- Email: [email protected]
Recombinant cytochrome c peroxidase folds properly without conformational “annealing”
Vol.
176,
May
15, 1991
No.
3, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS Pages
1469-1472
RECOMBINANT CYTOCHROME c PEROXIDASE FOLDS PROPERLY WITHOUT CONFORMATIONAL “ANNEALING” Juan C. Ferrer, Mark Ring, and A. Grant Mauk Department of Biochemistry, University of British Columbia, Vancouver, British Columbia V6T 1W5 Canada Received
March
20,
1991
Recombinant cytochrome c peroxidase isolated from Esclzerichiu cd has recently been reported to exhibit an abnormal electronic absorption spectrum that is converted to the normal spectrum after conformational “annealing” of the recombinant enzyme by passage over a cytochrome c affinity column. The current report provides evidence that the abnormal spectrum observed in somepreparations of recombinant cytochrome c peroxidase arises from the presence of contaminant, damaged forms cytochrome c peroxidase with altered spectra. Removal of these contaminant forms produces a major cytochrome c peroxidase fraction with a normal spectrum. We conclude that elution of recombinant cytochrome c peroxidase over a cytochrome c affinity column does not produce normal enzyme through conformational “annealing” but that it produces purified enzyme through removal of contaminants. 0 1991 Academic Press, Inc.
A recent report in this journal communicated experimental results concerning the purification of recombinant yeast cytochrome c peroxidase (CcP) from Eschericlzia cob that led the authors to conclude that the recombinant CcP does not fold in the same fashion as CcP prepared from yeast (Saccharomyces cerevisiue)(l).
This conclusion arose from
observation of a perturbation in the electronic absorption spectrum of recombinant CcP that was found to be corrected after passageof the recombinant enzyme over a cytochrome c affinity column. The salutary effect of the affinity column was attributed to a conformational “annealing” of incorrectly folded recombinant enzyme that resulted from interaction with one of its substrates, cytochrome c, during chromatography (1). In related work involving isolation of cytochrome c peroxidase from E. cd, we have observed some preparations of the recombinant enzyme that exhibit an aberrant electronic absorption spectrum similar to the spectrum reported recently (1). In this communication, we demonstrate that this abnormal spectrum results from contamination of native CcP with damaged peroxidase that has a suitably modified electronic absorption spectrum. The damaged, contaminant enzyme is readily removed by anion exchange chromatography to produce native CcP exhibiting the spectrum expected for high-spin, five coordinate Fe(III)-
1469
All
Copyright 0 1991 rights of reproduction
0006-291X/91 $1.50 by Academic Press, Inc. in any form reserved.
Vol.
176,
CCP.
No.
We
conclude
cytochrome
BIOCHEMICAL
3, 1991
that
elution
AND
BIOPHYSICAL
of recombinant
RESEARCH
cytochrome
COMMUNICATIONS
c peroxidase
over
a
c affinity column does not produce the native enzyme through conformational
“annealing”
but through removal of contaminant, MATERIALS
modified forms of CcP.
AND METHODS
Cytochrome c peroxidase was expressed from the cloned gene (2,3) in E. coli using the T7 expression system (4). Enzyme purification was conducted (4) essentially as described by Fishel et al. (5). In some preparations, the Soret band of the recrystallized ferric enzyme was found to occur at 409.5 nm rather than at 408 nm as reported for the enzyme isolated from S. cerevtiiue (6,7). CcP with this abnormal spectrum was further purified by FPLC ion exchange chromatography over a Mono-Q column (Pharmacia). Electronic spectra were recorded with a Cary Model 219 spectrophotometer interfaced to a Zenith Model Z-248 microcomputer (On-Line-Instrument-Systems, Jefferson, GA) or with a Shimadzu Model 250 spectrophotometer. RESULTS
The elution profile obtained by FPLC anion exchange chromatography of CCP exhibiting the aberrant electronic absorption spectrum is shown in Figure 1.
The
heterogeneity observed in this chromatogram is remarkable considering that the enzyme preparation from which it is produced has been previously treated by preparative anion exchange chromatography on DEAE-Sepharose,
gel filtration
over Sephadex G-75
(Pharmacia), and crystallization. The spectrum of the contaminant eluting with a retention time of cu. 7 minutes is shown in Figure 2 (thin line). In this preparation, the contaminant
120 100 80 60
0.08
40 20
0 5 10 15 20 RetentionTime@in)
0
240
360
480
600
720
Wavelength(nm)
Figure 1. The chromatogramof recrystallizedCCPproducedwith an FPLC system fitted with a Mono-Q (Pharmacia) anion exchange column (monitored at 280 nm). The protein fractions were resolved with a gradient formed by mixing 20 mM 3-(N-morpholino)propanesulfonic acid (MOPS) (pH 6.0) with 20 mM MOPS containing 0.2 M citrate (pH6.0). Fieure 2, Electronic absorption spectra of components resolved from a preparation of crystalline recombinant CcP (0.1 M 2-(N-morpholino)-ethanesulfonic acid (MES), pH 7, 25-C): recombinant CCP after elution over the Mono-Q column (bold line); the contaminant fraction eluting with a retention time of cu. 7 minutes in Figure 1 (thin line).
1470
Vol.
176,
No.
3, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Table 1, Comparisonof absorptionratios calculatedfrom the spectrumof recombinant
Fe(III)CcP after purification by FPLC and from the spectrumof Fe(III)CcP isolated from yeast (7) Ratio
AdAm AxdAm bdb~ ‘%a&%47 4izdA.w
RecombinantCCP’
CkP from veast”
1.28
1.21
1.52
1.51
8.71
8.68
3.28
3.35
0.76
0.73
’ 0.1 M 2-(N-morpholino)-ethanesulfonic acid (MES), pH 7.0. ’ 0.1 M potassiumphosphatebuffer, pH 7.0 (7). has the spectrum of the six-coordinate, low-spin form of CCP, though the spectroscopic characteristics of the contaminant can vary with the history of the sample. The major enzyme component seen in Figure 1 has the spectrum reported for native, resting Fe(III)CkP (Figure 1, bold line).
Absorption maxima and relevant absorption ratios determined
for the major (native) component shown in Figure 1 compare favorably with those reported (7) for native CcP isolated from yeast (Table 1) The contaminant material eluting with a retention time of ca. 7 min. (Figure 1) exhibits a Soret maximum at 412.5 nm. This spectrum did not change after mixing the contaminant with yeast iso-1-cytochrome c in potassium phosphate buffer (50 mM, pH 6) for one hour. The FPLC retention time of the contaminant material treated in this manner is identical to its retention prior to treatment. SDS polyacrylamide gel electrophoresis of this contaminant showed it to possessa major component with electrophoretic mobility identical to that of CkP (data not shown) and a number of minor components. These additional minor components account for the relatively greater absorbance of this material at 280 nm (Figure 2, thin line). The aberrant spectrum observed for CcP prior to FPLC separation can be reproduced by mixing FPLC-purified, native CcP with the isolated contaminant. DISCUSSION
The results presented here demonstrate clearly that the changes in the electronic spectrum of recombinant CcP previously attributed
(1) to formation of the native
conformation of the enzyme through binding to a cytochrome c affinity column can be explained simply on the basisof removal of contaminant, damaged forms of the enzyme that co-purify and co-crystallize with the native enzyme in somepreparations. These authors also report (1) that binding of cytochrome c to improperly folded CcP in solution causes “annealing” asobserved by electronic difference spectroscopy usinga technique first reported by Erman and Vitello (8). Based on our previous use of this technique (e.g., ref. 9), we 1471
Vol.
176,
No.
BIOCHEMICAL
3, 1991
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
explain this result by suggesting that loss of small amounts of cytochrome
c from solution
on pipet or cuvette surfaces during mixing of the two protein solutions can readily produce spectroscopic
changes identical to those mentioned
by these authors (1).
Aside from our current findings, there is extensive evidence from the work of Kraut and co-workers
CcP isolated from E. coli folds in a fashion
(e.g., 5,lO) that recombinant
identical to that of CcP isolated from bakers’ chaperone structure
proteins
or conformational
of recombinant
yeast without
“annealing.”
the known
assistance of yeast
Most notably, the three-dimensional
CcP has been determined
by X-ray
diffraction
techniques
and
found to be virtually identical to that of the protein prepared from yeast (10). We conclude from this work
and from our current
explained by the presence recombinant
of contaminant
that the previous
of non-recombinant
results
(1) can be
forms of CcP present in some preparations
enzyme and that there is no indication
assume the conformation
that recombinant
of
CcP is unable to
enzyme isolated from yeast without
assistance.
This work was supported by MRC of Canada Operating Grant MT-7182
Acknowledgments:
(to A.G.M.).
observations
J.C.F. is the recipient of a postdoctoral fellowship from the Ministerio de
Education y Ciencia of Spain. We thank Dr. David Goodin for the T7 expression system used for production of recombinant CcP in this work.
REFERENCES 1.
2. 3. 4. 5.
Hake, R., McLendon, G., and Corin A. F. (1990) Biochem. Biophys. Res. Common. 172, 1157-1162. Goltz, S., Kaput, J., and Blobel, G. (1982) J. Biol. C/zem. 257, 11186-11190. Kaput, J., Goltz, S., and Blobel, G. (1982) J. Biol. Chem. 257, 15054-15058. Goodin, D.B., Davidson, M.G., and Roe, J.A., in preparation. Fishel, L. A., Villafranca, J. E., Mauro, J. M., and Kraut, J. (1987) Biochemistry 26,
351-360. 6. 7. 8. 9. 10.
Yonetani, T., and Anni, A. (1987) J. Biol. Chem. 262, 9547-9554. Vitello, L. B., Huang, M., and Erman, J. E. (1990) Biochemistry 29, 4283-4288. Erman, J. E., and Vitello, L. B. (1980) J. Biol. Chem. 255, 6224-6227. Mauk, M. R., Reid, L. S., and Mauk, A. G. (1982) Biochemistry 21, 1843-1846. Wang, J., Mauro, J. M., Edwards, S. L., Oatley, S. J., Fishel, L. A., Ashford, V. A., Xuong, N-h., and Kraut, J. (1990) Biochemistry 29, 7160-7173.
1472
176,
May
15, 1991
No.
3, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS Pages
1469-1472
RECOMBINANT CYTOCHROME c PEROXIDASE FOLDS PROPERLY WITHOUT CONFORMATIONAL “ANNEALING” Juan C. Ferrer, Mark Ring, and A. Grant Mauk Department of Biochemistry, University of British Columbia, Vancouver, British Columbia V6T 1W5 Canada Received
March
20,
1991
Recombinant cytochrome c peroxidase isolated from Esclzerichiu cd has recently been reported to exhibit an abnormal electronic absorption spectrum that is converted to the normal spectrum after conformational “annealing” of the recombinant enzyme by passage over a cytochrome c affinity column. The current report provides evidence that the abnormal spectrum observed in somepreparations of recombinant cytochrome c peroxidase arises from the presence of contaminant, damaged forms cytochrome c peroxidase with altered spectra. Removal of these contaminant forms produces a major cytochrome c peroxidase fraction with a normal spectrum. We conclude that elution of recombinant cytochrome c peroxidase over a cytochrome c affinity column does not produce normal enzyme through conformational “annealing” but that it produces purified enzyme through removal of contaminants. 0 1991 Academic Press, Inc.
A recent report in this journal communicated experimental results concerning the purification of recombinant yeast cytochrome c peroxidase (CcP) from Eschericlzia cob that led the authors to conclude that the recombinant CcP does not fold in the same fashion as CcP prepared from yeast (Saccharomyces cerevisiue)(l).
This conclusion arose from
observation of a perturbation in the electronic absorption spectrum of recombinant CcP that was found to be corrected after passageof the recombinant enzyme over a cytochrome c affinity column. The salutary effect of the affinity column was attributed to a conformational “annealing” of incorrectly folded recombinant enzyme that resulted from interaction with one of its substrates, cytochrome c, during chromatography (1). In related work involving isolation of cytochrome c peroxidase from E. cd, we have observed some preparations of the recombinant enzyme that exhibit an aberrant electronic absorption spectrum similar to the spectrum reported recently (1). In this communication, we demonstrate that this abnormal spectrum results from contamination of native CcP with damaged peroxidase that has a suitably modified electronic absorption spectrum. The damaged, contaminant enzyme is readily removed by anion exchange chromatography to produce native CcP exhibiting the spectrum expected for high-spin, five coordinate Fe(III)-
1469
All
Copyright 0 1991 rights of reproduction
0006-291X/91 $1.50 by Academic Press, Inc. in any form reserved.
Vol.
176,
CCP.
No.
We
conclude
cytochrome
BIOCHEMICAL
3, 1991
that
elution
AND
BIOPHYSICAL
of recombinant
RESEARCH
cytochrome
COMMUNICATIONS
c peroxidase
over
a
c affinity column does not produce the native enzyme through conformational
“annealing”
but through removal of contaminant, MATERIALS
modified forms of CcP.
AND METHODS
Cytochrome c peroxidase was expressed from the cloned gene (2,3) in E. coli using the T7 expression system (4). Enzyme purification was conducted (4) essentially as described by Fishel et al. (5). In some preparations, the Soret band of the recrystallized ferric enzyme was found to occur at 409.5 nm rather than at 408 nm as reported for the enzyme isolated from S. cerevtiiue (6,7). CcP with this abnormal spectrum was further purified by FPLC ion exchange chromatography over a Mono-Q column (Pharmacia). Electronic spectra were recorded with a Cary Model 219 spectrophotometer interfaced to a Zenith Model Z-248 microcomputer (On-Line-Instrument-Systems, Jefferson, GA) or with a Shimadzu Model 250 spectrophotometer. RESULTS
The elution profile obtained by FPLC anion exchange chromatography of CCP exhibiting the aberrant electronic absorption spectrum is shown in Figure 1.
The
heterogeneity observed in this chromatogram is remarkable considering that the enzyme preparation from which it is produced has been previously treated by preparative anion exchange chromatography on DEAE-Sepharose,
gel filtration
over Sephadex G-75
(Pharmacia), and crystallization. The spectrum of the contaminant eluting with a retention time of cu. 7 minutes is shown in Figure 2 (thin line). In this preparation, the contaminant
120 100 80 60
0.08
40 20
0 5 10 15 20 RetentionTime@in)
0
240
360
480
600
720
Wavelength(nm)
Figure 1. The chromatogramof recrystallizedCCPproducedwith an FPLC system fitted with a Mono-Q (Pharmacia) anion exchange column (monitored at 280 nm). The protein fractions were resolved with a gradient formed by mixing 20 mM 3-(N-morpholino)propanesulfonic acid (MOPS) (pH 6.0) with 20 mM MOPS containing 0.2 M citrate (pH6.0). Fieure 2, Electronic absorption spectra of components resolved from a preparation of crystalline recombinant CcP (0.1 M 2-(N-morpholino)-ethanesulfonic acid (MES), pH 7, 25-C): recombinant CCP after elution over the Mono-Q column (bold line); the contaminant fraction eluting with a retention time of cu. 7 minutes in Figure 1 (thin line).
1470
Vol.
176,
No.
3, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Table 1, Comparisonof absorptionratios calculatedfrom the spectrumof recombinant
Fe(III)CcP after purification by FPLC and from the spectrumof Fe(III)CcP isolated from yeast (7) Ratio
AdAm AxdAm bdb~ ‘%a&%47 4izdA.w
RecombinantCCP’
CkP from veast”
1.28
1.21
1.52
1.51
8.71
8.68
3.28
3.35
0.76
0.73
’ 0.1 M 2-(N-morpholino)-ethanesulfonic acid (MES), pH 7.0. ’ 0.1 M potassiumphosphatebuffer, pH 7.0 (7). has the spectrum of the six-coordinate, low-spin form of CCP, though the spectroscopic characteristics of the contaminant can vary with the history of the sample. The major enzyme component seen in Figure 1 has the spectrum reported for native, resting Fe(III)CkP (Figure 1, bold line).
Absorption maxima and relevant absorption ratios determined
for the major (native) component shown in Figure 1 compare favorably with those reported (7) for native CcP isolated from yeast (Table 1) The contaminant material eluting with a retention time of ca. 7 min. (Figure 1) exhibits a Soret maximum at 412.5 nm. This spectrum did not change after mixing the contaminant with yeast iso-1-cytochrome c in potassium phosphate buffer (50 mM, pH 6) for one hour. The FPLC retention time of the contaminant material treated in this manner is identical to its retention prior to treatment. SDS polyacrylamide gel electrophoresis of this contaminant showed it to possessa major component with electrophoretic mobility identical to that of CkP (data not shown) and a number of minor components. These additional minor components account for the relatively greater absorbance of this material at 280 nm (Figure 2, thin line). The aberrant spectrum observed for CcP prior to FPLC separation can be reproduced by mixing FPLC-purified, native CcP with the isolated contaminant. DISCUSSION
The results presented here demonstrate clearly that the changes in the electronic spectrum of recombinant CcP previously attributed
(1) to formation of the native
conformation of the enzyme through binding to a cytochrome c affinity column can be explained simply on the basisof removal of contaminant, damaged forms of the enzyme that co-purify and co-crystallize with the native enzyme in somepreparations. These authors also report (1) that binding of cytochrome c to improperly folded CcP in solution causes “annealing” asobserved by electronic difference spectroscopy usinga technique first reported by Erman and Vitello (8). Based on our previous use of this technique (e.g., ref. 9), we 1471
Vol.
176,
No.
BIOCHEMICAL
3, 1991
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
explain this result by suggesting that loss of small amounts of cytochrome
c from solution
on pipet or cuvette surfaces during mixing of the two protein solutions can readily produce spectroscopic
changes identical to those mentioned
by these authors (1).
Aside from our current findings, there is extensive evidence from the work of Kraut and co-workers
CcP isolated from E. coli folds in a fashion
(e.g., 5,lO) that recombinant
identical to that of CcP isolated from bakers’ chaperone structure
proteins
or conformational
of recombinant
yeast without
“annealing.”
the known
assistance of yeast
Most notably, the three-dimensional
CcP has been determined
by X-ray
diffraction
techniques
and
found to be virtually identical to that of the protein prepared from yeast (10). We conclude from this work
and from our current
explained by the presence recombinant
of contaminant
that the previous
of non-recombinant
results
(1) can be
forms of CcP present in some preparations
enzyme and that there is no indication
assume the conformation
that recombinant
of
CcP is unable to
enzyme isolated from yeast without
assistance.
This work was supported by MRC of Canada Operating Grant MT-7182
Acknowledgments:
(to A.G.M.).
observations
J.C.F. is the recipient of a postdoctoral fellowship from the Ministerio de
Education y Ciencia of Spain. We thank Dr. David Goodin for the T7 expression system used for production of recombinant CcP in this work.
REFERENCES 1.
2. 3. 4. 5.
Hake, R., McLendon, G., and Corin A. F. (1990) Biochem. Biophys. Res. Common. 172, 1157-1162. Goltz, S., Kaput, J., and Blobel, G. (1982) J. Biol. C/zem. 257, 11186-11190. Kaput, J., Goltz, S., and Blobel, G. (1982) J. Biol. Chem. 257, 15054-15058. Goodin, D.B., Davidson, M.G., and Roe, J.A., in preparation. Fishel, L. A., Villafranca, J. E., Mauro, J. M., and Kraut, J. (1987) Biochemistry 26,
351-360. 6. 7. 8. 9. 10.
Yonetani, T., and Anni, A. (1987) J. Biol. Chem. 262, 9547-9554. Vitello, L. B., Huang, M., and Erman, J. E. (1990) Biochemistry 29, 4283-4288. Erman, J. E., and Vitello, L. B. (1980) J. Biol. Chem. 255, 6224-6227. Mauk, M. R., Reid, L. S., and Mauk, A. G. (1982) Biochemistry 21, 1843-1846. Wang, J., Mauro, J. M., Edwards, S. L., Oatley, S. J., Fishel, L. A., Ashford, V. A., Xuong, N-h., and Kraut, J. (1990) Biochemistry 29, 7160-7173.
1472