Subject: mitochondrial P450s
Dear Michael,
I was checking the ever-improving annotations of the P450s and I have some
considerations that apply to 11 of them, these that are putative
mitochondrial enzymes. While I detected the mitochondrial P450s at the
jamboree and discussed some of them with David Nelson, only now I had the
time to look at all of them and communicate my observations to FlyBase with
a suggestion to correct the annotations. I thought that since the evidence
below applies to all 11 P450s, I would rather write to you and ask you to
communicate it to the FlyBase annotators (and file it as a personal
communication if you wish) rather than using 11 'fix annotation' entries.
In case it is more practical I will do the 'fix annotaions'.
While most P450s are indeed microsomal, some are mitochondrial. Among the
fly P450s 11 appear to be good candidates for mitochondrial enzymes. These
are: all CYP12s \- CYP12A4, CYP12A5, CYP12B2, CYP12C1, CYP12D1, and CYP12E1,
CYP49A1, CYP301A1, CYP302A1, CYP314A1, and CYP315A1.
Several lines of evidence suggest that these are putative mitochondrial
P450s: 1) They are most closely related to mitochondrial P450s from other
organisms as judged by overall sequence similarity; 2) Presence of a
putative mitochondrial targeting sequences at their N-termini \-
amphipathic, rich in basic residues (mainly R and K) in contrast to the
N-termini of microsomal P450s which are highly hydrophobic; 3) presence of
a conserved pair of basic residues (with the exception of CYP314A1 where Q
is present instead of the second basic residue)in the C-terminal half of
the sequences, probably orthologous to the RxxxK pair of the putative
adrenodoxin binding site of mammalian mitochondrial P450s.
Additional strong functional evidence that these are mitochondrial P450s
derives from two recent studies:
1) The first family 12 member, CYP12A1 (from the house fly Musca domestica)
has been firmly established as a mitochondrial enzyme by localization in
the mitochondria by immunocytochemistry and by functional characterization
of the recombinant protein. The later was efficiently reduced only by the
mitochondrial electron transfer partners adrenodoxin reductase and
adrenodoxin and not by the microsomal cytochrome P450 reductase (Guzov et
al., Arch.Biochem.Biophys.  359:231-249 , 1998). Thus, the other CYP12s are
more than likely to be also mitochondrial enzymes.
2) In the case of CYP302A1 (disembodied, Chavez et al., Development
 127:4115-4126 , 2000), all biochemical and genetic evidence points to a
mitochondrial localization of the enzyme. Surprisingly, the GadFly summary
incorrectly states that 'It encodes a cytochrome P450 (EC:1.14.14.1)
putatively involved in metabolism which is a component of the microsome.',
rather than of the mitochondrion (mitochondrial inner membrane).
The distinction between microsomal and mitochondrial P450s is interesting
and important and I suggest that it is reflected in the annotations of the
sequences to avoid confusion \- now they are all listed as microsomal. Let
me know if further justification of the proposed changes in the annotations
is needed.
P.S. I understand that David Nelson has communicated to you the corrections
he made to the Drosophila P450 protein sequences. They are all correrct now
on his site and I think it would be best if they are used for the
annotations in GadFly. I still do not understand how 30 % of the P450
protein sequences given by Celera to GenBank are wrong....
Best regards,
Boris
Boris Dunkov
University of Arizona