alcohol (Shimada et al., 1997a,b), and acted on it moderately in hydrolysis of tuna oil
(Shimada et al., 1995b). These results showed that the fatty acid specificity of a
lipase changed slightly in different reaction system. Thus it is important to evaluate
the fatty acid specificity under the reaction conditions employed in order to outline
the lipase characteristics to maximal effect.
Regiospecific analysis of the transesterified oil was carried out by Grignard de-
gradation with allyl magnesium bromide, followed by isolation and analysis of the
1,3-diglyceride fraction (Shimada et al., 1996a). Table 4 shows the fatty acid com-
positions of the 1(3)- and 2-positions in the original tuna oil and transesterified oil
obtained by acidolysis of tuna oil with CA. The fatty acid compositions at the 2-
positions in the oils before and after the reaction were almost the same, and CA
was incorporated mainly at the 1,3-positions. It was confirmed from these results
that only fatty acids at 1(3)-position in tuna oil were exchanged for CA byRhizopus
lipase.
Triglycerides in the original tuna oil and transesterified oil were analyzed by
HPLC (Figure 5) (Shimada et al., 1996a). Most triglycerides in tuna oil were eluted
from the octadecyl silica (ODS) column after 25-min retention. In contrast, trigly-
cerides in the transesterified oil were eluted before 25 min, and found to be new
components. These results showed that all the transesterified oil contained one or
two CA(s) at 1(3)-position(s) of the triglycerides. Main peaks were collected and
their fatty acid compositions analyzed. The structure of each peak was estimated
on the basis of molar ratio of the fatty acids, and is shown in Figure 5. The desired
1,3-caproyl-2-docosahexaenoyl-glycerol (8D8) was contained in peak I, and was
eluted together with 1,3-caproyl-2-eicosapentaenoyl-glycerol (8E8).
As shown in Figure 4, acidolysis of tuna oil with CA reached the steady state after
40 h. Hence, the acidolysis was repeated by transferring immobilized lipase to a fresh
tuna oil/CA substrate mixture every 2 days (Table 5) (Shimada et al., 1996a). The CA
content in triglycerides was 42.5 mol% in the first cycle of the reaction, and immo-
bilized lipase could be used during 15 cycles (30 days) without a significant decrease
in the CA content. The CA content gradually decreased after prolonged use, and
reached 31 mol% after the 20th reaction. The CA incorporation was, however, re-
covered to the original level by extending the reaction period of the 21st reaction to 4
days. The 22nd reaction was performed for 2 days in a mixture containing 2 % water,
but the incorporation of CAwas not recovered. These facts suggested that the decline
8.3 Production of highly absorbable structured lipid 137
Table 4.Fatty acid composition of 1(3)- and 2-positions in tuna oil and transesterified oil obtained by
acidolysis.
Fatty acid composition (mol%)Oil Position
8 : 0 16 : 0 16 : 1 18 : 0 18 : 1 20 : 4 20 : 5 22 : 6
Original tuna oil 1,3 n.d. 14.7 4.9 5.0 15.4 2.0 3.5 8.8
2 n.d. 7.2 0.8 0.4 3.3 0.7 2.2 11.9Transesterified oil 1,3 41.9 3.7 1.1 0.7 2.5 0.7 1.6 7.8
2 0.5 7.1 1.1 0.4 3.2 0.6 2.1 12.4