and is time consuming. It was observed that even the production of various heterologous monoheme cytochromes in E. coli failed in many cases and, when successful, it required a wide and different range of growth conditions (22). Among them, Bradyrhizobium japonicum cytochrome c 550 (23) and Paracoccus denitrificans cytochrome c 552 (24) could be expressed in anaerobically grown E. coli but in low yields. However, high levels of those holocytochromes c were produced in aerobically grown E. coli cotransformed with plasmid pEC86, that expresses the E. coli cytochrome c genes ccmABCDEFGH (ccm gene cluster) coding for several membrane proteins involved in the maturation of c-type cytochromes (24, 25). Recently, another monoheme cytochrome, cytochrome c⬙ from Methylophilus methylotrophus, could also be produced in E. coli using this expression system (26). Thus, expression of D. desulfuricans ATCC 27774 (Dd) cytochrome c 3 in aerobically grown E. coli cotransformed with pEC86 was attempted and here we report that under these conditions E. coli is able to produce a mature tetraheme cytochrome c 3. MATERIALS AND METHODS Production of Dd cytochrome c 3 in E. coli. Standard molecular biology protocols described in (27) were used in all DNA manipulations. Plasmid pEC86 containing the complete E. coli ccmA-H cluster (25) was a generous gift of Prof. L. Thony-Meyer. Using the Dd
cytochrome c 3 gene, previously cloned on a pBluescript SK(⫾) vector from Stratagene (our unpublished results), restriction sites were introduced by PCR and a 485-bp fragment was generated. This NdeI/EcoRI fragment was then cloned in pT7-7 (27) and transformed in E. coli DH5␣. The resulting plasmid, named pT7Ddc 3, was then introduced into E. coli BL21-Gold (DE3) (Promega) and cotransformed with plasmid pEC86. E. coli BL21-Gold (DE3) cells transformed with one plasmid only, pEC86 or pT7Ddc 3, were also prepared and used to test the presence of c-type cytochromes and expression of Dd cytochrome c 3, respectively. Cells containing one or both plasmids were aerobically grown, at 37°C, in Luria-Bertani medium supplemented with 1 mM FeSO 4. Antibiotics were added to media as required with the following concentrations: ampicillin (100 g/ml) and chloramphenicol (20 g/ml). When the culture reached a cell density of OD600nm ⫽ 0.5, 1 mM of isopropyl-thio--D-galactoside (IPTG) was added and after 3 hours cells were harvested by centrifugation. Purification and characterization of the recombinant Dd cytochrome c 3. The harvested cells were resuspended into 10 mM Tris/ HCl, pH 6.5, broken in a French Pressure cell (SLM/Aminco) at 8,000 psi and the membranes and cell debris removed by centrifugation. Heme staining on soluble fractions was performed as in (28). The purification was performed on a Pharmacia High-Load System at 4°C. The supernatant, containing the soluble proteins was dialyzed against 10 mM Tris/HCl pH 6.5 and loaded on a cationic exchange column (S-Sepharose), previously equilibrated with the same buffer. A gradient of 10 mM Tris/HCl-10 mM Tris/HCl, 1 M NaCl was then applied and Dd cytochrome c 3 was eluted at 250 mM NaCl. Further purification of Dd cytochrome c 3 was achieved by gel filtration using a Superdex 75 column equilibrated with 50 mM Tris/HCl, 100 mM NaCl pH 7.6. The degree of purity was confirmed by SDS-PAGE. The N-terminal sequence was determined with an Applied Biosystem 477A protein sequencer coupled to an Applied Biosystem 120A analyser. Ultraviolet/visible electronic spectra were recorded with a Shimadzu UV-360 spectrophotometer. EPR spectroscopy studies were carried out with a Bruker ESP 380 spectrometer equipped with a continuous flow helium cryostat (Oxford Instruments Co.).
689
Vol. 268, No. 3, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
onances characteristic of high-spin ferric hemes were not observed in the EPR spectrum indicating that all four hemes are six-coordinated. Altogether, these results unequivocally prove that the heterologous expression of Dd cytochrome c 3 in E. coli yield a four-heme containing cytochrome c 3 with no structural changes when compared with cytochrome c 3 isolated from the native source. CONCLUSION
FIG. 2. UV-visible spectra of oxidized and reduced forms of native Dd cytochrome c 3 (A) and recombinant Dd cytochrome c 3 overproduced in E. coli (B). Note: The ␣-region is multiplied 2 times relative to the Soret region both in spectra A and B; B spectrum is multiplied 4.5 times relative to A spectrum. Oxidized forms showing bands at 532 nm and 408.5 nm; reduced forms showing bands at 551 nm, 523 nm, and 418 nm.
RESULTS AND DISCUSSION The soluble fractions were analyzed spectroscopically and by heme staining (Fig. 1). Expression of c-type cytochromes was neither detected in cells transformed with pEC86 alone (Fig. 1-2) nor with pT7Ddc 3 (not shown). However, the expression of Dd tetraheme cytochrome c 3 was successful when cells were transformed with both the plasmid encoding the Dd cytochrome c 3 gene and plasmid pEC86 (Fig. 1-3). In order to investigate any structural differences, the recombinant Dd cytochrome c 3 was purified and analyzed. Cells grown as described above yield approximately 0.8 mg of protein per liter of culture, which is four times more than the amount obtained for tetraheme D. vulgaris cytochrome produced in the related host D. desulfuricans G200 (21). The Dd cytochrome c 3 gene encodes a signal peptide sequence (our unpublished results) that was cleaved during the production of the protein in E. coli as it is in D. desulfuricans, as judged by the identical N-terminal sequence of the first fifteen residues, when compared with the native cytochrome (29). The UV-visible spectra of the recombinant Dd cytochrome c 3 (oxidized and reduced form) show the same absorption maxima as the native protein (Fig. 2). The EPR spectrum of oxidized recombinant Dd cytochrome c 3 (Fig. 3) is identical to that of native cytochrome and the experimental spectrum could be fully reproduced by the sum of four subcomponents due to each one of the four hemes, with g-values identical to those reported for native cytochrome c 3 (30). Also, res-
The present work showing that Dd cytochrome c 3 can be produced in aerobically grown E. coli, cotransformed with the ccm gene cluster, has several implications. So far, non-mutated Desulfovibrio strains, i.e., strains in which the endogenous cytochrome c 3 gene was not deleted from the chromosome, have been used as hosts. Thus, the expression of both native and heterologous protein occurs and the host Desulfovibrio strain has to be chosen according to the differences in the physical properties of both proteins to allow their complete separation during purification. The present work shows that an efficient and easy system for production of Desulfovibrio tetraheme cytochromes can be used. Also, the fact that only in the presence of E. coli ccm gene cluster heterologous expression of the tetraheme cytochrome c 3 is achieved suggests that a similar set of genes is operative in Desulfovibrio species, as it is in many other bacteria (22). It may now be possible to produce in E. coli the various known tetraheme cyto-
FIG. 3. EPR spectrum of the oxidized recombinant Dd cytochrome c 3 (A) and theoretical simulation of the total spectrum (full line) obtained by the sum of the four subcomponents (B–E) using the g-values reported for the wild-type protein in (28). Temperature, 10K; microwave frequency, 9.64 GHz; microwave power, 2.4 mW. In the g ⬃ 2 region a resonance due to the EPR cavity and exogenous copper was removed.
690
Vol. 268, No. 3, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
chromes, independently from their native source, as well as other cytochromes with a higher heme content. Finally, this work proves that the expression of Desulfovibrio tetraheme cytochrome c 3 is not restricted to these species and contradicts previous believes that a specific system for heme insertion into cytochrome c 3 existent in Desulfovibrio strains was absent in E. coli (16). ACKNOWLEDGMENTS We are indebted to Professor L. Thony-Meyer for providing plasmid pEC86, Professor Anto´nio V. Xavier for helpful discussions, Professor Miguel Teixeira for his help with the EPR studies, Nicholas. J. Price for helpful suggestions, and M. Regalla for the N-terminal sequence. This work was supported by PRAXIS PCNA BIO/74/96 and E. C. Project FMRX-CT-98-0218 grants.
REFERENCES 1. Pereira, I. A. C., Teixeira, M., and Xavier, A. V. (1998) Structure and Bonding 91, 65– 89. 2. Nitschke, W., and Dracheva, S. (1995) in Anoxygenic Photosynthetic Bacteria (Blankenship, E. E., Madigan, M. T., and Bauer, C. E., Eds.), Chap. 36, pp. 775– 805, Kluwer Academic Press. 3. Tsapin, A. I., Nealson, K. H., Meyers, T., Cusanovich, M. A., Van Beummen, J., Crosby, L. D., Feinberg, B. A., and Zhang, C. (1996) J. Bacteriol. 178, 6386 – 6388. 4. Kolb, S., Seeliger, S., Springer, N., Ludwig, W., and Schink, B. (1998) System. Appl. Microbiol. 21, 340 –345. 5. Arciero, D. M., Balny, C., and Hooper, A. B. (1991) Biochemistry 30, 11466 –11472. 6. Louro, R. O., Catarino, T., LeGall, J., and Xavier, A. V. (1997) J. Biol. Inorg. Chem. 2, 488 – 491. 7. Saraiva, L. M., Salgueiro, C. A., da Costa, P. N., Messias, A. C., LeGall. J., van Dongen, W. M. A. M., and Xavier, A. V. (1998) Biochemistry 37, 12160 –12165. 8. Czjzek, M., Guerlesquin, F., Bruschi, M., and Haser, R. (1996) Structure 4, 395– 404. 9. Fraza˜o, C., Sieker, L., Sheldrick, G., Lamzin, V., LeGall, J., and Carrondo, M. A. (1999) J. Biol. Inorg. Chem. 4, 162–165. 10. Matias, P., Saraiva, L. M., Soares, C. M., Coelho, A. V. J., LeGall, J., and Carrondo, M. A. (1999) J. Biol. Inorg. Chem. 4, 478 – 494. 11. Pollock, W. B. R., Loutfi, M., Bruschi, M., Rapp-Giles, B. J., Wall, J. D., and Voordouw, G. (1991) J. Bacteriol. 173, 220 –228.
12. Voordouw, G., and Brenner, S. (1986) Eur. J. Biochem. 159, 347–351. 13. da Costa, P. N., Marujo, P. E., van Dongen, W. M. A. M., Arraiano, C., and Saraiva, L. M. (1999) GenBank: AJ250346. 14. Magro, V., Pieulle, L., Forget, N., Guigliarelli, B., Petillot, Y., and Hatchikian, E. C. (1997) GenBank: Y09717. 15. Magro, V., Pieulle, L., Forget, N., Guigliarelli, B., Petillot, Y., and Hatchikian, E. C. (1997) GenBank: Y09718. 16. Pollock, W. B. R., Chemerika, P. J., Forrest, M. E., Beatty, J. T., and Voordouw, G. (1989) J. Gen. Microbiol. 135, 2319 –2328. 17. Coutinho, I. B, and Xavier, A. V. (1994) Methods Enzymol. 243, 119 –140. 18. Pollock, W. B. R., and Voordouw, G. (1994) Biochimie 76, 554 – 560. 19. Van den Berg, W. A. M., Stokkermans, J. P. W. G., and Van Dongen, W. M. A. M. (1989) J. Biotechnol. 12, 173–184. 20. Van Dongen, W. M. A. M., Stokkermans, J. P. W. G., and Van den Berg, W. A. M. (1994) Methods Enzymol. 243, 319 –3307. 21. Voordouw, G., Pollock, W. B. R., Bruschi, M., Guerlesquin, F., Rapp-Giles, B. J., and Wall, J. D. (1990) J. Bacteriol. 172, 6122– 6126. 22. Thony-Meyer, L. (1997) Microb. Mol. Biol. Rev. 61, 337–376. 23. Thony-Meyer, L., Kunzler, P., and Hennecke, H. (1996) Eur. J. Biochem. 235, 754 –761. 24. Reincke, B., Thony-Meyer, L., Dannehl, C., Odenwald, A., Aidim, M., Witt, H., Ruterjans, H. and Ludwig, B. (1999) Biochim. Biophys. Acta 1411, 114 –120. 25. Arslan, E., Schulz, H., Zufferey, R., Kunzler, P., and ThonyMeyer, L. (1998) Biochem. Biophys. Res. Commun. 251, 744 – 747. 26. Price, N. J., Brennan, L., Faria, T. Q., Vijgenboom, E., Canters, G. W., Turner, D. L., and Santos, H. (1999) Submitted. 27. Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G. Smith, J. A., and Struhl, K. (1995) Current Protocols in Molecular Biology, Greene Publishing Associates and WileyInterscience, New York. 28. Goodhew, C. F., Brown, K. R., and Pettigrew, G. W. (1986) Biochim. Biophys. Acta 852, 288 –294. 29. Palma, P. N., Moura, I., LeGall, J., van Beeumen, J., Wampler, J. E., and Moura, J. J. G. (1994) Biochemistry 33, 6394 – 6407. 30. Morais, J., Palma, P. N., Fraza˜o, C., Caldeira, J., LeGall, J., Moura, I., Moura, J. J. G., and Carrondo, M. A. (1995) Biochemistry 34, 12830 –12841.
691
Expression of a Desulfovibrio Tetraheme Cytochrome c in Escherichia coli Patrı´cia N. da Costa, Cristiano Conte, and Lı´gia M. Saraiva 1 Instituto de Tecnologia Quı´mica e Biolo´gica, Universidade Nova de Lisboa, Rua da Quinta Grande, 6, Apartado 127, 2780-156 Oeiras, Portugal
Received December 20, 1999
A tetraheme cytochrome c was successfully overexpressed for the first time in Escherichia coli. Desulfovibrio desulfuricans ATCC 27774 tetraheme cytochrome c 3 was expressed in aerobically grown Escherichia coli cotransformed with Escherichia coli ccm gene cluster (Arslan et al. (1998) Bioch. Biophys. Res. Commun. 251, 744 –747). The analysis of the produced cytochrome showed that the signal peptide was correctly cleaved, the four heme groups were inserted and the electronic structure around the heme irons was conserved, i.e., the recombinant tetraheme cytochrome was identical to that isolated from the native source. Contradicting previous results which indicated that Escherichia coli was only capable of producing apocytochrome c 3 (Pollock et al. (1989) J. Gen. Microbiol. 135, 2319 –2328), the present work proves unequivocally that the holoform can also be obtained. © 2000 Academic Press
Key Words: multiheme cytochromes; overexpression; Desulfovibrio; E. coli.
The isolation of a diverse range of multiheme cytochromes c, acting either as electron transfer carriers or with enzymatic functions in the metabolic pathways of bacteria, such as reduction of nitrite and oxidation of hydroxylamine, showed that a variable number of heme groups can be present in a single polypeptide chain (1). These multiheme proteins and, in particular, tetraheme cytochromes have been found as part of the reaction center of photosynthetic bacteria (2), in the facultative anaerobe Shewanella frigidimarinus (3), in the microaerotolerant fermenting bacterium Malonomonas rubra (4) and in Nitrosomonas europea, an aerobic bacterium (5). Multiheme cytochromes also play an important role in anaerobic sulfate-reducing bacteria (SRB) since most of their cytochromes c contain more than one heme group (1). Among them, the tet1
To whom correspondence should be addressed. Fax: 351-14428766. E-mail: [email protected]. 0006-291X/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
raheme cytochrome c 3 is particularly well characterized due to several factors: (i) its presence in all Desulfovibrio (D.) species; (ii) its ability to work as a charge separation device which is indicative of a central role in the energy transduction mechanism of SRB (6); and (iii) the proposal that the architecture of its heme cores may constitute a structural model motif for cooperativity mechanisms (6, 7). Furthermore, several threedimensional structures indicate that the tetraheme structural arrangement is the building block of higher heme containing cytochromes c. For example, Desulfomicrobium norvegicum and D. gigas di-tetraheme cytochromes c 3 are dimers of tetraheme modules (8, 9) and D. desulfuricans ATCC 27774 nine-haem cytochrome contains two tetraheme c 3-like domains (10). The analysis of the sequence of the D. vulgaris sixteenheme high molecular weight cytochrome c also suggests that it is formed by three tetraheme domains, one triheme domain and one single heme motif (11). Thus, mechanistic studies of cytochrome c 3 are necessary to understand its functional behavior and that of higher heme content cytochromes c. In order to achieve those objectives, various tetraheme cytochromes c 3 were already cloned (12–15) but previous studies indicated that E. coli could not be used as an expression system. In fact, attempts to overexpress D. vulgaris tetraheme cytochrome c 3 in E. coli, in various conditions, were unsuccessful since only apocytochrome was obtained (16). The 3D-structures of Desulfovibrio tetraheme cytochromes c 3 show that the sixth histidine ligand to heme III is located between the fifth histidine ligand to heme I and its covalent binding motif (CXXCH) (17). This particular structural arrangement suggested the presence of a specific Desulfovibrio heme chaperone thus explaining the failure of heterologous production of holocytochrome c 3 in E. coli (18). In fact, when related hosts such as other Desulfovibrio strains were utilized, after development of technology for conjugative transfer from E. coli to Desulfovibrio strains (19, 20), the functional form of D. vulgaris cytochrome c 3 could be expressed in D. desulfuricans G200 (21). How-
688
Vol. 268, No. 3, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Heme staining after SDS-PAGE separation (A) and dithionite-reduced minus oxidized difference visible spectra (B) of native Dd cytochrome c 3 (1), soluble fraction from E. coli carrying pEC86 alone (2), and soluble fraction from E. coli carrying pEC86 and pT7Ddc 3 (3).
ever, this methodology requires special conditions and the overproduction of the recombinant proteins is not very efficient and is time consuming. It was observed that even the production of various heterologous monoheme cytochromes in E. coli failed in many cases and, when successful, it required a wide and different range of growth conditions (22). Among them, Bradyrhizobium japonicum cytochrome c 550 (23) and Paracoccus denitrificans cytochrome c 552 (24) could be expressed in anaerobically grown E. coli but in low yields. However, high levels of those holocytochromes c were produced in aerobically grown E. coli cotransformed with plasmid pEC86, that expresses the E. coli cytochrome c genes ccmABCDEFGH (ccm gene cluster) coding for several membrane proteins involved in the maturation of c-type cytochromes (24, 25). Recently, another monoheme cytochrome, cytochrome c⬙ from Methylophilus methylotrophus, could also be produced in E. coli using this expression system (26). Thus, expression of D. desulfuricans ATCC 27774 (Dd) cytochrome c 3 in aerobically grown E. coli cotransformed with pEC86 was attempted and here we report that under these conditions E. coli is able to produce a mature tetraheme cytochrome c 3. MATERIALS AND METHODS Production of Dd cytochrome c 3 in E. coli. Standard molecular biology protocols described in (27) were used in all DNA manipulations. Plasmid pEC86 containing the complete E. coli ccmA-H cluster (25) was a generous gift of Prof. L. Thony-Meyer. Using the Dd
cytochrome c 3 gene, previously cloned on a pBluescript SK(⫾) vector from Stratagene (our unpublished results), restriction sites were introduced by PCR and a 485-bp fragment was generated. This NdeI/EcoRI fragment was then cloned in pT7-7 (27) and transformed in E. coli DH5␣. The resulting plasmid, named pT7Ddc 3, was then introduced into E. coli BL21-Gold (DE3) (Promega) and cotransformed with plasmid pEC86. E. coli BL21-Gold (DE3) cells transformed with one plasmid only, pEC86 or pT7Ddc 3, were also prepared and used to test the presence of c-type cytochromes and expression of Dd cytochrome c 3, respectively. Cells containing one or both plasmids were aerobically grown, at 37°C, in Luria-Bertani medium supplemented with 1 mM FeSO 4. Antibiotics were added to media as required with the following concentrations: ampicillin (100 g/ml) and chloramphenicol (20 g/ml). When the culture reached a cell density of OD600nm ⫽ 0.5, 1 mM of isopropyl-thio--D-galactoside (IPTG) was added and after 3 hours cells were harvested by centrifugation. Purification and characterization of the recombinant Dd cytochrome c 3. The harvested cells were resuspended into 10 mM Tris/ HCl, pH 6.5, broken in a French Pressure cell (SLM/Aminco) at 8,000 psi and the membranes and cell debris removed by centrifugation. Heme staining on soluble fractions was performed as in (28). The purification was performed on a Pharmacia High-Load System at 4°C. The supernatant, containing the soluble proteins was dialyzed against 10 mM Tris/HCl pH 6.5 and loaded on a cationic exchange column (S-Sepharose), previously equilibrated with the same buffer. A gradient of 10 mM Tris/HCl-10 mM Tris/HCl, 1 M NaCl was then applied and Dd cytochrome c 3 was eluted at 250 mM NaCl. Further purification of Dd cytochrome c 3 was achieved by gel filtration using a Superdex 75 column equilibrated with 50 mM Tris/HCl, 100 mM NaCl pH 7.6. The degree of purity was confirmed by SDS-PAGE. The N-terminal sequence was determined with an Applied Biosystem 477A protein sequencer coupled to an Applied Biosystem 120A analyser. Ultraviolet/visible electronic spectra were recorded with a Shimadzu UV-360 spectrophotometer. EPR spectroscopy studies were carried out with a Bruker ESP 380 spectrometer equipped with a continuous flow helium cryostat (Oxford Instruments Co.).
689
Vol. 268, No. 3, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
onances characteristic of high-spin ferric hemes were not observed in the EPR spectrum indicating that all four hemes are six-coordinated. Altogether, these results unequivocally prove that the heterologous expression of Dd cytochrome c 3 in E. coli yield a four-heme containing cytochrome c 3 with no structural changes when compared with cytochrome c 3 isolated from the native source. CONCLUSION
FIG. 2. UV-visible spectra of oxidized and reduced forms of native Dd cytochrome c 3 (A) and recombinant Dd cytochrome c 3 overproduced in E. coli (B). Note: The ␣-region is multiplied 2 times relative to the Soret region both in spectra A and B; B spectrum is multiplied 4.5 times relative to A spectrum. Oxidized forms showing bands at 532 nm and 408.5 nm; reduced forms showing bands at 551 nm, 523 nm, and 418 nm.
RESULTS AND DISCUSSION The soluble fractions were analyzed spectroscopically and by heme staining (Fig. 1). Expression of c-type cytochromes was neither detected in cells transformed with pEC86 alone (Fig. 1-2) nor with pT7Ddc 3 (not shown). However, the expression of Dd tetraheme cytochrome c 3 was successful when cells were transformed with both the plasmid encoding the Dd cytochrome c 3 gene and plasmid pEC86 (Fig. 1-3). In order to investigate any structural differences, the recombinant Dd cytochrome c 3 was purified and analyzed. Cells grown as described above yield approximately 0.8 mg of protein per liter of culture, which is four times more than the amount obtained for tetraheme D. vulgaris cytochrome produced in the related host D. desulfuricans G200 (21). The Dd cytochrome c 3 gene encodes a signal peptide sequence (our unpublished results) that was cleaved during the production of the protein in E. coli as it is in D. desulfuricans, as judged by the identical N-terminal sequence of the first fifteen residues, when compared with the native cytochrome (29). The UV-visible spectra of the recombinant Dd cytochrome c 3 (oxidized and reduced form) show the same absorption maxima as the native protein (Fig. 2). The EPR spectrum of oxidized recombinant Dd cytochrome c 3 (Fig. 3) is identical to that of native cytochrome and the experimental spectrum could be fully reproduced by the sum of four subcomponents due to each one of the four hemes, with g-values identical to those reported for native cytochrome c 3 (30). Also, res-
The present work showing that Dd cytochrome c 3 can be produced in aerobically grown E. coli, cotransformed with the ccm gene cluster, has several implications. So far, non-mutated Desulfovibrio strains, i.e., strains in which the endogenous cytochrome c 3 gene was not deleted from the chromosome, have been used as hosts. Thus, the expression of both native and heterologous protein occurs and the host Desulfovibrio strain has to be chosen according to the differences in the physical properties of both proteins to allow their complete separation during purification. The present work shows that an efficient and easy system for production of Desulfovibrio tetraheme cytochromes can be used. Also, the fact that only in the presence of E. coli ccm gene cluster heterologous expression of the tetraheme cytochrome c 3 is achieved suggests that a similar set of genes is operative in Desulfovibrio species, as it is in many other bacteria (22). It may now be possible to produce in E. coli the various known tetraheme cyto-
FIG. 3. EPR spectrum of the oxidized recombinant Dd cytochrome c 3 (A) and theoretical simulation of the total spectrum (full line) obtained by the sum of the four subcomponents (B–E) using the g-values reported for the wild-type protein in (28). Temperature, 10K; microwave frequency, 9.64 GHz; microwave power, 2.4 mW. In the g ⬃ 2 region a resonance due to the EPR cavity and exogenous copper was removed.
690
Vol. 268, No. 3, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
chromes, independently from their native source, as well as other cytochromes with a higher heme content. Finally, this work proves that the expression of Desulfovibrio tetraheme cytochrome c 3 is not restricted to these species and contradicts previous believes that a specific system for heme insertion into cytochrome c 3 existent in Desulfovibrio strains was absent in E. coli (16). ACKNOWLEDGMENTS We are indebted to Professor L. Thony-Meyer for providing plasmid pEC86, Professor Anto´nio V. Xavier for helpful discussions, Professor Miguel Teixeira for his help with the EPR studies, Nicholas. J. Price for helpful suggestions, and M. Regalla for the N-terminal sequence. This work was supported by PRAXIS PCNA BIO/74/96 and E. C. Project FMRX-CT-98-0218 grants.
REFERENCES 1. Pereira, I. A. C., Teixeira, M., and Xavier, A. V. (1998) Structure and Bonding 91, 65– 89. 2. Nitschke, W., and Dracheva, S. (1995) in Anoxygenic Photosynthetic Bacteria (Blankenship, E. E., Madigan, M. T., and Bauer, C. E., Eds.), Chap. 36, pp. 775– 805, Kluwer Academic Press. 3. Tsapin, A. I., Nealson, K. H., Meyers, T., Cusanovich, M. A., Van Beummen, J., Crosby, L. D., Feinberg, B. A., and Zhang, C. (1996) J. Bacteriol. 178, 6386 – 6388. 4. Kolb, S., Seeliger, S., Springer, N., Ludwig, W., and Schink, B. (1998) System. Appl. Microbiol. 21, 340 –345. 5. Arciero, D. M., Balny, C., and Hooper, A. B. (1991) Biochemistry 30, 11466 –11472. 6. Louro, R. O., Catarino, T., LeGall, J., and Xavier, A. V. (1997) J. Biol. Inorg. Chem. 2, 488 – 491. 7. Saraiva, L. M., Salgueiro, C. A., da Costa, P. N., Messias, A. C., LeGall. J., van Dongen, W. M. A. M., and Xavier, A. V. (1998) Biochemistry 37, 12160 –12165. 8. Czjzek, M., Guerlesquin, F., Bruschi, M., and Haser, R. (1996) Structure 4, 395– 404. 9. Fraza˜o, C., Sieker, L., Sheldrick, G., Lamzin, V., LeGall, J., and Carrondo, M. A. (1999) J. Biol. Inorg. Chem. 4, 162–165. 10. Matias, P., Saraiva, L. M., Soares, C. M., Coelho, A. V. J., LeGall, J., and Carrondo, M. A. (1999) J. Biol. Inorg. Chem. 4, 478 – 494. 11. Pollock, W. B. R., Loutfi, M., Bruschi, M., Rapp-Giles, B. J., Wall, J. D., and Voordouw, G. (1991) J. Bacteriol. 173, 220 –228.
12. Voordouw, G., and Brenner, S. (1986) Eur. J. Biochem. 159, 347–351. 13. da Costa, P. N., Marujo, P. E., van Dongen, W. M. A. M., Arraiano, C., and Saraiva, L. M. (1999) GenBank: AJ250346. 14. Magro, V., Pieulle, L., Forget, N., Guigliarelli, B., Petillot, Y., and Hatchikian, E. C. (1997) GenBank: Y09717. 15. Magro, V., Pieulle, L., Forget, N., Guigliarelli, B., Petillot, Y., and Hatchikian, E. C. (1997) GenBank: Y09718. 16. Pollock, W. B. R., Chemerika, P. J., Forrest, M. E., Beatty, J. T., and Voordouw, G. (1989) J. Gen. Microbiol. 135, 2319 –2328. 17. Coutinho, I. B, and Xavier, A. V. (1994) Methods Enzymol. 243, 119 –140. 18. Pollock, W. B. R., and Voordouw, G. (1994) Biochimie 76, 554 – 560. 19. Van den Berg, W. A. M., Stokkermans, J. P. W. G., and Van Dongen, W. M. A. M. (1989) J. Biotechnol. 12, 173–184. 20. Van Dongen, W. M. A. M., Stokkermans, J. P. W. G., and Van den Berg, W. A. M. (1994) Methods Enzymol. 243, 319 –3307. 21. Voordouw, G., Pollock, W. B. R., Bruschi, M., Guerlesquin, F., Rapp-Giles, B. J., and Wall, J. D. (1990) J. Bacteriol. 172, 6122– 6126. 22. Thony-Meyer, L. (1997) Microb. Mol. Biol. Rev. 61, 337–376. 23. Thony-Meyer, L., Kunzler, P., and Hennecke, H. (1996) Eur. J. Biochem. 235, 754 –761. 24. Reincke, B., Thony-Meyer, L., Dannehl, C., Odenwald, A., Aidim, M., Witt, H., Ruterjans, H. and Ludwig, B. (1999) Biochim. Biophys. Acta 1411, 114 –120. 25. Arslan, E., Schulz, H., Zufferey, R., Kunzler, P., and ThonyMeyer, L. (1998) Biochem. Biophys. Res. Commun. 251, 744 – 747. 26. Price, N. J., Brennan, L., Faria, T. Q., Vijgenboom, E., Canters, G. W., Turner, D. L., and Santos, H. (1999) Submitted. 27. Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G. Smith, J. A., and Struhl, K. (1995) Current Protocols in Molecular Biology, Greene Publishing Associates and WileyInterscience, New York. 28. Goodhew, C. F., Brown, K. R., and Pettigrew, G. W. (1986) Biochim. Biophys. Acta 852, 288 –294. 29. Palma, P. N., Moura, I., LeGall, J., van Beeumen, J., Wampler, J. E., and Moura, J. J. G. (1994) Biochemistry 33, 6394 – 6407. 30. Morais, J., Palma, P. N., Fraza˜o, C., Caldeira, J., LeGall, J., Moura, I., Moura, J. J. G., and Carrondo, M. A. (1995) Biochemistry 34, 12830 –12841.
691