March 25, 1999
Note on P450 evolution in yeasts and early eukaryotes.
The yeast Saccharomyces cerevisiae has three cytochrome P450 genes CYP51 (lanosterol
14 alpha demethylase), CYP56 (DIT2 a dityrosine forming enzyme required for spore wall
biosynthesis), and CYP61 (22 sterol desaturase). The sequences of these proteins are
given below. The Fission yeast Schizosaccharomyces pombe is a second fungal genome
that is nearing completion (82.7% on Mar 22, 1999) and will be completed this year.
So far there are only two P450s found in S. pombe. These are homologs of the CYP51
and CYP61 sequences. The CYP56 sequence has no homolog in S. pombe yet. The
CYP56 sequence may be specific to the spore wall biosynthesis program of yeast so that it
is absent in S. pombe. A dityrosine bluish fluorescence has been seen in Candida albicans
Smail et al., 1995.
Candida albicans is closely related to S. cerevisiae and so it could have the DIT2
gene. S. pombe is more distantly related, perhaps branching from yeast 500-1000
million years ago. A tree of fungal species showing S. cerevisiae, C. albicans, S.
pombe, Neurospora crassa and others is given here .
The Candida albicans genome is also being sequenced. The results from these three
genomes should help to clarify the history of P450s in fungi. The two P450s CYP51 and
CYP61 are found in all three species, but CYP56 is only in S. cerevisiae so far. If
CY56 is in C. albicans, but not in pombe, then it seems probable that CYP56 evolved
from CYP51 or CYP61. The ancestral yeast may have had only the two P450s and not
Since CYP61 is a later enzyme in the ergosterol pathway, it is probable that CYP61
evolved from a duplication and divergence of the CYP51 gene. This would suggest that
the ancestral eukaryote had only CYP51, essential for the formation of cholesterol and
related sterols. These are characteristic of eukaryotes and it is expected that
cholesterol/ergosterol biosynthesis must have existed in the very earliest eukaryotes.
It should be noted that Candida species also have the CYP52 family of P450s that
metabolize alkanes. These are not found in S. cerevisiae or pombe. The question arises
did a progenitor CYP52 P450 exist in the common ancestor that was lost in S. cerevisiae
and pombe, or did alkane hydroxylases also evolve from the CYP51 ancestor. This
question may also be reversed. Did the common ancestor to all eukaryotes bring with it an
alkane or fatty acid hydroxylase that gave rise to CYP51. The CYP52 sequences are most
like a cluster of bacterial P450s called the eukaryote-like bacterial P450s. These include
CYP102 (a fatty acid hydroxylase) and CYP110. This cluster next joins with the CYP4
clan that also includes fatty acid hydroxylases. This may be the more parsimonious
history of eukaryotic P450s, that they derived from a eukaryote-like bacterial fatty
acid hydroxylase and gave rise to CYP51 and cholesterol biosynthesis very early on in
the eukaryote lineage. Mutagenesis experiments on CYP2C2 have shown that a single
amino acid substitution S473V allows CYP2C2 to accept progesterone as a substrate,
when CYP2C2 is normally a lauric acid hdroxylase, so this is not a far fetched idea.
Ramarao M, Kemper B, 1995.
It will be useful to know the distribution of CYP52 sequences among the fungi. If
they are only found among the fungi related to Candida, but not in Neurospora,
Penicillium, or S. Pombe, then these might represent a novel development of the
Candida related yeasts.
The genome sequence of Giardia is also being done Giardia genome project
page. The position of Giardia on the tree of life is in flux at the moment. It
was thought to be very ancient, one of the earliest eukaryotic branches, but this is
being revisited in the light of more recent studies. Several amitochondriate
eukaryotes are being reexamined, since HSP sequences in these organisms appear to be
derived from mitochondrial genes. Therefore, the loss of mitochondria is now thought
to have happened in these species. In any event, the Giardia branch is still
earlier than the plant-animal-fungi divergences. Any P450s in Giardia will possibly
be primitive features. The only problem with this organism is its anaerobic habitat.
It may have dumped any primitve P450s it had, since it is an anaerobe. After 12000
sequence reads, no P450 has been found in Giardia, but a solid match has been found to
the NADPH cytochrome P450 reductase 26% identical to the N terminal half (e-9).
query seq is Giardia, Sbjct is rat reductase.
Query: 414 IPRGTIIYMTCTFFAGEHPPASKEFIAWLQTVNPSLRPFRDIRFAVFGMGSKNYTTFCAA 593
I + +++ T+ G+ +++F WLQ + L ++FAVFG+G+K Y F A
Sbjct: 128 IDKSLVVFCMATYGEGDPTDNAQDFYDWLQETDVDLT---GVKFAVFGLGNKTYEHFNAM 184
Query: 594 SKNADKSIEIFGGTRILDALHLDRDEFKSDDSAYIHWKK-------DLFKVLGLSEQPVI 752
K D+ +E G RI + D D +D +I W++ + F V E+ I
Sbjct: 185 GKYVDQRLEQLGAQRIFELGLGDDDGNLEED--FITWREQFWPAVCEFFGVEATGEESSI 242
Query: 753 STNKIIVTKNTSLPDKWVCDVS----------PLGYKRGIMSKV---KVLSDGKVDGVVH 893
+++V ++ + + ++ P K ++ V + L+ G ++H
Sbjct: 243 RQYELVVHEDMDVAKVYTGEMGRLKSYENQKPPFDAKNPFLAAVTANRKLNQGTERHLMH 302
Query: 894 L-YEITCPCMKYEAGGHCAILPRN 962
L +I+ ++YE+G H A+ P N
Sbjct: 303 LELDISDSKIRYESGDHVAVYPAN 326
This sequence is from the Giardia genome project at the Marine Biological
Laboratories, Woods Hole, MA.
Dictyostelium discoideum is another primitive eukaryote whose genome is being
sequenced. In this case there are already 45 ESTs representing 18 different P450
genes(Dictyostelium sequences). This shows that
eukaryotes more ancient than yeast can have numerous P450s. One of these ESTs is the
probable CYP51 of Dictyostelium (AU033519). One of these 18 sequences is complete. The other full length sequences will be needed to compare them to the fungal sequences, but there is no strong candidate for a CYP61.
>S. pombe CYP61 C22 sterol desaturase
>Sacchromyces CYP61 C22 sterol desaturase
>Candida albicans AL033396 71% to Saccharomyces CYP61
>S. pombe CYP51 lanosterol 14 alpha demethylase
>Saccharomyces CYP51 lanosterol 14 alpha demethylase
>Candida albicans CYP51 X13296
Percent identities from Do-it-yourself WU-BLAST (some partial alignments)
51 Sc 51 Sp 51 Ca 61 Sc 61 Sp 61 Ca 56 Sc
CYP51 Sc 49% 65% 26% 25% 26% 23%
CYP51 Sp 47% 24% 23% 24% 22%
CYP51 Ca 29% 24% 24% 26%
CYP61 Sc 51% 69% 27%
CYP61 Sp 52% 28%
CYP61 Ca 26%
The C. albicans sequences are more closely related to S. cerevisiae than to pombe as
expected. If a CYP56 homolog exists in albicans it should be about 65-69% identical.