mutagenesis of I593 (I593A) was carried out and the mutant enzyme converted AA
to 12- and 15-HETE in a ratio of about 1 : 1 (Borngra ̈ber et al., 1999). To compare the
relative importance of the sequence determinants, we mutated each of them to a
small alanine (ala-scan) or to a space-filling phenylalanine (phe-scan) so that the
volume of the pocket was either increased or decreased. The results of these experi-
ments suggested that enlargement or alteration in packing density in the substrate-
binding pocket of the rabbit 15-LOX increased the share of 12-LOX products,
whereas a smaller active site appeared to favor 15-lipoxygenation (Borngra ̈ber et
al., 1999). Moreover, we found that the sequence determinants functionally interact
with each other. Alteration in the positional specificity was observed when a space-
filling amino acid was mutated to a smaller residue, but this was reversed when a
large amino acid was introduced at another critical position (Borngra ̈ber et al., 1999).
A schematic view of the topology of substrate binding at the active site of reticu-
locyte-type 15-LOXs is shown in Figure 9.
In order to go beyond a 15-LOX/12-LOX exchange and to convert the 15-LOX to
an 8- or even a 5-lipoxygenating mutant, it was attempted to increase the volume of
the substrate-binding site as much as possible by creating double, triple, and quad-
ruple mutants. Double mutations, such as F353A+I418A or F353L+I593Awere well
tolerated and the mutant enzyme converted AA to 12-HETE. However, after triple
alanine mutation (F353A+I418A+I593A), the enzyme exhibited a residual activity
of only about 15 %. Here again, 12-HETE was detected as major product. This result
may suggest that the pocket became too large to align properly with the fatty acid
substrate. In addition, several other multiple mutants were created aimed at increas-
ing the volume of the substrate-binding pocket, but unfortunately, all of them turned
out to be catalytically inactive. Thus, one may conclude that excessive re-engineer-
ing of the LOX active site may not be possible without a loss in functionality (Born-
gra ̈ber et al., 1999).
So far, all studies on the enzyme/substrate interaction and on the positional spe-
cificity were carried out with free fatty acids as substrates. In contrast, the reticu-
locyte 15-LOXs are capable of oxygenating more complex substrates (Murray and
Brash, 1988; Ku ̈hn et al., 1990a) and there is experimental evidence suggesting that
membrane phospholipids and lipoprotein cholesterol esters may be the preferred
natural substrates (Ku ̈hn and Brash, 1990). Since these substrates are much more
space-filling than polyenoic fatty acids, it remains to be investigated how they
326 15 Application of Lipoxygenases and Related Enzymes
Figure 9. Schematic view of the topology of substrate binding at the active site of reticulocyte-type 15-
LOXs. The solid circles represent the hydrogen acceptor of the enzyme (nonheme iron) and the ‘horse-
shoe-like’ structure symbolizes the hydrophobic substrate-binding pocket of the enzyme.