Fatty Acid Binding by Soybean Lipoxygenase-1

(Click here to go to picture gallery--no text)

Two different research groups have published three papers describing the crystal structure of soybean lipoxygenase-1. All of these structures are of the inactive ferrous form of the enzyme. The references are as follows:

  1. Boyington, J. C., Gaffney, B. J., & Amzel, L. M. (1993). "The three-dimensional structure of an arachidonic acid 15-lipoxygenase". Science, 260, 1482-1486.
  2. Minor, W., Steczko, J., Bolin, J. T., Otwinowski, Z., & Axelrod, B. (1993). "Crystallographic determination of the active site iron and its ligands in soybean lipoxygenase-1". Biochemistry, 32, 6320-6323.
  3. Minor, W., Steczko, J., Stec, B., Otwinowski, Z., Bolin, J. T., Walter, R., & Axelrod, B. (1996). "Crystal structure of soybean lipoxygenase L-1 at 1.4 Å resolution". Biochemistry, 35, 10687-10701.

The pictures that follows were based on coordinates of the 1.4 Å structure (1996 paper). These were kindly provided by Bernard Axelrod prior to publication of the structure. Updated coordinates have been deposited with the Brookhaven Protein Structure Databank.

Tertiary Structure of SLO-1

Soybean Lipoxygenase-1 is mostly alpha helical. The structure has been described as composed of two domains (reference 1) or as composed of five domains (reference 3). When I look at it, I see 9 regions. Region 1 is the beta sheet part, and is roughly coincident with what both crystallographic groups have called domain 1. Regions 2, 4, 6 and 8 wrap around the outside of the protein, and regions 3, 5, 7 and 9 form the core of the protein and provide all the residues at the interior surface of the fatty-acid-binding cavity (see below) and the residues which bind the non-heme iron atom. These regions are shown in the picture below (double-click to enlarge it).

Fatty-Acid Binding Cavity

The first crystallographic structure reports noted a long thin cavity running through the lipoxygenase molecule. In the 1.4 Å structure, the cavity has closed up so that it is no longer fully accessible to bulk solvent. However, protein motions are likely to make it possible for fatty-acid molecules (i.e. substrate and product) to move in and out of the cavity.

A 1.4 Å radius Connolly surface is traced out in the diagram below. The dots are only visible in the enlarged picture. This picture is in the same orientation as that above, but shows only the backbone ribbons for the "core" regions 3, 5, 7 and 9. Also shown in this picture are the ligands to the iron: His 499 and His 504 from region 5, His 690 and Asn 694 from region 7, and the terminal carboxylate (from Ile 839) in region 9.

Molecular Modelling Shows Arachidonic Acid Fits in Cavity

Compared to linoleic acid, the "natural" substrate for SLO-1, arachidonic acid is slightly bigger (20 vs. 18 carbon atoms) and more sterically constrained (four double bonds vs. two) . However, arachidonic acid is known to be regio- and stereospecifically converted to a hydroperoxide at the 15 position. I placed the arachidonic acid molecule in the cavity and performed molecular modelling to show that a plausible ES structure could be obtained with only minimal rearrangements of atoms from the crystal structure.

In this molecular modelling, the iron was assumed to be iron(III) and made six-coordinate with the addition of a hydroxide ligand, based on results from a previous EXAFS study performed by students working with me and in collaboration with Mark Nelson and Rebecca Cowling of DuPont: Scarrow, R. C., Trimitsis, M. G., Buck, C. P., Grove, G. N., Cowling, R. A., & Nelson, M. J. (1994). "X-ray Spectroscopy of the Iron Site in Soybean Lipoxygenase-1: Changes in Coordination upon Oxidation or Addition of Methanol.". Biochemistry, 33, 15023-15035.

All three views below show the modelled arachidonic acid as a grey stick figure. The lower panel is in the same orientation as the pictures above and shows mainly residues from region 5 (yellow in the pictures above) as well as a few residues from region 7 around Arg 707 (which may act as a gate at the end of the cavity--see reference 1). The upper panel shows what it would look like from the other side. Imagine folding the pictures at the red line and folding the top half down on top of the bottom half--this is the spatial orientation between the two views.

Double-click on any picture to get a better view.

Showing the backbone ribbon structure and stick figures for the side chains which form the walls of the hydrophobic cavity in which the arachidonic acid resides.

Atoms colored by type--white =hydrogen, green = carbon, red = oxygen, blue = nitrogen, purple = iron.

Atoms colored by type of residue, with hydrophobic in green.

Insights into Mechanism from Molecular Modelling

The following pathways have been proposed for lipoxygenase catalysis

Possibility of iron-hydroxide acting as active site base (BH)

In our initial docking of linoleic (or arachidonic) acid bound to SLOA, both the pro-R and pro-S protons at C11 (or C13) were about 4 Å from the hydroxide oxygen.

To see how close the pro-S proton could approach the hydroxide, we restrained the distance to be 2.0 Å, performed dynamics, then released the restraint, per formed 500 fs of dynamics and minimized. The views shown so far have been the results of these calculations. The geometry about the hydroxyl oxygen appear set up to form a hydrogen bond with the carbonyl oxygen of Ile 839 and to act as a proton acceptor of the pro-S hydrogen.

These calculations show the plausibility of Fe-OH acting as the active site base for abstraction of the pro-S hydrogen. They are not meant to imply that the conformation shown is a global energy minimum, and the results of our ESFF force field calculations are still problematic in terms of giving chemically reasonable coordination geometry around iron. In particular, the calculations locate the hydroxyl H at position a rather than the chemically-intuitive position b.

Iron - substrate bond would require major conformational change

We were unable to dock deprotonated fatty acids to form an organometalic Fe-C bond with the substrate. Formation of such an ES complex would require major rearrangement of the protein structure. This does not disprove, but makes more unlikely, the organometallic mechanism (path B of mechanistic picture above).

Return to Rob Scarrow's home page. Or send e-mail to rscarrow@haverford.edu.

This page maintained by Rob Scarrow, Last updated 12/2/97.