Front Matter

or N425M exchange (A424 and N425 of the human 5-LOX align with I418 and

M419 of the rabbit and human 15-LOX) transformed the wild-type 5-LOX to en-

zyme species which produced significant amounts of 8-HPETE (8–15 %) in addi-

tion to 5-HPETE. In contrast, an A603I exchange (A603 of the human 5-LOX aligns

with I593 of the rabbit enzyme) did not influence the positional specificity. The

fourth sequence determinant of the 15-LOX (F353) aligns with F359 in 5-LOXs.

In order to reduce the volume of the binding pocket, we mutated F359 to an

even more bulky tryptophan. This F359W mutant turned out to be a major 5-

LOX with a significant share of 8-HPETE formation (5-/8-HPETE ratio of about

2 : 1). Unfortunately, no 15-HPETE was observed with any of the single mutants.

To reduce further the volume of the substrate-binding pocket, we combined the ef-

fective single mutations to create double, triple, and even quadruple mutants. When

the N425M was combined with a F359W and A424I (F359W+N425M and

A424I+N425M) the share of 8-HPETE formation was strongly increased (60–

90 %), but no 15-HPETE formation was observed. However, when the F359W

was combined with the A424I, a small (5–10 %) but significant share of 15-lipox-

ygenation was detected. This effect was even more pronounced when the

F359W+A424I+N425M triple mutant was constructed. Here, an almost 1 : 1 distri-

bution of 15- and 8-HPETE was found, and both products turned out to be chiral with

a strong preponderance of theS-isomer. For the rabbit 15-LOX it has been reported

that I593A exchange altered the positional specificity (Borngra ̈ber et al., 1996), but

inverse mutagenesis on the human 5-LOX (A603I) did not have major effects. It

might be possible that the loss of the pocket volume achieved by this mutation

for 5-LOXs was not strong enough to alter the positional specificity of the wild-

type enzyme. However, when this amino acid exchange was performed on the

F359W+A424I+N425M triple mutant the product pattern was further shifted to-

wards 15-HPETE formation. In fact, with this quadruple mutant, AA was mainly

oxygenated to 15-HPETE (85–95 %), with 8-HPETE being a minor product.

The 15-lipoxygenating quadruple mutant was purified by ATP–agarose chroma-

tography and its basic enzymatic characteristics were determined. Although we al-

tered the positional specificity of AA oxygenation, no major effects were observed

on other enzymatic characteristics. Both the wild-type and the mutant enzyme were

activated by Ca2+, ATP and phosphatidylcholine, and the substrate specificity with

different fatty acids was very similar. Biological membranes and lipoproteins, which

are suitable substrates for reticulocyte-type 15-LOXs but not for 5-LOXs, were not

oxygenated by the mutant. Although the arachidonate dioxygenase activity of the

mutant enzyme was only 30 % of that of the wild-type counterpart, the mutant en-

zyme exhibited a comparable leukotriene synthase activity with 5-HPETE as sub-

strate. When the wild-type 5-LOX was incubated with 5-HPETE the leukotriene A 4

hydrolysis products were the major compounds formed (80–90 %), and only small

amounts (5–20 %) of double oxygenation products were detected. In contrast, with

the 15-lipoxygenating quadruple mutant the product mixture was the other way

around. With this enzyme the double oxygenation products (5S,15S)- and

(5S,12S)-DiH(P)ETE were the major reaction products, and leukotriene A 4 was

only formed in small amounts (7–15 %). However, since the quadruple mutant con-

verted 5-HPETE much faster than the wild-type enzyme the overall yield of leuko-

triene A 4 formation was similar with the two enzyme species.

328 15 Application of Lipoxygenases and Related Enzymes