110
Cytochrome P450
exciting, the structures of mammalian,
membrane-bound CYPs are becoming
available. The Frst is a rabbit microso-
mal steroid hydroxylase, CYP2C5, and a
commercial Frm has reported structure
determination of two human hepatic CYPs
(2C9 and 3A4). Because eukaryotic CYPs
are integral membrane proteins and are
consequently more hydrophobic, they are
more difFcult to crystallize. However, that
is true of proteins in general where more
than 2,000 structures of soluble proteins
are available, yet only about 50 membrane
protein structures have been solved. The
recent success in structure determination
of eukaryotic P450s has required modi-
Fcation by mutagenesis to generate less
hydrophobic molecules. We can expect
additional structures of human P450s to
appear in the near future.
Generally speaking, there is little se-
quence conservation between
CYP
gene
families. However, the CYP structures all
look the same from a distance, assuming
what is now called the
P450 fold
.C
lo
se
r
analysis reveals, however, subtle but im-
portant differences. More than 10 years
ago, Gotoh described in the CYP2 fam-
ily six ‘‘substrate recognition sequences’’
(SRS). These regions are predicted to inter-
act with the substrate and are hypervariable
between gene families but are more con-
served within P450s that carry out similar
o
rt
h
es
am
er
e
a
c
t
i
o
n
s
.H
a
v
i
n
gc
o
n
s
i
d
-
erable structural information on P450s,
we now think of these regions in terms
of secondary structural elements. SRS1
is generally found in the region of B
0
-
(or B)helix and BC loop, SRS2 in the C-
terminal region of the ±-helix, SRS3 in the
N-terminal region of G-helix, SRS4 in the
I-helix, SRS5 in
β
1–4, and SRS6 in
β
4–2.
The lengths of these secondary structural
elements vary between P450 families to ac-
commodate substrates of different shape,
size, and polarity. ±rom structural studies,
we know that the active site in P450s is
sequestered inside the protein molecule,
anditisnotc
learatpresenthowthesub
-
strate gains access to the enzyme’s active
site. Only in two cases has the structure of
the same CYP plus and minus substrate
been determined – the bacterial enzymes
P450cam and P450BMP. In the case of
P450BMP (CYP102A1), there is a channel
(substrate access channel) that connects
the surface of the molecule and the buried
active site, which is open in the substrate-
free form of the enzyme. Binding of
the substrate causes signiFcant conforma-
tional changes and results in a closure of
the substrate access channel. This is in
contrast to P450cam (CYP101), in which
there is no obvious route to the active site
and substrate binding is accompanied by
only minor conformational changes.
P450s in the endoplasmic reticulum
have a very hydrophobic amino-terminal
region, which serves as a signal anchor
sequence. All such P450s are inserted co-
translationally into the endoplasmic retic-
ulum membrane by the signal recognition
particle pathway, and this amino-terminal
anchor participates both in association of
the P450 with the membrane and in direct-
ing the proper folding pathway for the pro-
tein. However, removal of this hydropho-
bic tail from the primary sequence of the
P450 does not generally lead to synthesis
of a soluble P450. Rather, such truncated
P450s are still found to be associated with
the membrane, indicating that other hy-
drophobic stretches in the P450 molecule
participate in membrane association. Even
P450s in the mitochondrion do not have
discernible hydrophobic signal anchors.
They, however, have amino-terminal pre-
cursor extensions that are removed prote-
olytically as the unfolded P450 is taken up
previous page 1430 Encyclopedia of Molecular Cell Biology and Molecular Medicine read online next page 1432 Encyclopedia of Molecular Cell Biology and Molecular Medicine read online Home Toggle text on/off