Great wave off the coast of Kanagawa
Last modified Nov. 19, 2002
D. Nelson
Great wave off the coast of Kanagawa
You must have a Chime plug-in to view these images. These plug-ins can be obtained from the Chime resources web page. There are some browser limitations for this plug-in. Netscape 4.08, 4.7x and 4.8x work, but Netscape 6.x and 7.x do not work. Windows Internet Explorer works, but not Mac Internet Explorer. All these images work fine on a Mac G4 with Netscape 4.7 using the Chemscape Chime 2.6 SP3 plug-in. I have viewed these images on two different PCs using Internet Explorer (IE) and there are some problems. Some of the images are not colored properly or they have wireframe showing over the top of the desired structures, sometimes the orientation is wrong, i.e an edge view of the whole molecule. I discovered that this was due to calling the same PDB filename more than once. It worked on the first call, but on all later calls of the same PDB filename it did not work. I fixed this by making duplicate PDB files with different names, so the same file was not called twice. Here is a link to a Chime troubleshooting page where it says "Chime versions before 2.6SP3 work in IE versions before 5.5SP2, while Chime 2.6SP3 is required in IE5.5SP2 and later."
Go to the images of a P450cam
mutant PDB 1J51 and some comparable views of CYP2C5. later in the views, some CYP2C5 views deal with non-bacterial motifs.
X.CHEN,A.CHRISTOPHER,J.JONES,Q.GUO,F.XU,R.CAO, L-L.WONG,Z.RAO
Later images of CYP2C5 are from the structure 1DT6
Structure Of Mammalian Cytochrome P450 2c5.
P.A.Williams, J.Cosme, V.Sridhar, E.F.Johnson & D.E.Mcree
Notes on the 36 views. You may want to open this page in a separate window, so you can jump back and forth, or you can print it. (I cheated, there are actually 37 views).
This series starts with the raw pdb file imported into RasMol. There are four
P450 molecules in the unit cell. The second view isolates just one of these four
with the heme and the inhibitor shown in stick display and in color. The third
view converts the wireframe to a ribbon display that is more familiar and easier
to interpret. The fourth view colors the I-helix blue and the L-helix magenta, to
aid in orientation. This view also colors the Cys357 yellow in stick display
format. The fifth view zooms in on the oxygen binding pocket. Here the I-helix
has been selected along with the heme, Cys357 and the inhibitor. The four
conserved residues in the I-helix signature GGXDT have been color coded and
labeled to make them easier to see. You can see that the Gly248 (labeled on the
alpha carbon) has to be glycine or at least a small side chain because it points
directly at the center of the heme. Any side chain here would block access for
oxygen binding to the heme iron.
I strongly recommend holding down the mouse to get the popup menu to stop the
rotation and then display this view in spacefill format. Look at the Gly249, also
labeled on the alpha carbon. This alpha carbon is very near to one of the heme
vinyl groups. Any larger side chain would probably push against the heme,
interfering with its insertion into this pocket.
Genbank M12546 P450 cam wt sequence (mutations in this structure are shown as (Xnnn) following the wt amino acid. The Numbering convention does not count the N-terminal Met. Residues discussed in the text are highlighted red and numbered.
Thr252 points directly at the oxygen binding site, while Asp251 points away from
the heme and the inhibitor binding site. The Asp251 group is also well conserved,
but it seems to be interacting with another part of the structure, not visible in
this view. It is closest to Asn255 in the I-helix. One of the oxygens from the
Asp251 carboxyl seems to be in contact with the Asn255 side chain nitrogen. The
255 location is often occupied by an Asn or another polar side chain in bacterial
P450s. Thr185 is within three angstroms of the Asp251 carboxyl and so is Arg186.
The sixth view shows this interaction. Arg186 and Asp251 are almost kissing, with
the two carboxyl oxygens nestled inbetween the two external nitrogens of the
guanidinium group. This interaction is probably quite strong, but the Arg186 is
not conserved. It lies at the end of the F-helix. This charge pair interaction
will anchor this end of the F-helix relative to the I-helix. The paper by Poulos et al. (J. Mol. Biol. 195, 687-700 1987) discusses this charge pair in terms of a key distortion of the I-helix needed to allow oxygen to bind to the heme iron. Without a slight kink in the helix here there is not enough room for the oxygen to fit.
In addition to Arg186, Lys178, also in the F-helix, is within 3 angstroms of the Asp251 carboxyl. This is shown in the 7th view in magenta.
The WXXXR motif is a highly conserved region in P450s. The W is usually seen in eukaryotic p450s, especially in animals. Histidine is found in place of W fairly often in fungi, plants and bacteria. Bacteria do not generally have Trp, they most often have His, so the His is probably the ancestral condition. This protein (CYP101) has a Gln here, which is less common. The WXXXR motif is found in the C-helix. Views 8 and 9a show this motif. The two conserved residues are neigbors on the same side of the C-helix, one turn apart. They coordinate one of the heme propionates. This probably accounts for their strong conservation over more than 3 billion years of evolution. They are part of the heme cradling architecture of the P450 fold. The view in 9b shows the homologous amino acids Trp120 and Arg124 in CYP2C5. Notice that the Arg112 and Gln108 or Trp120, Arg124 only coordinate one of the oxygens of the carboxyl group on this propionate. The other oxygen interacts with His 355 (P450cam), only two amino acids away from Cys357 the heme iron ligand.
These two additional side chains are shown in view 10. The heme propionate carboxyl is shown in spacefill format to emphasize the proximity of the coordinating side chains. View 11 focuses on a closeup of his355, gln108, arg112 and the propionate carboxyl. The His 355 lies inside the P450 signature sequence.
FXXGX(R,H)XCXG. This location is most often Arg or His. An exception is in the CYP7 and CYP8 families that have polar uncharged sidechains there. CYP39 has Phe in this location. Surprisingly, Lys is not tolerated. Only Mycobacterium tuberculosis CYP137A1 has Lys at this location in over 300 sequences in an alignment. His is found in almost all bacteria. His is also seen in some fungi and a few plants. Most eukaryotes have Arg. There must have been an early substitution for Arg in eukaryotes, though CYP51 still has His. This may reflect an ancestral condition.
The other heme propionate is coordinated by Arg299 and Asp297. This is discussed in the paper by Poulos et al. (see above). The Asp propionate interaction is a rare carboxyl-carboxyl hydrogen bond usually seen in crystals grown at low pH, so it is an anomaly here. These two residues are in a beta strand designated beta 1-4 by J. A. Peterson and S. E. Graham-Lorence in Bacterial P450s structural similarities and functional differences Chapter 5 of Cytochrome P450 Structure Mechanism and Biochemistry second ed., edited by Paul ortiz de Montellano. pp. 151-180. This same strand is called beta 3 in the Poulos paper.
View 12 shows Arg299 and Asp297 coordinating the heme propionate in the whole molecule. View 13 zooms in and looks only at the heme and the beta rich region that includes Arg299 and Asp297. The five yellow strands compose a single beta sheet, while the orange strands are beta 2 in Petersons nomenclature or beta 4 in the Poulos nomenclature.
The Arg299 position lies between the EXXR motif and the PKG motif. It is almost always Arg or His, with Lys occuring in some plant sequences. There are very few exceptions to this positive charge and one can predict that it serves a conserved function in the protein. One exception is CYP102 or P450BM3. It has a Leu in this location. CYP505A1 or P450foxy also has Leu there. These two P450s CYP102A1 and CYP505A1 are related. They both have fused NADPH cytochrome P450 reductase domains. CYP505A1 probably arose from a lateral gene transfer across kingdoms. CYP61 sequences have a Tyr at this position.
The Asp297 is not conserved and is presumably a unique interaction.
Pro268 is found at the junction of the I-helix and the J-helix. As expected this introduces a turn. View 14 shows the proline at this location with part of the I-helix and the J-helix. This pro is conserved in many but not all P450s.
Also in this region is the conserved aromatic side chain at position 263. This is Phe in P450 cam, but it is often Trp in mammalian P450s. Tyr is also found here.
View 15 shows the interaction of F263 with H270. These rings are perpendicular to one another and touching in a herringbone stack. Val338 in the non-helical non-beta region between beta 1-3 and the L-helix, is also in contact with both of these rings. This contact between F263 and H270 may help fix the geometry of the I-and J-helices relative to each other. The V338 is only four residues from R342, a part of the PERF motif region. The PERF sequence is absent in CYP101, but the Arg side chain is conserved and it is probably doing the same thing as R in the PERF motif.
The EXXR motif is one of the most conserved motifs in all P450s. There are no exceptions to the Glu at this location and only a few recent unpublished exceptions to the Arg. These sidechains are on the same side of the K-helix, one turn apart, and they form a charge pair. The Asp side chain may be too short for this pair to form, thus Glu is the only amino acid observed here. Arg342 from the modified PERF motif joins them to help lock in the "meander" sequence just upstream of the heme Cys357. View 16 shows these side chains. Two nitrogens from Arg342 and one nitrogen from Arg290 point away from this contact. Three carbonyl oxygens from Ala333, Ala334 and Pro335 are facing the Arg nitrogens less than 3 angstroms away. So there is a crescent of three adjacent carbonyls above the guanidinium groups in the charge triad. View 17 shows this with the AAP sequence in yellow, carbonyls in blue. In the wt sequence A334 is a Cys. Therefore, it is interesting to look at the WT P450 cam structure to see if the carbonyl crown is still intact.
P450cam wt 1DZ6 was examined at the EXXR PERF triad region to see if the carbonyl crown of the triad was the same even though C334 was different from A334 in the mutant structure. View 18 shows that the structures look the same.
We now look at the mammalian P450 CYP2C5 to see the N-terminal motifs not found in the bacterial P450s.
Rabbit CYP2C5 M55664
The KYG motif is CYG (60-62) in CYP2C5. It occurs at a turn between the A helix and the first strand of the large beta sheet that contains the WXXXR motif (View 19). Very often the next amino acid after the KYG motif is a Pro. This is true here also. This pro lies right at the beginning of the first beta strand, where it induces a sharp turn. Gly62 is very tight against pro63 and the gly alpha carbon points away from the turn out into space. It is not obvious from the structure why this has to be a Gly, other than Gly residues facilitate turns. A larger side chain would stick out into space. Perhaps the first 29 amino acids that are missing from this structure lie close to this space and force this side chain to be small.
The Tyr is surrounded by PTP (33-35) and Phe57 shown in yellow. P33 and T34 are in close contact with the Tyr61 ring. P35 and F57 wrap around and under the tyr ring. It is true that there are few N-terminal conserved motifs. Here we see two of these rare conserved regions are close in space and in contact. The PPG sequence before this (30-32) sticks out in space. They are the first three amino acids in the structure. Though the PPGP motif is highly conserved in mammalian P450s, we cannot see what it is doing, since the rest of the N-terminal is missing.
After P33 T34 P35 there is another conserved motif. PIIGNI in CYP2C5, This is
P (hydrophobic)x2 GXL in many other sequences. This short sequence forms a 180 degree bend with Ile 39 and Gly40 at the corners of this very sharp turn. The conservation of this six amino acid sequence in many eukaryotic P450s suggests the abrupt turn structure is conserved. The turn is on one of the corners of the trigonal P450 prism shape. It does not interact with any other part of the P450.
View 20 shows the location of this turn in the intact P450 structure.
View 21 shows the PKG motif. The PKG motif is found in the sequence (in green) YFIPKG amino acids 376-381 of CYP2C5. This sequence is closely associated with VRFRN 370-375 (blue). Together they form a hairpin structure between beta 1-3 (orange) and beta 1-4 (violet). These amino acids correspond in the sequence to beta-2 of P450 cam, though they are not shown as beta structure in CYP2C5 crystal structure 1DT6. They clearly occupy the same structural space as the beta-2 structure. Lys380 points out into space and may form a contact with a carboxyl on the reductase. This Lys is equivalent to Lys384 of rat CYP2B1, that has been implicated in charge-pair interactions with the reductase (Shen S, Strobel HW Role of lysine and arginine residues of cytochrome P450 in the interaction between
cytochrome P4502B1 and NADPH-cytochrome P450 reductase. Arch Biochem Biophys 1993 Jul;304(1):257-65.) The highly conserved position must be conserved for a reason. If it points into space, a contact with another protein is very likely.
View 22 shows the same region in P450 cam. The KG sequence is conserved, though the rest of this sequence region is not conserved. Lys314 points out in the same direction as Lys 380 in CYP2C5.
The next six views, 23-28 show the heme signature sequence and how the side chains in the signature interact with the heme and each other.
There is a region in the middle of some P450s I call the ETAM exon because this exon begins with ETAM or sometimes QTAM or KTAM. The joint at this exon is phase 1 so the first nucleotide comes from the previous exon. This means that GAG = Glu can become CAG = Gln or AAG = Lys in the first amino acid. This sequence is seen in human 4V2 54 amino acids after the WXXXR motif in the C-helix.
There are numerous other sequences.
This region corresponds to
The conserved Asp is an Asn in CYP2C5 (view 31). It points away from the E and I-helices and must be interacting with other polar sidechains in another part of the sequence. This interaction is with Glu148 and Arg186 shown in Views 34 and 35. This Arg186 is not the same as the Arg186 of P450 cam shown in views 6 and 7.
The sidechains of Cys175, Ile178, Cys179, Ile182 and Phe183 form a hydrophobic pocket (View 32) that is filled by Phe292 of the I-helix (View 33). The I182 F183 sequence is equivalent to the AM of ETAM. The SV sequence that is equivalent to ET in ETAM is on the back of the E-helix behind the hydrophobic pocket. So is the H (in the same location as G in ETAMG).
Phe292 is in the sequence FGAGTET that is part of the I-helix oxygen binding pocket. This snug interaction of F292 and the E-helix probably helps to solidify the architecture around the oxygen binding pocket.
The Asn176 side chain that pointed away from the E-helix interacts with a charge pair between Glu148 on the D-helix and Arg186 from the loop between the E and F-helices. To orient the viewer the whole molecule is shown in view 34 with helices D, E and F colored (View 34). The charge pair interaction is shown in view 35.
Many mammalian P450s have a PR or P (hydrophobic) sequence at the end or very near the end. The end of CYP2C5 is in a beta strand. The C-terminal PI sequence of CYP2C5 486-487 sticks out at the end and two His residues have been added, (part of a His tag?). However, YQLCFI 480-485 interacts closely with the loop between the D-helix and the E-helix ASPCDP 161-166 (view 36a). In P450 cam this loop contains a beta strand, but it is not shown as a beta strand in CYP2C5. The other side of this interaction is another beta strand 453-458. This is shown in view 36b in white.
MTTETIQSNANLAPLPPHVPEHLVFDFDMYNPSNLSAGVQEAWA
VLQESNVPDLVWTRCNGGHWIATRGQLIREAYEDYRHFSSECPF(W87)
IPREAGEAY(F96)DFIPTSMDPPE Q108 RQF R112 ALAN
QVVGMPVVDKLENRIQELACSLIESLRPQGQCNFTEDYAEPFPI
RIFMLLAGLPEEDIPHLKYLTDQMT R186 PDGSMTFAEAKEA
LYDYLIPIIEQRRQKPGTDAISIVANGQVNGRPITSDEAKRMCG
LLLV(L247) G248 G249 L D251 T252 VVNFLSFSMEF
LAKSPEHRQELIERPERIPAACEELLRRFSLVADG R299 ILT
SDYEFHGVQLKKGDQILLPQMLSGLDERENAC(A334)PMHVDF
SRQKVSHTTFGHGS H355 L C357 LGQHLARREIIVTLKEW
LTRIPDFSIAPGAQIQHKSGIVSGVQALPLVWDPATTKAV
MDPVVVLVLG LCCLLLLSIW KQNSGRGKLP PGPTPFPIIG NILQ
IDAKDISKSLTKFSECYGPVFTVYLGMKPTVVLHGYEAVKEALVDLGE
EFAGRGSVPILEKVSKGLGIAFSNAKTWKEMRRFSLMTLRNFGMGKRS
IEDRIQEEARCLVEELRKTNASPCDPTFILGCAPCNVICSVIFHNRFD
YKDEEFLKLMESLNENVRILSSPWLQVYNNFPALLDYFPGIHKTLLKN
ADYIKNFIMEKVKEHQKLLDVNNPRDFIDCFLIKMEQENNLEFTLESL
VIAVSDLFGAGTETTSTTLRYSLLLLLKHPEVAARVQEEIERVIGRHR
SPCMQDRSRMPYTDAVIHEIQRFIDLLPTNLPHAVTRDVRFRNYFIPK
GTDIITSLTSVLHDEKAFPNPKVFDPGHFLDESGNFKKSDYFMPFSAG
KRMCVGEGLARMELFLFLTSILQNFKLQSLVEPKDLDITAVVNGFVSV
PPSYQLCFIPI
human CYP4V2 CALDIIC ETAMG
C. elegans 29A2 CALDIIS ETAMG
C. elegans 29A3 CTLDTIC KTAMG
C. elegans 29A4 CTLDIIC GTAMG
C. elegans 31A1P CTLDIIC ETSMG
C. elegans 31A2 CTLDIIC ETSMG
C. elegans 31A3 CTLDIIC ETSMG
C. elegans 32A1 CTLDIIC ETAMG
C. elegans 37A1 CALDIIC ETAMG
Drosophila 4C3 CTLDIVC ETAMG
Drosophila 4D1 CTMDTIC ETAMG
Drosophila 4D2 TALDIIA ETAMG
Drosophila 4D8 AALDIIA ETAMG
Drosophila 4E1 LTLDVIC DTAMG
Rat 4A1 MTLDTVM KCAFS
RABBIT 2C5 APCNVIC SVIFH
But the intron location is different in these sequences. The intron boundary is retained in CYP4V2. The structural feature associated with this sequence is the E-helix. Ile178 in the VIC sequence is in contact with a heme vinyl sidechain (view 29). This occurs at a close contact with the I-helix (view 30).