Tài liệu Báo cáo khoa học: Mitochondrial targeting of intact CYP2B1 and CYP2E1 and N-terminal truncated CYP1A1 proteins in Saccharomyces cerevisiae ) role of protein kinase A in the mitochondrial targeting of CYP2E1 pdf

Mitochondrial targeting of intact CYP2B1 and CYP2E1 and
N-terminal truncated CYP1A1 proteins in Saccharomyces
cerevisiae
)
role of protein kinase A in the mitochondrial
targeting of CYP2E1
Naresh B. V. Sepuri, Sanjay Yadav, Hindupur K. Anandatheerthavarada and Narayan G. Avadhani
Department of Animal Biology, School of Veterinary Medicine, University of Pennsylvania, Philadelphia, PA, USA
Cytochrome P450s (CYPs) belong to a superfamily of
heme-thiolate enzymes that catalyze the oxidation of
xenobiotic as well as endogenous compounds [1–3]. A
majority of the constitutively expressed and inducible
CYPs belonging to families 1–4 are primarily localized
in the endoplasmic reticulum (ER), hereafter referred to
as microsomes. However, there is increasing evidence
suggesting that some of the inducible CYPs are also
bimodally targeted to the mitochondrial compartment
[4–7]. Studies from our laboratory and others demon-
strated that b-naphthoflavone-inducible CYP1A1,
pyrazole-inducible CYP2E1, and phenobarbital-induci-
ble CYP2B1, known to be bona fide microsomal forms,
are also targeted to mitochondria [5,6,8–10]. These
Keywords
chimeric targeting signals; CYP2E1;
evolutionary conservations; mitochondrial
protein targeting; xenobiotic metabolism
Correspondence
N. G. Avadhani, Department of Animal
Biology, School of Veterinary Medicine,
University of Pennsylvania, 3800 Spruce
Street, Philadelphia, PA 19104, USA

Fax: +1 215 573 6651
Tel: +1 215 898 8819
E-mail:
(Received 30 April 2007, revised 6 July
2007, accepted 13 July 2007)
doi:10.1111/j.1742-4658.2007.05990.x
Previously we showed that intact rat cytochrome P450 2E1, cytochrome
P450 2B1 and truncated cytochrome P450 1A1 are targeted to mito-
chondria in rat tissues and COS cells. However, some reports suggest that
truncated cytochrome P450 2E1 is targeted to mitochondria. In this study,
we used a heterologous yeast system to ascertain the conservation of
targeting mechanisms and the nature of mitochondria-targeted proteins.
Mitochondrial integrity and purity were established using electron
microscopy, and treatment with digitonin and protease. Full-length cyto-
chrome P450 2E1 and cytochrome P450 2B1 were targeted both to micro-
somes and mitochondria, whereas truncated cytochrome P450 1A1 (+ 5
and + 33 ⁄ cytochrome P450 1A1) were targeted to mitochondria. Inability
to target intact cytochrome P450 1A1 was probably due to lack of cyto-
solic endoprotease activity in yeast cells. Mitochondrial targeting of cyto-
chrome P450 2E1 was severely impaired in protein kinase A-deficient cells.
Similarly, a phosphorylation site mutant cytochrome P450 2E1 (Ser129A)
was poorly targeted to the mitochondria, thus confirming the importance
of protein kinase A-mediated protein phosphorylation in mitochondrial
targeting. Mitochondria-targeted proteins were localized in the matrix
compartment peripherally associated with the inner membrane and their
ethoxyresorufin O-dealkylation, erythromycin N-demethylase, benzoxyres-
orufin O-dealkylation and nitrosodimethylamine N-demethylase activities
were fully supported by yeast mitochondrial ferredoxin and ferredoxin
reductase.
Abbreviations

BROD, benzoxyresorufin O-dealkylation; CCPO, cytochrome c peroxidase; CYP, cytochrome P450; DHFR, dihydrofolate reductase; DMPS,
dolichol mannose phosphate synthase; ERND, erythromycin N-demethylase; ER, endoplasmic reticulum; FDX, ferredoxin 1; FDXR, ferredoxin
reductase; NDMA, nitrosodimethylamine; NDMA-d, nitrosodimethylamine N-demethylase; PKA, protein kinase A; Put2, D1-pyrroline-5-
carboxylate dehydrogenase; TIM, translocase of inner membrane; TOM, translocase of outer membrane.
FEBS Journal 274 (2007) 4615–4630 ª 2007 University of Pennsylvania. Journal compilation ª 2007 FEBS 4615
studies led us to propose the concept of chimeric pro-
tein-targeting signals that drive the bimodal targeting of
the same primary translation product to more than one
subcellular compartment [10–12].
Protein targeting to the microsomes requires the co-
translational insertion of the newly synthesized protein
into the microsomal membrane, where the N-terminal
hydrophobic signal sequence of the protein interacts
with a signal recognition particle. This interaction sub-
sequently results in the association of the translational
complex with the microsomal membrane [13,14]. Thus,
the N-terminal hydrophobic sequences of CYPs are
important for their targeting to and retention in the
ER [15–17]. Protein translocation into the mitochon-
dria requires a cytosolic chaperone-mediated associa-
tion of precursor protein with peripheral translocase of
outer membrane (TOM) receptors (TOM20, TOM22
and TOM70), which enables the translocation of
proteins through the outer membrane and the inner
membrane channel-forming proteins, TOM40 and
translocase of inner membrane 23 (TIM23) [18–20].
Our studies defined two distinct mechanisms of acti-
vation of cryptic mitochondria-targeting signals at the
N-terminus of mammalian CYP proteins [4,10]. We
found that post-translational processing of CYP1A1

at the 4th and 32nd amino acid residues by a cytosolic
endoprotease is critical for the activation of cryptic
mitochondria-targeting signal at the 33–44 positions
of CYP1A1 [4,8,9]. This endoprotease was unable to
cleave the N-termini of CYP2E1 and CYP2B1 [6,12].
In the case of CYP2B1 and CYP2E1, uncleaved intact
proteins were targeted to the mitochondria in both
inducer-treated rat liver and transiently transfected
COS cells [6,10]. In both of these cases, protein
kinase A (PKA)-mediated phosphorylation at Ser128
or Ser129 [12] was necessary for the activation of a
cryptic mitochondria-targeting signal at positions
21–36 of the protein [6,12]. In contrast to our obser-
vations on the targeting of intact CYP2E1 to mito-
chondria, Neve & Ingelman-Sundberg [21] showed
that an N-terminally truncated CYP2E1 was targeted
to mitochondria in transiently transfected hepatoma
cells [22]. The nature and specificity of the endopro-
tease and the site of proteolytic cleavage remain
unknown. The same investigators were unable to see
any significant intramitochondrial localization of
intact or N-terminal truncated CYP2E1 in yeast cells
[23]. The precise nature of CYP proteins targeted to
mitochondria and the conservation of targeting mech-
anism is important for understanding the evolution
and regulation of bimodal targeting. Our primary
objective here was to address this important question
using rigorous approaches.
Yeast provides an ideal system for the heterologous
expression of genes to study both gene function and

metabolism. The protein translocation machineries of
both the mitochondria and microsomes are highly con-
served among mammals and yeast [24,25]. As protein
trafficking has been very well characterized in budding
yeast and is thought to involve a similar translocation
mechanism as that in mammalian cells, the yeast
expression system is well suited for the study of the
bimodal targeting mechanism described mostly in tran-
siently transfected mammalian cells. As targeting of
intact CYP2E1 and the requirement for PKA-mediated
phosphorylation for mitochondrial targeting are con-
tradicted by other studies [22,23,26], we sought cell
systems lacking specific PKA subunits to address this
important question. The availability of PKA gene dele-
tion yeast strains provided another advantage for the
present study.
We show here that mammalian CYPs are targeted
efficiently to both the microsomes and mitochondria in
yeast cells, depending on the nature of the chimeric
signals that they carry. In transformed yeast cells,
+33⁄ 1A1 was exclusively localized to the mito-
chondria, whereas + 5 ⁄ 1A1 was localized in both the
mitochondria and microsomes. Also, we found that full-
length CYP2E1 and CYP2B1 were targeted to the mito-
chondria as well as microsomes. By using PKA-deficient
cells, we further show the importance of PKA-mediated
phosphorylation in the mitochondrial targeting of
CYP2E1. Most importantly, substrate conversion by
mitochondria-targeted CYPs was fully supported by
yeast mitochondrial ferredoxin (FDX) + ferredoxin

reductase (FDXR), homologs of mammalian ferredoxin
and ferrodoxin reductase [27,28].
Results
Expression of intact and truncated rat CYPs in
Saccharomyces cerevisiae
The levels of expression of various CYP cDNAs were
measured by resolving whole cell extracts on SDS ⁄
PAGE and western blot analysis using CYP-specific
antibodies. As shown in Fig. 1A,B, the western blot
patterns of CYP proteins in cells transformed with
plasmids dihydrofolate reductase (DHFR)-1A1, intact
CYP1A1, + 5 ⁄ 1A1, + 33 ⁄ 1A1 and CYP2E1 were con-
sistent with their predicted molecular masses. Yeast
strain BY4741, transformed with plasmid CYP2B1
cDNA, showed extensive degradation of the CYP2B1
protein (Fig. 1B). This result was consistent with a pre-
vious report showing similar degradation of CYP2B1
in yeast cells [29]. Use of the protease-deficient strain
Mitochondrial targeting of rat CYPs in yeast cells N. B. V. Sepuri et al.
4616 FEBS Journal 274 (2007) 4615–4630 ª 2007 University of Pennsylvania. Journal compilation ª 2007 FEBS
pepD as suggested by Liao et al. [29] yielded more
intact CYP2B1 protein (Fig. 1B). Consistent with the
reported size of CYP2E1 protein, cells transformed
with CYP2E1 cDNA plasmid yielded a 52 kDa band.
This extract also yielded additional antibody-reactive
bands of about 28–32 kDa, which probably represent
degradation products.
Mitochondrial localization of N-terminal
truncated CYP1A1
The analysis of the microsomal fractions from yeast

strains expressing full-length CYP1A1, + 5 ⁄ 1A1 and
+ 331A1 showed significant levels of full-length
CYP1A1 protein, reduced levels of + 5 ⁄ 1A1 protein,
and vastly reduced levels of + 33 ⁄ 1A1 protein
(Fig. 2A, first four lanes). We also found nearly un-
detectable full-length CYP1A1 and clearly visible
+5⁄ 1A1 and + 33 ⁄ 1A1 in the mitochondrial fraction
(Fig. 2A, last four lanes). As expected, full-length
CYP1A1 and + 5 ⁄ 1A1 from the microsomal mem-
brane fraction were degraded by trypsin treatment
(Fig. 2A, first four lanes). This is consistent with the
model suggesting a transmembrane topology of CYPs
with a single N-terminal membrane anchor and most
of the remaining protein exposed to the cytosolic side
[15,17,30,31]. The intramitochondrial localization
of CYPs and their topologies were studied using a
combination of treatment with trypsin, treatment with
digitonin plus trypsin, and extraction with alkaline
Na
2
CO
3
.+5⁄ 1A1, + 33 ⁄ 1A1, and TIM23, which was
used as an internal control, were protected fully
against trypsin up to 100 lgÆmL
)1
, whereas full-length
CYP1A1 was completely digested (Fig. 2A, last four
lanes). These results suggest that full-length CYP1A1
is peripherally associated with the mitochondria. We

found that both + 5 ⁄ 1A1 and + 33 ⁄ 1A1 were resistant
to protease digestion even after selective removal of
the outer membrane by digitonin treatment (Fig. 2B).
Under these conditions, TIM23, with a significant pro-
portion of its sequence exposed outside the inner mem-
brane, facing the intermembrane space, was degraded
significantly. These results suggested that the imported
+5⁄ 1A1 and + 33 ⁄ 1A1 proteins were localized inside
of the inner membrane. Moreover, the imported
+5⁄ 1A1 and + 33 ⁄ 1A1 proteins were extractable with
alkaline Na
2
CO
3
as shown in Fig. 2C, suggesting that
both proteins are localized in the matrix compartment
in a membrane-extrinsic topology.
We found that DHFR-1A1 was peripherally associ-
ated with the mitochondria and microsomes, as it was
vulnerable to very low concentrations of trypsin, sug-
gesting that the intracellular distribution of CYPs was
not due to random insertion into the microsomal or
mitochondrial compartments (Fig. 2D). As expected,
TIM23 used as an internal control for the mitochon-
drial fraction was protected from protease, whereas
dolichol phosphate mannose synthase (DMPS) used as
internal control for the microsomes was not protected
from the externally added trypsin (Fig. 2D). In previ-
ous studies, we showed that the positively charged
amino acids at positions 34 and 39 were important for

targeting of + 5 ⁄ 1A1 and + 33 ⁄ 1A1 proteins to the
mitochondrial compartment. In keeping with these
observations, the results in Fig. 2E show that the asso-
ciation of a single mutant (R34D) or double mutants
(R34D and K39I) of + 331A1 with the mitochondrial
membrane was sensitive to protease treatment
(Fig. 2E). These results suggest that, as in the mamma-
lian cell system, the cryptic signal sequence at amino
acids 33–44 serves as a mitochondria-targeting signal
in the yeast system.
Mitochondrial localization of intact CYP2E1
in yeast cells
Because of the existing ambiguity in the literature on
the nature and extent of CYP2E1 import into mito-
chondria, we first established the relative purity of
mitochondrial preparations by biochemical and elec-
tron microscopy techniques. Figure 3A (top left panel)
+33/1A1
+5/1A1
DHFR1A1
1A1
35
47
62
81
kDa
AB
1234
2B1/BY4746
2E1

2B1/pep4Δ
'
25
35
47
62
81
kDa
56
7
16
25
35
47
62
81
kDa
16
Fig. 1. The levels of expression of CYP1A1, CYP2B1 and CYP2E1
proteins in yeast cells. Yeast strains BY4741 or PeP4D, trans-
formed with cDNA constructs, were grown to log phase, and cell
lysates were analyzed by SDS ⁄ PAGE and western blotting. (A)
Extracts from cells transformed with the indicated CYP1A1 con-
structs were probed with antibody to CYP1A1. (B) Extracts from
cells transformed with CYP2B1 (lanes 5 and 6) and CYP2E1 con-
struct were probed with either CYP2B1 antibody (left panel, lanes
5 and 6) or CYP2E1 antibody (right-most panel). All transformations
were done in the BY4741 strain, except for the middle panel of (B),
where the protease-deficient strain PeP4D was used. In each case,
50 lg of protein was used for immunoblot analysis.

N. B. V. Sepuri et al. Mitochondrial targeting of rat CYPs in yeast cells
FEBS Journal 274 (2007) 4615–4630 ª 2007 University of Pennsylvania. Journal compilation ª 2007 FEBS 4617
shows the transmission electron microscopy pattern of a
representative mitochondrial preparation. A representa-
tive field shows several well-defined mitochondrial parti-
cles with minor membrane contamination. As shown in
the insets, a large majority of mitochondrial prepara-
tions showed intact inner and outer membrane compo-
nents, confirming the structural integrity of mito-
chondrial isolates.
As shown in Fig. 3B, mitochondrial preparations
from cells transfected with plasmid pNS61(CYP2E1
cDNA) lacked significant levels of the microsomal
marker protein DMPS, and also the cytosolic protein
Trypsin (µg/mL)
Microsomes
1A1
25 5010025
A
C
DE
B
––50 100
TIM23
+33/1A1
TIM23
+5/1A1
TIM23
Digitonin
+Trypsin:

+–
Mitochondria
Mitochondria
1A1
TIM23
+33/1A1
TIM23
+5/1A1
TIM23
Alk. insol.
Alk. Sol.
Input
+5/1A1 mito
TIM23ab
1A1ab
+33/1A1 mito
Alk. insol.
Alk. Sol.
Input
+
TIM23
Micro.
Mito.
Trypsin:
DPMS
+––
DHFR/1A1
Micro.
Mito.
+33(R34D)

+33(R34D&K39I)
+33(R34D)
+33(R34D&K39I)
Trypsin:
+
+33/1A1
+33/1A1
–+–+–
+33/1A1
TIM23
Fig. 2. Mitochondrial targeting of truncated CYP1A1 in yeast cells. Mitochondrial and microsomal fractions of yeast cells expressing CYP1A1,
+5⁄ 1A1, DHFR-1A1 and + 33 ⁄ 1A1 were separated by SDS ⁄ PAGE and subjected to western blotting. Membrane topologies of mitochondria-
associated + 5 ⁄ 1A1, + 33 ⁄ 1A1, CYP1A1 (A, B, C), DHFR-1A1 (D) + 33 ⁄ 1A1, + 33 ⁄ 1A1(R34D) and + 331A1(R34D and K39I) (E) were
determined by protease treatment of microsomal and mitochondrial isolates before (A, D, E) or after (B) digitonin treatment. In (C), digitonin-
treated mitochondria were subjected to alkaline Na
2
CO
3
extraction. In (A), increasing concentrations of trypsin (0–100 lgÆmL
)1
) were used,
and in (B), (D) and (E), a fixed concentration (50 lgÆmg
)1
) of trypsin were used. Fifty micrograms of protein in each case was subjected to
immunoblot analysis. Stripped blots were redeveloped with antibodies to marker proteins, TIM23 (mitochondrial marker) or DMPS, a micro-
somal marker.
Mitochondrial targeting of rat CYPs in yeast cells N. B. V. Sepuri et al.
4618 FEBS Journal 274 (2007) 4615–4630 ª 2007 University of Pennsylvania. Journal compilation ª 2007 FEBS
3-phospho glycerate kinase (3-PGK), but contained
mitochondria-specific TIM23 protein. Additionally, both

the ER and mitochondrial fractions showed CYP2E1
antibody crossreactivity, although the latter fraction
showed 60% lower band intensity. The preparations
lacked significant levels of oligomycin-insensitive
NADPH cytochrome c reductase, a microsome-specific
marker enzyme (results not shown). Furthermore, the
mitochondrial preparations contained < 90% of total
cellular cytochrome c oxidase activity (results not
shown). The results in Fig. 3C show that both FDXR
(matrix protein) and TIM23 (inner membrane) were
resistant to protease treatment. However, addition of
trypsin to digitonin-treated fraction reduced the level
of TIM23 but not that of FDXR. The antibody to
mouse FDXR used in this study weakly crossreacts
with the yeast homolog. Addition of trypsin to Triton-
solubilized samples completely degraded TIM23 and
FDXR. Figure 3D represents a full view of an immu-
noblot of mitochondrial and microsomal proteins
developed with a combination of CYP2E1 and TIM23
antibodies. It is seen that both microsomal and mito-
chondria-associated CYP2E1 consisted of a major
antibody-reactive full-length protein and minor faster-
migrating bands. Furthermore, the mitochondrial frac-
tion crossreacted with TIM23 antibody, whereas the
microsomal fraction lacked detectable TIM23. These
results suggested that the full-length CYP2E1 is tar-
geted to mitochondria in yeast cells.
Localization of CYP2E1 in the mitochondrial inner
membrane–matrix compartment
We used multiple approaches to determine the precise

intramitochondrial localization of CYP2E1 in trans-
formed yeast cells. In the first approach, we assessed the
effects of treatment of mitochondria and mitoplasts
with trypsin. As shown in Fig. 4A, we found that mito-
chondria-associated P4502E1 was relatively resistant to
trypsin treatment, whereas the outer membrane
protein TOM20 was completely degraded. Additionally,
CYP2E1 was resistant to trypsin when the outer
Micro
Mito
2E1
TIM23
TIM23
B
D
A
C
PGK
2E1
Cyto. Micro. Mito.
DPMS
Trypsin:
Digitonin:
Tx-100:
TIM23
++
+
+
+
+

-
-
-

-
-
-
-
Adx-red
Fig. 3. The nature of mitochondria-associated CYP2E1 in yeast cells. (A) Assessment of the integrity of the isolated mitochondria. Mitochon-
drial isolates were subjected to scanning electron microscopy as described in Experimental procedures (magnification: · 30 000). (B) The rel-
ative purity of the isolated mitochondrial, microsomal and cytosolic fractions. Mitochondria, microsomes and cytosol were isolated by
differential centrifugation as described in Experimental procedures. Fifty micrograms of protein from each fraction was subjected to
SDS ⁄ PAGE and probed with antibodies specific for microsomes (DMPS), mitochondria (TIM23), and cytosol (3-phosphoglycerokinase), as
indicated. (C) Isolated mitochondrial fractions were treated with or without digitonin, trypsin or Triton X-100 (Tx-100; 0.2%) and probed with
antibodies against human FDXR and TIM23 as indicated. (D) A full gel pattern of the microsomal and mitochondrial fractions of cells express-
ing full-length P4502E1.
N. B. V. Sepuri et al. Mitochondrial targeting of rat CYPs in yeast cells
FEBS Journal 274 (2007) 4615–4630 ª 2007 University of Pennsylvania. Journal compilation ª 2007 FEBS 4619
membrane was stripped by digitonin. Under these treat-
ment conditions, inner membrane-associated TIM23,
which is exposed out of the membrane lipid bilayer
towards the intermembrane space, was degraded in a
concentration-dependent manner (Fig. 4A). The case
was similar with the intermembrane space protein,
cytochrome c peroxidase (CCPO). By contrast, D1-pyrr-
oline-5-carboxylate dehydrogenase (Put2), a matrix-
localized protein, was protected against protease
treatment even after stripping of the outer membrane.
These results suggest that mitochondria-associated

CYP2E1 is localized inside the inner membrane.
In the second series of experiments, we treated intact
mitochondria with various concentrations of digitonin.
It is known that low concentrations of digitonin (about
0.05%) selectively damage the outer membrane, and
higher concentrations (about 0.1%) damage the inner
membrane. In this experiment, we determined the con-
centration of digitonin required to release mitochon-
dria-associated CYP2E1 into the soluble fraction, and
compare it with the amounts needed to release the
outer membrane-specific marker protein porin and the
mitochondrial matrix protein Put2. Figure 4B shows
that significant CYP2E1 release occurred at digitonin
concentrations between 0.05% and 0.1% (w⁄ v), at
which concentrations Put2 was also released to the sol-
uble fraction to a large extent. The release of porin
started at a much lower concentration of 0.025%.
These results further support the possibility that mito-
chondria-associated CYP2E1 is located inside the
innermembrane compartment.
In the third approach, we used alkaline Na
2
CO
3
extraction to determine whether mitochondrial CYP2E1
is a membrane-intrinsic or membrane-extrinsic protein.
The results showed that most part of the microsomal-
associated CYP2E1 resisted Na
2
CO

3
extraction,
suggesting a transmembrane topology. The mitochon-
dria-associated CYP2E1, on the other hand, was mostly
partitioned in the soluble fraction, indicating a mem-
brane-extrinsic topology (Fig. 4C). TIM23, a bona fide
inner membrane protein partitioned mostly in the
Na
2
CO
3
-insoluble fraction, whereas Put2, a bona fide
matrix protein, was nearly completely extracted in the
soluble phase (Fig. 4C). These results suggest that mito-
chondrial CYP2E1 is mostly a membrane-extrinsic
protein localized in the matrix compartment. However,
Trypsin (µg/mL):
A
B
C
-
2E1
25
50
CCPO
PUT2
TIM23
Mito
Mito
Mitoplast

Mito
Mitoplast
Mito
Mitoplast
Mitoplast
TOM20
Mitoplast
Mito
Pellet
pPut2
Porin
2E1
Digitonin%:
Supernatant
0.01
0.025
0.05
0.1
0.2
0.4
0.01
0.025
0.05
0.1
0.2
0.4
pPut2
Porin
2E1
Micro

Mito
Mito
2E1
2E1
TIM23
pPut2
Pellet
Sup
Mito
Fig. 4. Intramitochondrial localization of CYP2E1 in transformed yeast cells. (A) Mitochondria and mitoplasts from cells expressing CYP2E1
were treated with 0–50 lgÆmL
)1
trypsin as indicated. Mitochondria reisolated by banding through a sucrose layer were analyzed by immuno-
blotting (50 lg protein each) with antibody to CYP2E1. Put2, TIM23 and CCPO and TOM20 were used as matrix, inner membrane, inter-
membrane space and outer membrane markers, respectively. (B) Isolated mitochondria were incubated with increasing concentrations of
digitonin (0–0.4%; 0–400 lgÆmg
)1
protein) for 30 min as described in the text. The digitonin-insoluble (pellet, left panel) and digitonin-soluble
(supernatant, right panel) fractions were separated by centrifugation (14 000 g for 10 min), analyzed by immunoblot, and probed with anti-
bodies to porin (outer membrane marker protein), pPut2 (matrix marker protein), and CYP2E1. (C) Mitochondria of yeast cells expressing
wild-type CYP2E1 were treated with bicarbonate, and both the soluble and insoluble fractions were separated and analyzed by immunoblot
analysis with CYP2E1, TIM23 and Put2 antibodies as indicated.
Mitochondrial targeting of rat CYPs in yeast cells N. B. V. Sepuri et al.
4620 FEBS Journal 274 (2007) 4615–4630 ª 2007 University of Pennsylvania. Journal compilation ª 2007 FEBS
the protein was peripherally associated with the inner
membrane and required washing with a high salt con-
centration (0.2 m NaCl) to be released from the mem-
brane (data not shown).
Role of PKA in the mitochondrial targeting of
CYP2E1

The role of PKA in mitochondrial targeting of
CYP2E1 was investigated using two approaches. The
first approach involved measuring the level of mito-
chondrial targeting of wild-type CYP2E1 in PKA-defi-
cient (a ⁄ c deleted or a ⁄ b deleted) yeast strains. The
western blot in Fig. 5A shows that in control yeast
cells, the microsomal CYP2E1 content was approxi-
mately 4–6-fold higher than the mitochondrial content,
and the microsome-localized CYP2E1 was highly sensi-
tive to trypsin (Fig. 5A, compare lanes 1 and 3). The
mitochondrial CYP2E1 was resistant to externally
added trypsin, and in this regard was similar to the
inner membrane protein TIM23 (Fig. 5A, compare
lanes 2 and 4). The microsome-associated CYP2E1
levels in both the PKA subunit a ⁄ c and a ⁄ b deleted
strains were similar to that in the control yeast strain
(Fig. 5A, lane 1). As observed with the control yeast,
the microsome-associated CYP2E1 in PKA mutant
strains was sensitive to trypsin treatment. Quantitation
of the gel pattern presented in Fig. 5B showed that the
mitochondrial CYP2E1 levels were reduced to < 10%
in the a ⁄ c mutant and < 3% in the a ⁄ b mutant, as
compared to about 25% in the control strain. We also
tested the targeting to mitochondria of Su9-DHFR, in
which the presequence of ATPase subunit 9 of Neuros-
pora crassa was fused to a passenger protein, DHFR.
As seen in Fig. 5A, the level of mitochondrial targeting
of Su9-DHFR, which lacks a canonical PKA phos-
phorylation site, was similar in all three cell lines
tested. CYP2E1 contained a single PKA target site at

Ser129, which was shown to be important in mito-
chondrial targeting of the protein in COS cells. In the
second approach, we tested the level of mitochondrial
targeting of S129A mutant CYP2E1 in transformed
0
10
20
30
40
50
60
70
80
90
100
WT
Micro
Mito
PKA
Δ α/γ
Δ α/β
Δ α/γ
PKA
Δ α/β
% distribution of 2E1
Strain background
TIM23
TIM23
2E1
2E1

Su9-DHFR
Su9-DHFR
*
TIM23
Micro
A
B
C
Mito Micro Mito
Trypsin:
+
2E1
WT
Su9-DHFR
+
*
Micro Mito
Micro
Mito
Trypsin:
++
2E1(S129A)
TIM23
100 72 2 10
% distribution
% distribution
10041
96
12
3

4
Fig. 5. Role of PKA-mediated phosphorylation in mitochondrial targeting of CYP2E1. (A) Isolated mitochondrial and microsomal fractions from
the wild-type and PKA deletion strains were transformed with CYP2E1 or SU9-DHFR expression cDNA constructs or empty vectors. Subcel-
lular fractions were subjected to immunoblot analysis following treatment with or without trypsin as indicated. TIM23 was used as a mito-
chondrial marker. (B) The amount of CYP present in the microsomal or mitochondrial fractions obtained from various PKA mutants were
quantified and represented as a bar graph. (C) Reduced mitochondrial targeting of phosphorylation site mutated CYP2E1 (S129A). Mitochon-
dria and microsomal fractions of yeast cells expressing S129A mutant CYP2E1 were treated with or without trypsin and subjected to immu-
noblot analysis with CYP2E1 and TIM23 antibodies.
N. B. V. Sepuri et al. Mitochondrial targeting of rat CYPs in yeast cells
FEBS Journal 274 (2007) 4615–4630 ª 2007 University of Pennsylvania. Journal compilation ª 2007 FEBS 4621
yeast cells. The results in Fig. 5C show that, as with
the wild-type protein, full-length mutant protein was
targeted to mitochondria, although at a markedly
reduced level (30–40% as compared to the wild-type
construct). These results collectively show that PKA-
mediated phosphorylation, most likely targeted to the
S129 consensus site, is very important for mitochon-
drial targeting of CYP2E1 protein.
Mitochondrial targeting of CYP2B1 in yeast cells
We used the protease-deficient strain pep4D for testing
the mitochondrial targeting of rat CYP2B1. The
western blot in Fig. 6A shows that both ER and
mitochondrial preparations from transformed yeast
cells contained CYP2B1 protein, with about 25% of
protein in the mitochondrial fraction. As in previous
experiments, the mitochondrial preparations contained
TIM23 protein, whereas the microsomes lacked signifi-
cant levels of TIM23. The western blot in Fig. 6B also
shows that the microsome-targeted CYP2B1 was sensi-
tive to trypsin treatment, whereas the mitochondrial

protein showed significant resistance, suggesting an
intramitochondrial location.
CYP contents and catalytic activities
With the aim of correlating the levels of expression of
various apoproteins in yeast with CYP contents, we
measured the P450-heme contents by CO-reduced spec-
tra. As shown in Fig. 7, mitochondrial isolates from
+33⁄ 1A1-expressing yeast cells yielded a CO reduced
and dithionite reduced spectrum with a peak at
448 nm. No peak was observed with mitochondria
from cells transformed with empty vector (data not
shown). Additionally, we did not detect any character-
istic spectrum with mitochondria from cells expressing
+33⁄ 1A1 mutant constructs (data not shown).
Figure 7B shows the P450-heme contents of mito-
chondria and microsomal fractions from yeast strains
expressing various CYP constructs based on CO differ-
ence spectral analysis. Consistent with the negligible
mitochondrial localization of full-length CYP1A1, we
detected no significant CYP in the mitochondrial
isolates. However, the microsomal fraction showed a
high (6.5 nmolÆ mg
)1
) CYP content. Cells expressing
+5⁄ 1A1 showed nearly equal CYP contents in the
mitochondrial and microsomal fraction. Cells express-
ing + 33 ⁄ 1A1 showed no detectable CYP in the
microsomal fraction, but a high (3.5 nmolÆmg
)1
) level

of CYP in mitochondria. Cells expressing full-length
CYP2E1 showed about 300 pmolÆmg
)1
CYP in the
microsomes and 55 pmolÆ mg
)1
in mitochondria. The
mitochondrial fraction from CYP2B1-expressing cells
showed about 50 pmolÆmg
)1
CYP.
As shown in Fig. 8A, the microsomal fraction from
full-length CYP1A1- and + 5 ⁄ 1A1-expressing cells
0.005
0.055
0.105
0.155
0.205
400 420 440 460 480 500
Wave length (nm)
+331A1 mito
1A1 6500 ND
+51A1 250 200
+331A1 ND 3500
CYP450
2E1 300 55
2B1
Vector ND
BA
ND

DHFR 1A1 ND ND
+331A1 (m) ND ND
Not determined 50
pmole/mg microsomal
protein
pmole/mg mitochondrial
protein
Fig. 7. Mitochondrial CYP contents in yeast
cells expressing CYP1A1, CYP2E1 and
CYP2B1 proteins. (A) The reduced CO spec-
tra of the mitochondrial fraction expressing
+ 331A1. The reduced CO spectrum was
performed essentially as described by
Anandatheerthavarada et al. [5]. (B) Relative
levels of CYP in mitochondria and micro-
somes from cells expressing different CYP
proteins. CYP content was measured by CO
difference spectra as described in (A). The
values are the mean of three experiments.
Micro
Mito
25
35
47
62
81
kDa
AB
TIM23
2B1

Micro
Mito
Trypsin
++
Fig. 6. Immunoblot analysis of PeP4D strain expressing full-length
CYP2B1. Isolated mitochondrial or microsomal fractions were trea-
ted with (B) or without (A) trypsin, and subjected to SDS ⁄ PAGE
and immunoblot analysis with CYP2B1 and TIM23 antibodies.
Mitochondrial targeting of rat CYPs in yeast cells N. B. V. Sepuri et al.
4622 FEBS Journal 274 (2007) 4615–4630 ª 2007 University of Pennsylvania. Journal compilation ª 2007 FEBS
showed high activity of ethoxyresorufin O-dealkylation
(EROD), which is a specific marker enzyme for ER-
associated CYP1A1. +33 ⁄ 1A1-expressing cells, how-
ever, showed very low ERND activity. The latter is
consistent with the microsomal CYP content of cells
expressing + 33 ⁄ CYP1A1 (Fig. 7). The mitochondrial
isolates from all these three cell types showed very low
EROD activity. The erythromycin N-demethylase
(ERND) activity pattern (Fig. 8B) was significantly
different from the EROD activity pattern (Fig. 8A).
The microsomal fraction of cells expressing full-length
CYP1A1 and + 5 ⁄ 1A1 showed relatively low ERND
activity. Similarly, consistent with the low or non-
significant mitochondrial localization of full-length
CYP1A1, mitochondria from these cells also showed
very low activity. The mitochondrial isolates from
+5⁄ 1A1- and + 33 ⁄ 1A1-expressing cells, however,
showed high ERND activity (2.0–2.5 nmolÆmg
)1
) with

the endogenous yeast FDX + FDXR. Expression of
mutant 33 ⁄ 1A1 with impaired mitochondrial targeting
showed vastly reduced mitochondrial ERND activity.
Our results on ERND activity of mitochondria-
targeted CYP1A1 supported by mitochondrial
FDX1 + FDXR are consistent with previous studies
from our laboratory [32,33] showing altered catalytic
property of mitochondria-targeted rat and mouse
CYP1A1.
As shown in Fig. 8C, both microsomal and mito-
chondrial fractions from CYP2B1 cDNA-transformed
cells show benzoxyresorufin O-dealkylation (BROD)
activity. The BROD was reduced by about 60% when
the mitochondrial or microsomal fractions were prein-
cubated with CYP2B1 antibody, indicating the specific-
ity of the assay.
As shown in Fig. 9A, both microsomal and mitochon-
drial fractions from wild-type yeast cells transformed
with CYP2E1 cDNA showed nitrosodimethylamine
N-demethylase (NDMA-d) activity. The activities of
both the microsomal CYP2E1 and mitochondrial
CYP2E1 were dependent on the addition of NADPH
(Fig. 9A). The catalytic activities were reduced when
the mitochondrial or microsomal enzymes were prein-
cubated with CYP2E1 antibody or SKF-525, a general
inhibitor of CYPs. These results suggest the specificity
of the assay. We did not observe any significant
increases in the activity of mitochondrial CYP2E1 after
supplementing the reaction with purified bovine
FDX1 + FDXR, possibly because of adequate endo-

genous FDX1 + FDXR in these mitochondrial prepa-
rations (Fig. 9A). Additionally, the activity with the
mitochondrial fraction was inhibited by 50% after addi-
tion of polyclonal antibody to human FDX1 (Fig. 9A),
confirming the role of endogenous FDX + FDXR in
supporting the activity.
To further confirm the role of endogenous
FDX + FDXR in supporting the catalytic activity, we
transformed the FDX (Yah1)- and FDXR (Arh1)-
depleted yeast cells with CYP2E1. Expression of
0
0.05
0.1
0.15
0.2
0.25
Micro
Mito
CYP2B1:

- -
+ +
+ +
Anti-2B1:
+ +
pmoles of resorufin/min/mg
BROD ACTIVITY
ERND ACTIVITY
nmoles/mg/min
0

0.5
1
1.5
2
2.5
3
1A1 +5 +33 1A1 +5 +33 33(M)
Micro
Mito
nmoles/mg/min
EROD ACTIVITY
A
C
B
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
Micro
Mito
Fig. 8. Metabolic activities of the micro-
somal and mitochondrial CYPs in trans-
formed yeast cells. Microsomes and

digitonin-treated mitoplasts from cells
expressing various CYP variants were
assayed for their enzyme activities as indi-
cated in Experimental procedures. (A) and
(B) represent the EROD and ERND activities
of cells transformed with various CYP1A1s,
and (C) represents the BROD activity of
cells transformed with CYP2B1 constructs
as indicated.
N. B. V. Sepuri et al. Mitochondrial targeting of rat CYPs in yeast cells
FEBS Journal 274 (2007) 4615–4630 ª 2007 University of Pennsylvania. Journal compilation ª 2007 FEBS 4623
CYP2E1 in Yah1-depleted cells turned out to be lethal.
We therefore analyzed the Arh1-depleted cells express-
ing CYP2E1. The catalytic activity of the mitochondrial
fraction was reduced by about 60% in FDXR-depleted
cells (Fig. 9B). However, the activity with the mito-
chondrial fraction was restored by the addition of puri-
fied bovine FDX + Fdr. These results confirm that
yeast mitochondrial FDX + FDXR is capable of sup-
porting the catalytic activity of mammalian CYPs.
Discussion
A large majority of mitochondrial proteins are
encoded by nuclear genes, synthesized in the cytoplasm
and post-translationally transported to mitochondria.
The mitochondrial proteome in mammalian cells is
estimated to consist of well over 1500 proteins
imported from the cytoplasm [34]. Several mitochon-
drial matrix-targeted proteins contain an N-terminal
extension or ‘presequence’ that is cleaved upon import
into mitochondria [35]. However, the current estimates

are that more than 50% of the mitochondrial-associ-
ated proteins lack the canonical mitochondria-targeting
signals, and the precise mechanisms by which these
proteins are translocated to the mitochondrial com-
partment remain unclear [35,36]. The mitochondrial
inner membrane-associated carrier protein, uncoupler
proteins and outer membrane proteins belong to this
latter class [37,38]. Additionally, the bimodal targeting
of CYPs to the ER and mitochondria, Alzheimer’s
amyloid precursor protein to the plasma membrane
and mitochondria, and translocation of the cytosolic
glutathione S-transferases to the mitochondrial matrix
compartment, probably represent the targeting of non-
canonical signal-containing proteins to the mitochon-
drial compartment [4–6,12,39]. We have shown that
xenobiotic-inducible CYPs such as rat CYP1A1,
CYP2E1 and CYP2B1, and mouse CYP1A1, contain
chimeric noncanonical-targeting signals that are capa-
ble of targeting proteins to both the ER and mitochon-
dria [5,11,12]. Our results showed that the cryptic
mitochondria-targeting signals present at residues
29–40 in various CYP proteins are activated by two
different mechanisms: (a) proteolytic processing at the
N-terminus by a cytosolic endoprotease, resulting in
the exposure of cryptic mitochondrial targeting signal,
as in the case of CYP1A1 [4]; and (b) PKA- or protein
kinase C-mediated phosphorylation of nascent chains
either at the N-terminus (Ser128 or Ser129) or the
C-terminus, which promotes the mitochondrial target-
ing of CYP2E1, CYP2B1, and GSTA4-4 [6,12,40]. In

this study, we show that bimodal targeting of
CYP1A1, CYP2E1 and CYP2B1 are conserved in the
heterologous cell system Saccharomyces cerevisiae.
The N-terminal four or 32 amino acid sequence
(+ 5 ⁄ 1A1 or + 33 ⁄ 1A1) of CYP1A1 exposes the cryp-
tic mitochondria-targeting sequence at positions 33–39
of the protein, thus targeting the + 331A1 protein
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
Micro Mito
Cell Fractions
NADPH
FDR& FDX
Ant-FDX
SKF-525
Anti-2E1
Anti-IgG
nmoles/HCHO/min/mg
Wt cells expressing CYP2E1
AB
0

0.5
1
1.5
2
2.5
3
3.5
4
4.5
Micro
Mito
nmoles/HCHO/min/mg
Cell Fractions
NADPH
FDR& FDX
FDXR depleted cells expressing CYP2E1
Fig. 9. Mitochondrial CYP2E1 activity in wild-type and FDXR-depleted yeast strains. Mitochondrial and microsomal NMDA-d activity in
(A) wild-type cells expressing CYP2E1 and (B) FDXR-depleted cells expressing CYP2E1. Reactions were carried out as described in Experi-
mental procedures, using 50 lg of mitochondria or microsomal proteins. NADPH (1 m
M), FDX + FDXR (1 lg each), antibody to FDX (2 lg),
antibody to CYP2E1 (1 lg), preimmune IgG (anti-IgG) and SKF-525 (0.1 m
M) were added to the reaction before initiating the enzyme activity
by adding the dimethylnitrosomine (4 m
M). Depletion of FDXR was carried out as described in Experimental procedures. Details of enzyme
assays are given in Experimental procedures.
Mitochondrial targeting of rat CYPs in yeast cells N. B. V. Sepuri et al.
4624 FEBS Journal 274 (2007) 4615–4630 ª 2007 University of Pennsylvania. Journal compilation ª 2007 FEBS
exclusively to the mitochondria. In the present study,
we observed no significant targeting of intact CYP1A1
to mitochondria, possibly due to the absence of cyto-

solic endoprotease specific for cleavage at the + 5 and
+ 33 positions (Fig. 2). Consistent with this possibility,
both the + 5 ⁄ 1A1 and + 33 ⁄ 1A1 proteins are
efficiently targeted to the yeast mitochondrial compart-
ment in transformed cells (Fig. 2). Mitochondria-
targeting sequences are usually rich in positively
charged residues, either as part of an amphiphilic heli-
cal structure or as random structures. The charged res-
idues have been suggested to play an important role
in binding to TOM20 and TOM22 outer membrane
receptors of the import complex [41]. The positive resi-
dues at the + 34 and + 39 positions of + 331A1 are
indeed important for targeting of the protein to mito-
chondria, as substitutions at these positions with neu-
tral or hydrophobic amino acids essentially abolished
mitochondrial import (Fig. 2D,E). Our results also
showed that the + 51A1 construct was mostly targeted
to mitochondria, with a minor fraction being found in
association with the ER (Fig. 2A). An N-terminally
blocked CYP1A1 (DHFR-1A1 fusion) was not tar-
geted to either the ER or the mitochondria, whereas
the full-length was CYP1A1 targeted efficiently to the
ER, suggesting that an open N-terminal signal domain
is critical for protein targeting to the ER (Fig. 2D).
In contrast to the targeting of CYP1A1 requiring
N-terminal truncation, we have shown that intact
CYP2B1 and CYP2E1 proteins are targeted to mito-
chondria in inducer-treated rat livers [5], in transiently
transfected COS cells, and also in an in vitro transport
system with isolated rat liver mitochondria [6,10,12].

Our results from the mutational analysis also showed
that the two positively charged residues at positions 24
and 25 of CYP2E1 formed a critical part of the mito-
chondria-targeting signal [12]. Accordingly, N-terminal
truncated proteins (+ 29 ⁄ 2E1 and + 36 ⁄ 2E1) failed to
show significant mitochondrial targeting under in vitro
and in vivo conditions [10]. In sharp contrast to our
results, Ingelman-Sunderberg’s group first reported
that N-terminal truncated + 29 ⁄ 2E1 expressed in hepa-
toma cells is targeted to mitochondria as a 50 kDa
soluble protein [21], suggesting proteolytic processing
at an uncharacterized internal site. These same investi-
gators reported that N-terminal truncated + 29 ⁄ 2E1
and + 82 ⁄ 2E1 expressed in yeast cells failed to enter
mitochondria, but existed as outer membrane-bound
forms [23]. In the present study, using rigorous con-
trols on mitochondrial integrity and selective markers
for the outer membrane, intermembrane space, inner
membrane and matrix compartment, we demonstrated
that CYP2E1 is targeted to yeast mitochondria as an
intact protein, and that PKA-mediated phosphoryla-
tion is critical for mitochondrial targeting. These find-
ings are consistent with results from three different
groups showing nearly identical gel migration of mito-
chondrial and microsomal CYP2E1 in the mouse and
rat liver under different pathophysiologic conditions
[5,10,26,42,43].
Although the precise reasons for this sharp differ-
ence in the targeting patterns of CYP2E1 remains
unclear, it is likely that the hepatoma cells used by

Neve & Ingelman-Sundberg [22] contain an unusual
cytosolic endoprotease that clips the protein and acti-
vates an internal cryptic mitochondria-targeting signal.
It is known that randomly generated peptides, even
from Escherichia coli, can potentially function as mito-
chondria-targeting signals when fused to the N-termi-
nus of a reporter protein, DHFR [44]. This does not
mean that such internal signals are functional under
physiologic settings while working with an intact pro-
tein. A major difference that may account for the
observed difference in the results is that we have used
well-defined and intact mitochondrial isolates for ana-
lyzing CYP proteins, whereas Neve & Ingelman-Sund-
berg [21,22,22,23] have consistently used total cell
homogenates with undefined mitochondrial integrity.
In this study, we have ensured mitochondrial purity
and integrity by the use of several criteria: (a) mito-
chondria sedimented through a 0.65 m sucrose layer,
which reduces microsomal contamination; (b) mito-
chondria intact as seen by scanning electron micros-
copy; and (c) mitochondrial markers for the outer
membrane, intermembrane space, inner membrane and
matrix compartments sequentially released at expected
digitonin concentrations. We have observed that mito-
chondria-associated CYP2E1 is resistant to limited
protease treatment, similar to a bona fide matrix pro-
tein, Put2 (Fig. 4A). However, both CYP2E1 and Put2
are readily degraded by trypsin following treatment
with Triton X-100 (data not shown). Furthermore,
mitochondrial CYP2E1 was released by digitonin only

when the matrix protein Put2 was released (Fig. 4B).
These results provide rigorous proof for the intra-
mitochondrial location of CYP2E1. Furthermore, as
shown for rat liver mitochondrial and COS cell mito-
chondrial CYP2E1, the yeast mitochondrial CYP2E1
is readily extracted by alkaline Na
2
CO
3
, suggesting its
membrane-extrinsic orientation (Fig. 4C).
PKA is known to play important roles in cellular
regulation, cell growth, metabolism, stress resistance,
and filamentous invasive growth [45,46]. Studies have
also shown that in yeast, PKA controls the nuclear
localization of certain transcription factors, such as
Msn2 and Msn4, and of snf1 kinase in the cytosol in
N. B. V. Sepuri et al. Mitochondrial targeting of rat CYPs in yeast cells
FEBS Journal 274 (2007) 4615–4630 ª 2007 University of Pennsylvania. Journal compilation ª 2007 FEBS 4625
response to glucose, suggesting the importance of PKA
in protein trafficking [47,48]. In this study, we used
PKA knockout mutants to determine the role of PKA
in mitochondrial targeting of CYP2E1. In mutants
lacking PKA, CYP2E1 was exclusively localized to the
microsomal fraction, indicating that PKA is required
for CYP2E1 targeting to the mitochondria (Fig. 5A).
The present findings suggest that a conserved PKA-
mediated pathway is required for targeting of a subset
of precursor proteins to mitochondria. It is likely that
PKA phosphorylation is a more general pathway for

the targeting of proteins lacking the canonical signals
to mitochondria.
In mammalian cells, mitochondrial CYP activity is
dependent on FDX and FDXR proteins. Yeast homo-
logs of FDXR, Arh1, and FDX, Yah1, are known to
be involved in both heme A biosynthesis and iron–sul-
fur cluster assembly [49]. It is not known whether these
yeast homologs can serve as electron donors for mam-
malian CYPs. Our results show that yeast homologs
can serve as electron donors for mammalian CYPs
(Fig. 9), as mitochondria without externally added
FDX + FDXR are able to support CYP-mediated
ERND, BROD and NDMA-d metabolism. CYP2E1
expressed in yeast strains with depleted FDXR resulted
in decreased activity (Fig. 9B) that was restored by
adding bovine FDX + FDXR (Fig. 9B). Additionally,
the enzyme activity was inhibited by antibody to
human FDX (Fig. 9A). These results provide evidence
that yeast FDX + FDXR fully supports that activity
of mammalian CYPs.
As found before with rat liver mitochondrial
enzymes, yeast mitochondrial CYP1A1 shows high
ERND activity and mitochondrial CYP2E1 exhibits
high NDMA-d activity. The activities of both micro-
somal and mitochondrial enzymes were dependent on
addition of NADPH and inhibited by SKF-525.
In summary, our results show that the mitochondrial
import of mammalian CYP family 1 and family 2 pro-
teins is highly conserved in yeast, except that the cyto-
solic processing activity for CYP1A1 seems to be

lacking. Our results provide rigorous documentation of
the import of intact CYP2E1 into mitochondria, which
is also dependent on PKA-mediated phosphorylation
at Ser129. The precise physiologic significance of mito-
chondria-targeted CYPs remains unknown, although
several studies, including ours, suggest roles in drug
metabolism and reactive oxygen species production
[26,50]. Increased levels of mitochondrial CYP2E1 in
streptozotocin-induced diabetes have been implicated
in reactive oxygen species production and depletion of
the mitochondrial glutathione pool, thus contributing
to oxidative stress [26,50].
Experimental procedures
Materials
Mata his3D1 leu2D0 met15D0 ura3D0 strain BY4741 was
obtained from Research Genetics Inc. (Huntsville, AL,
USA). The protease-deficient strain (pep4D) ade2–101
met2 his3D200 lys2–801 ura3–52 was a kind gift from E
Johnson (Thomas Jefferson University). BY4741 and pep4D
strains were used to express rat CYP cDNAs driven by the
elongation factor promoter on either a 2 lm or centromeric
URA3 or Leu2 plasmids. W303 strain MATa ade2-1 trp1-1
his3-11,15 can1-100 ura3-1 leu2-3,112, PKA a ⁄ cDstrain
MATa tpk1::URA3 tpk2::HIS3 leu2-3112 ura3-1 trp1-1 his3-
11,15 ade2-1 can1-100, PKA a ⁄ bDstrain MAa tpk1::URA3
tpk3::TRP1 leu2-3112 ura3-1 trp1-1 his3-11,15 ade2-1 can1-
100 and PKA b ⁄ cDstrain MATa tpk2::HIS3 tpk3::TRP1
leu2-3112 ura3-1 trp1-1 his3-11,15 ade2-1 can1-100 were kind
gifts from M Carlson (Columbia University, New York).
MATa ura3-52 lys2-80(amber) his3-D200 trp1-D63 leu2-D1

URA3::pGalArh1pHA and MATa ura3-52 lys2-80(amber)
his3-D200 trp1-D63 leu2-D1 URA3::pGalYah1pHA were
kind gifts from A Dancis (University of Pennsylvania). Plas-
mids were transformed into different yeast strains by using
the standard LioAc method. The yeast cells were grown to
log phase at 30 °C in supplemental minimal medium con-
taining 2% dextrose (SD) or 2% raffinose (SG), and appro-
priate amino acids was used to select the various plasmids.
To induce the Gal promoter (Gal-Yah1 or Gal-Arh1), cells
were grown in 2% raffinose and 0.5% galactose. Expression
from the Gal promoter was turned off by shifting the
cultures to identical medium without galactose [51].
Plasmids and cloning strategy
The full-length and truncated CYP cDNAs were amplified
by PCR using Taq polymerase (Qiagen, Chatsworth, CA,
USA) [8]. An EcoRI site was used for the in-frame fusion
of DHFR to the N-terminus of CYP1A1 by overlap PCR.
The double mutants (R34D and K39D) and single (R34D)
mutants of + 33 ⁄ 1A1 were generated by using appropri-
ately substituted forward primers in PCR reactions. The
S129A mutant CYP2E1 cDNA was generated as described
previously [10]. The cDNAs were cloned into the 2 lm vec-
tor pTEF-URA3 or Leu2 or the centromeric pTEF-URA3
or Leu2 plasmid [52]. The details of cDNAs, restriction
sites used and plasmid designations are listed in Table 1.
Subcellular fractionation and isolation of
mitochondria and mitoplasts
Mitochondria were isolated from yeast strains containing
high copy levels of various CYP cDNAs as described previ-
ously [20]. Briefly, yeast strains expressing various CYPs

were grown on selective synthetic medium in the presence
Mitochondrial targeting of rat CYPs in yeast cells N. B. V. Sepuri et al.
4626 FEBS Journal 274 (2007) 4615–4630 ª 2007 University of Pennsylvania. Journal compilation ª 2007 FEBS
of 2% glucose or raffinose and were harvested during log
phase and treated with zymolyase to produce spheroplasts.
Spheroplasts were homogenized in SEM buffer (250 mm
sucrose, 1 mm EDTA, 20 mm Mops, pH 7.2) containing
protease inhibitors and 0.4% BSA. The homogenate was
centrifuged twice for 5 min at 2500 g (Sorvall RC5B,
SA600) to remove unbroken cells and nuclei; the superna-
tant was then centrifuged at 12 000 g for 10 min (Sorvall
RC5B, SA600). The pellet was washed twice with SEM buf-
fer, and the final pellet was resuspended in SEM buffer and
passed through a 0.65 m sucrose cushion. The pellet was
again resuspended in either SEM buffer or CYP buffer
(50 mm potassium phosphate, pH 7.4, 20% glycerol,
0.5 mm dithiothreitol, 1 mm EDTA, 0.1 mm phenyl-
methanesulfonyl fluoride). The 12 000 g supernatant was
spun at 100 000 g to isolate microsomes (Beckman L7
ultracentrifuge, SW 50.1 rotor), which were then suspended
in the CYP buffer. The purity of the mitochondria was rou-
tinely assessed by immunoblotting the subcellular fractions
with antibodies specific for mitochondria (TIM23), micro-
somes (DMPS), and cytosol (3-phosphoglycerate kinase).
To remove proteins that are peripherally associated with
the mitochondrial fraction, isolated mitochondria were trea-
ted with trypsin in SEM buffer for 20 min on ice. Trypsin
was then inhibited by adding a 10-fold excess of soybean
trypsin inhibitor followed by washing the mitochondria
with SEM buffer containing phenylmethanesulfonyl fluo-

ride.
Digitonin fractionation
Isolated mitochondria (100 lg) were resuspended in SEM
buffer containing indicated concentrations of digitonin
(0.05% to 0.4%, w ⁄ v) for 30 min on ice with occasional
mixing. Digitonin-soluble and digitonin-insoluble fractions
were separated by centrifugation at 12 000 g for 15 min
(Labnet, Hermlez-233M, 220.59 rotor), and the pellet was
washed once again with SEM buffer. The pellet and the
supernatant fractions were mixed with SDS sample buffer
and processed as described [53].
Alkaline extraction of membrane proteins
To determine the mode of association of CYPs with the
membrane, we used an alkaline extraction method as
described by Clark & Waterman [54]. Mitochondrial and
microsomal fractions were treated with 0.1 m Na
2
CO
3
(pH 11.0) for approximately 20 min on ice, and both the sol-
uble and insoluble fractions were separated by centrifugation
(Labnet, Hermlez-233M, 220.59 rotor); the samples were
processed as described by Anandatheerthavarada et al. [6].
Table 1. Plasmids used in this study.
Plasmid, gene Marker Cloning Source ⁄ Reference
pNS45, 1A1 pTEF 2l-Ura3 5¢ BamH1 sense and 3¢ HindIII antisense primers
were used to amplify the full length of 1A1 and
cloned into the same sites of yeast vector
[5]
pNS47, DHFR1A1 pTEF 2l-Ura3 5¢ BamH1 sense and 3¢ HindIII antisense primers

were used to amplify the DHFR ⁄ 1A1 and cloned
into the same sites of yeast vector
[5]
pNS48, + 331A1
point mutation (R34D)
pTEF 2l-Ura3 The point mutant generated by incorporating
appropriate base substitutions in the forward primer
and cloned into yeast vector as above
This study
pNS49, + 331A1 pTEF2l-Ura3 Cloned as above [1]
pNS56, + 51A1 pTEF2l-Ura3 Cloned as above [1]
pNS58, + 331A1
(double mutant, R34D; K39I)
pTEF2l-Ura3 The double mutant was generated as pNS48 This study
pNS59, 2B1 pTEF2l-Ura3 5¢ EcoRI sense and 3¢ XhoI antisense primers
were used to clone into yeast vector as above
[3]
pNS61, 2E1 pTEF2l-Ura3 5¢ EcoRI sense and 3¢ XhoI antisense primers
were used to clone into yeast vector as above
[7]
pNS62, 2E1
(point mutation S128A)
pTEF2l-Ura3 5¢ EcoRI sense and 3¢ XhoI antisense primers
were used to clone the phosphomutant into
yeast vector as above
[7]
pNS69, su9-DHFR pTEF2l-Ura3 5¢ BamH1 sense and 3¢ HindIII antisense primers
were used to amplify the su9-DHFR and cloned
into the same sites of yeast vector
This study

pNS120, 2E1 pTEF2l-Leu2 pNS61 was cut with
Spe1 and Xho1 and cloned
into same sites of yeast vector
This study
N. B. V. Sepuri et al. Mitochondrial targeting of rat CYPs in yeast cells
FEBS Journal 274 (2007) 4615–4630 ª 2007 University of Pennsylvania. Journal compilation ª 2007 FEBS 4627
Scanning electron microscopy
Mitochondria were isolated as described above, and the
final mitochondrial pellets were fixed in NaCl ⁄ P
i
containing
4% formaldehyde and 2% glutaraldehyde. Mitochondria
were dehydrated through a graded ethanol series and
embedded in hard grade LR White Resin (Sigma-Aldrich,
St Louis, MO, USA). The sections were then examined and
photographed using a JEOL (Hammarbaccan, Sollen Tuna,
Sweden) 100CX electron microscope.
Catalytic activity
The EROD and BROD activities were determined by mea-
suring the formation of resorufin as described previously
[55]. The mitoplasts were isolated by treating the mitochon-
drial fraction with 0.0075% digitonin and isolation through
a sucrose cushion. The reaction mixture was composed of
20 mm Tris ⁄ HCl (pH 7.8), 20 mm MgCl
2
,10lm dicumarol,
3.2 mgÆmL
)1
BSA, 200 lg of protein, and 20 lm resorufin
derivatives as a substrate. Reactions were initiated by the

addition of 3 mm NADPH, and incubation was continued
for 30 min at 37 °C in a shaking water bath. The reactions
were terminated by adding 2 mL of ice-cold methanol, and
insoluble particles were sedimented by centrifugation at
10 000 g for 10 min at room temperature (Labnet, Hermlez-
233M, 220.59 rotor). Spectrophotometric determinations of
the supernatant containing resorufin were made at the exci-
tation and emission wavelengths of 528 nm and 590 nm,
respectively. The ERND activities of mitochondria and
mitoplast fractions expressing yeast CYP1A1 were measured
as described by Anandatheerthavarada et al. [32]. The assay
mixture, containing 50 mm Tris ⁄ HCl (pH 7.4), 20 mm
MgCl
2
, 500 lg of protein, and 1 mm erythromycin, was
preincubated for 3 min. The reaction was initiated by the
addition of 3 mm NADPH, and continued for 30 min at
37 °C in a shaking water bath. The reaction was terminated
by the addition of 250 lL of ice-cold 10% trichloroacetic
acid. The reaction product, formaldehyde, was measured as
previously described [56]. In all cases, both enzyme and zero-
time blanks were also analyzed.
NDMA-d activity was assayed by the Anderson & Angel
method [57] as modified by Yadav et al. [58]. The assay
mixture contained 50 lg of protein, 70 mm Tris ⁄ HCl
(pH 7.4), 10 mm semicarbazide, 14 mm MgCl
2
, 215 mm
KCl, 1 mm NADPH, and 4 mm NDMA in a 1.0 mL final
volume. The reaction mixture was incubated at 37 °C for

30 min, and the reaction was stopped by the addition of
0.1 mL of 25% zinc sulfate and 0.1 mL of saturated barium
hydroxide. After centrifugation at 2000 g for 10 min
(Labnet, Hermlez-233M, 220.59 rotor), 0.7 mL of the
supernatant was mixed with an equal amount of Nash
reagent. The tubes were then incubated at 70 °C for
20 min, and the formaldehyde that was formed was mea-
sured at 415 nm.
The CYP contents of microsomes and mitochondria were
measured by reduced CO difference spectra as previously
described [5].
Acknowledgements
We are grateful to Dr Carlson (Columbia University,
New York, NY) for providing the yeast PKA mutant
strains, Dr Erica Johnson (Thomas Jefferson Univer-
sity, Philadelphia, PA) for providing the Pep4D yeast
strain, and Dr Andy Dancis for providing with FDX-
and FDXR-depleted strains. We also thank members
of the Avadhani laboratory for valuable suggestions
and comments. This research was supported by NIH
grant GM-34883.
References
1 Guengerich FP (2003) Cytochromes P450, drugs, and
diseases. Mol Interv 3, 194–204.
2 Guengerich FP (2006) Cytochrome P450s and other
enzymes in drug metabolism and toxicity. AAPS J 8,
E101–E111.
3 Nebert DW & Dalton TP (2006) The role of cyto-
chrome P450 enzymes in endogenous signalling path-
ways and environmental carcinogenesis. Nat Rev Cancer

6, 947–960.
4 Addya S, Anandatheerthavarada HK, Biswas G,
Bhagwat SV, Mullick J & Avadhani NG (1997) Target-
ing of NH2-terminal-processed microsomal protein to
mitochondria: a novel pathway for the biogenesis of
hepatic mitochondrial P450MT2. J Cell Biol 139,
589–599.
5 Anandatheerthavarada HK, Addya S, Dwivedi RS,
Biswas G, Mullick J & Avadhani NG (1997) Locali-
zation of multiple forms of inducible cytochromes P450
in rat liver mitochondria: immunological characteristics
and patterns of xenobiotic substrate metabolism. Arch
Biochem Biophys 339, 136–150.
6 Anandatheerthavarada HK, Biswas G, Mullick J, Sepu-
ri NB, Otvos L, Pain D & Avadhani NG (1999) Dual
targeting of cytochrome P4502B1 to endoplasmic reticu-
lum and mitochondria involves a novel signal activation
by cyclic AMP-dependent phosphorylation at ser128.
EMBO J 18, 5494–5504.
7 Omura T (2006) Mitochondrial P450s. Chem Biol Inter-
act 163, 86–93.
8 Bhagwat SV, Biswas G, Anandatheerthavarada HK,
Addya S, Pandak W & Avadhani NG (1999) Dual
targeting property of the N-terminal signal sequence
of P4501A1. Targeting of heterologous proteins to
endoplasmic reticulum and mitochondria. J Biol Chem
274, 24014–24022.
Mitochondrial targeting of rat CYPs in yeast cells N. B. V. Sepuri et al.
4628 FEBS Journal 274 (2007) 4615–4630 ª 2007 University of Pennsylvania. Journal compilation ª 2007 FEBS
9 Boopathi E, Anandatheerthavarada HK, Bhagwat SV,

Biswas G, Fang JK & Avadhani NG (2000) Accumula-
tion of mitochondrial P450MT2, NH(2)-terminal trun-
cated cytochrome P4501A1 in rat brain during chronic
treatment with beta-naphthoflavone. A role in the
metabolism of neuroactive drugs. J Biol Chem 275,
34415–34423.
10 Robin MA, Anandatheerthavarada HK, Fang JK,
Cudic M, Otvos L & Avadhani NG (2001) Mitochon-
drial targeted cytochrome P450 2E1 (P450 MT5) con-
tains an intact N terminus and requires mitochondrial
specific electron transfer proteins for activity. J Biol
Chem 276, 24680–24689.
11 Dasari VR, Anandatheerthavarada HK, Robin MA,
Boopathi E, Biswas G, Fang JK, Nebert DW &
Avadhani NG (2006) Role of protein kinase C-mediated
protein phosphorylation in mitochondrial translocation
of mouse CYP1A1, which contains a non-canonical
targeting signal. J Biol Chem 281, 30834–30847.
12 Robin MA, Anandatheerthavarada HK, Biswas G,
Sepuri NB, Gordon DM, Pain D & Avadhani NG
(2002) Bimodal targeting of microsomal CYP2E1 to
mitochondria through activation of an N-terminal
chimeric signal by cAMP-mediated phosphorylation.
J Biol Chem 277, 40583–40593.
13 Gilmore R, Blobel G & Walter P (1982) Protein translo-
cation across the endoplasmic reticulum. I. Detection in
the microsomal membrane of a receptor for the signal
recognition particle. J Cell Biol 95, 463–469.
14 Walter P & Blobel G (1981) Translocation of proteins
across the endoplasmic reticulum. II. Signal recognition

protein (SRP) mediates the selective binding to
microsomal membranes of in-vitro-assembled poly-
somes synthesizing secretory protein. J Cell Biol 91,
551–556.
15 Black SD & Coon MJ (1982) Structural features of liver
microsomal NADPH-cytochrome P-450 reductase.
Hydrophobic domain, hydrophilic domain, and connect-
ing region. J Biol Chem 257, 5929–5938.
16 Fujii-Kuriyama Y, Negishi M, Mikawa R & Tashiro Y
(1979) Biosynthesis of cytochrome P-450 on membrane-
bound ribosomes and its subsequent incorporation into
rough and smooth microsomes in rat hepatocytes. J Cell
Biol 81, 510–519.
17 Sakaguchi M, Mihara K & Sato R (1987) A short
amino-terminal segment of microsomal cytochrome
P-450 functions both as an insertion signal and as a
stop-transfer sequence. EMBO J 6, 2425–2431.
18 Davis AJ, Sepuri NB, Holder J, Johnson AE & Jensen
RE (2000) Two intermembrane space TIM complexes
interact with different domains of Tim23p during its
import into mitochondria. J Cell Biol 150, 1271–1282.
19 Sepuri NB, Gordon DM & Pain D (1998) A GTP-
dependent ‘push’ is generally required for efficient
protein translocation across the mitochondrial inner
membrane into the matrix. J Biol Chem 273, 20941–
20950.
20 Sepuri NB, Schulke N & Pain D (1998) GTP hydrolysis
is essential for protein import into the mitochondrial
matrix. J Biol Chem 273, 1420–1424.
21 Neve EP & Ingelman-Sundberg M (1999) A soluble

NH(2)-terminally truncated catalytically active form of
rat cytochrome P450 2E1 targeted to liver mitochondria
(1). FEBS Lett 460, 309–314.
22 Neve EP & Ingelman-Sundberg M (2001) Identification
and characterization of a mitochondrial targeting signal
in rat cytochrome P450 2E1 (CYP2E1). J Biol Chem
276, 11317–11322.
23 Neve EP, Hidestrand M & Ingelman-Sundberg M
(2003) Identification of sequences responsible for intra-
cellular targeting and membrane binding of rat CYP2E1
in yeast. Biochemistry 42, 14566–14575.
24 Bauer MF, Rothbauer U, Muhlenbein N, Smith RJ,
Gerbitz K, Neupert W, Brunner M & Hofmann S
(1999) The mitochondrial TIM22 preprotein translocase
is highly conserved throughout the eukaryotic kingdom.
FEBS Lett 464, 41–47.
25 Hartl FU & Neupert W (1990) Protein sorting to mito-
chondria: evolutionary conservations of folding and
assembly. Science 247, 930–938.
26 Bai J & Cederbaum AI (2006) Overexpression of
CYP2E1 in mitochondria sensitizes HepG2 cells to the
toxicity caused by depletion of glutathione. J Biol Chem
281, 5128–5136.
27 Barros MH & Nobrega FG (1999) YAH1 of Saccharo-
myces cerevisiae: a new essential gene that codes for a
protein homologous to human adrenodoxin. Gene 233,
197–203.
28 Manzella L, Barros MH & Nobrega FG (1998) ARH1
of Saccharomyces cerevisiae: a new essential gene that
codes for a protein homologous to the human adreno-

doxin reductase. Yeast 14, 839–846.
29 Liao M, Zgoda VG, Murray BP & Correia MA (2005)
Vacuolar degradation of rat liver CYP2B1 in Saccharo-
myces cerevisiae: further validation of the yeast model
and structural implications for the degradation of
mammalian endoplasmic reticulum P450 proteins. Mol
Pharmacol 67, 1460–1469.
30 Bar-Nun S, Kreibich G, Adesnik M, Alterman L,
Negishi M & Sabatini DD (1980) Synthesis and
insertion of cytochrome P-450 into endoplasmic
reticulum membranes. Proc Natl Acad Sci USA 77,
965–969.
31 Monier SLP, Kreibich G, Sabatini DD & Adesnik M
(1988) Signals for the incorporation and orientation of
cytochrome P450 in the endoplasmic reticulum mem-
brane. J Cell Biol 107, 457–470.
32 Anandatheerthavarada HK, Addya S, Mullick J &
Avadhani NG (1998) Interaction of adrenodoxin with
P4501A1 and its truncated form P450MT2 through
N. B. V. Sepuri et al. Mitochondrial targeting of rat CYPs in yeast cells
FEBS Journal 274 (2007) 4615–4630 ª 2007 University of Pennsylvania. Journal compilation ª 2007 FEBS 4629
different domains: differential modulation of enzyme
activities. Biochemistry 37, 1150–1160.
33 Anandatheerthavarada HK, Vijayasarathy C, Bhagwat
SV, Biswas G, Mullick J & Avadhani NG (1999) Physio-
logical role of the N-terminal processed P4501A1 tar-
geted to mitochondria in erythromycin metabolism and
reversal of erythromycin-mediated inhibition of mito-
chondrial protein synthesis. J Biol Chem 274, 6617–6625.
34 Taylor SW, Fahy E & Ghosh SS (2003) Global organel-

lar proteomics. Trends Biotechnol 21, 82–88.
35 Neupert W (1997) Protein import into mitochondria.
Annu Rev Biochem 66, 863–917.
36 Pfanner N & Geissler A (2001) Versatility of the mito-
chondrial protein import machinery. Nat Rev Mol Cell
Biol 2, 339–349.
37 Kurz M, Martin H, Rassow J, Pfanner N & Ryan MT
(1999) Biogenesis of Tim proteins of the mitochondrial
carrier import pathway: differential targeting mecha-
nisms and crossing over with the main import pathway.
Mol Biol Cell 10, 2461–2474.
38 Ryan MT, Muller H & Pfanner N (1999) Functional
staging of ADP ⁄ ATP carrier translocation across the
outer mitochondrial membrane. J Biol Chem 274,
20619–20627.
39 Niranjan BG, Raza H, Shayiq RM, Jefcoate CR &
Avadhani NG (1988) Hepatic mitochondrial cytochrome
P-450 system. Identification and characterization of a
precursor form of mitochondrial cytochrome P-450
induced by 3-methylcholanthrene. J Biol Chem 263,
575–580.
40 Robin MA, Prabu SK, Raza H, Anandatheerthavarada
HK & Avadhani NG (2003) Phosphorylation enhances
mitochondrial targeting of GSTA4-4 through increased
affinity for binding to cytoplasmic Hsp70. J Biol Chem
278, 18960–18970.
41 Brix J, Rudiger S, Bukau B, Schneider-Mergener J &
Pfanner N (1999) Distribution of binding sequences for
the mitochondrial import receptors Tom20, Tom22, and
Tom70 in a presequence-carrying preprotein and a non-

cleavable preprotein. J Biol Chem 274, 16522–16530.
42 Raza H, Ahmed I, John A & Sharma AK (2000) Modu-
lation of xenobiotic metabolism and oxidative stress in
chronic streptozotocin-induced diabetic rats fed with
Momordica charantia fruit extract. J Biochem Mol
Toxicol 14, 131–139.
43 Robin MA, Sauvage I, Grandperret T, Descatoire V,
Pessayre D & Fromenty B (2005) Ethanol increases
mitochondrial cytochrome P450 2E1 in mouse liver and
rat hepatocytes. FEBS Lett 579, 6895–6902.
44 Lemire BD, Fankhauser C, Baker A & Schatz G
(1989) The mitochondrial targeting function of ran-
domly generated peptide sequences correlates with
predicted helical amphiphilicity. J Biol Chem 264,
20206–20215.
45 Thevelein JM & de Winde JH (1999) Novel sensing
mechanisms and targets for the cAMP-protein kinase A
pathway in the yeast Saccharomyces cerevisiae. Mol
Microbiol 33, 904–918.
46 Pagliarini DJ & Dixon JE (2006) Mitochondrial modu-
lation: reversible phosphorylation takes center stage?
Trends Biochem Sci 31, 26–34.
47 Hedbacker K, Townley R & Carlson M (2004) Cyclic
AMP-dependent protein kinase regulates the subcellular
localization of Snf1-Sip1 protein kinase. Mol Cell Biol
24, 1836–1843.
48 Wiatrowski HA & Carlson M (2003) Yap1 accumulates
in the nucleus in response to carbon stress in Saccharo-
myces cerevisiae. Eukaryot Cell 2, 19–26.
49 Alves R, Herrero E & Sorribas A (2004) Predictive

reconstruction of the mitochondrial iron–sulfur cluster
assembly metabolism. I. The role of the protein pair fer-
redoxin–ferredoxin reductase (Yah1–Arh1). Proteins 56,
354–366.
50 Raza H, Prabu SK, Robin MA & Avadhani NG (2004)
Elevated mitochondrial cytochrome P450 2E1 and glu-
tathione S-transferase A4-4 in streptozotocin-induced
diabetic rats: tissue-specific variations and roles in oxi-
dative stress. Diabetes 53, 185–194.
51 Li J, Saxena S, Pain D & Dancis A (2001) Adrenodoxin
reductase homolog (Arh1p) of yeast mitochondria
required for iron homeostasis. J Biol Chem 276, 1503–
1509.
52 Mumberg D, Muller R & Funk M (1994) Regulatable
promoters of Saccharomyces cerevisiae: comparison of
transcriptional activity and their use for heterologous
expression. Nucleic Acids Res 22, 5767–5768.
53 Mahlke K, Pfanner N, Martin J, Horwich AL, Hartl FU
& Neupert W (1990) Sorting pathways of mitochondrial
inner membrane proteins. Eur J Biochem 192, 551–555.
54 Clark BJ & Waterman MR (1991) The hydrophobic
amino-terminal sequence of bovine 17 alpha-hydroxylase
is required for the expression of a functional hemopro-
tein in COS 1 cells. J Biol Chem 266, 5898–5904.
55 Sinclair JF, Sinclair PR, Smith EL, Bement WJ, Pome-
roy J & Bonkowsky H (1981) Ethanol-mediated increase
in cytochrome P-450 in cultured hepatocytes. Biochem
Pharmacol 30, 2805–2809.
56 Reinke LA & Moyer MJ (1985) p-Nitrophenol hydroxy-
lation. A microsomal oxidation which is highly induc-

ible by ethanol. Drug Metab Dispos 13, 548–552.
57 Anderson LM & Angel M (1980) Induction of dimethyl-
nitrosamine demethylase activity in mouse liver by poly-
chlorinated biphenyls and 3-methylcholanthrene.
Biochem Pharmacol 29, 1375–1383.
58 Yadav S, Dhawan A, Singh RL, Seth PK & Parmar D
(2006) Expression of constitutive and inducible cyto-
chrome P450 2E1 in rat brain. Mol Cell Biochem 286,
171–180.
Mitochondrial targeting of rat CYPs in yeast cells N. B. V. Sepuri et al.
4630 FEBS Journal 274 (2007) 4615–4630 ª 2007 University of Pennsylvania. Journal compilation ª 2007 FEBS