738 



Fishery Bulletin 94(4), 1 996 



greatly reduced in reactivity with the secondary rab- 

 bit anti-goat IgG antiserum. The same anti-66 kDa 

 IgG was adsorbed two more times with fresh, 

 insolubilizedL. apodus saline extract for 24 h at 4°C, 

 and Western blots showed that the reaction of L.jocu 

 was completely eliminated, whereas a trace reaction 

 by L. apodus was detectable. The marker protein and 

 the L. griseus extract reactions were similar to those 

 in Figure 4 (data not shown). Therefore, we conclude 

 that the 66 kDa protein from L. griseus must have 

 different and multiple immunodominant regions 

 compared with the 66 kDa proteins of L. jocu or L. 

 apodus. Otherwise, after the third adsorption of the 



Figure 3 



Western blot analysis of the graduated reaction 

 of'lutjanid saline extracts to anti-66 kDa protein 

 IgG (primary antibody) followed by horseradish 

 peroxidase rabbit anti-goat IgG (secondary anti- 

 body). The samples are as follows: lane 1 = puri- 

 fied L. griseus 66 kDa protein, 0.34 ug total pro- 

 tein; lane 2 = L. griseus extract, 235 |ig total pro- 

 tein, classified as a strong reaction; lane 3 = 0. 

 chrysurus extract, 235 |ig total protein, classi- 

 fied as a weak reaction; and lane 4 = L. 

 cyanopterus extract. 235 ug total protein, classi- 

 fied as not detected. The arrow marks the posi- 

 tion of the 66 kDa protein. Table 2 shows the 

 reactions of soluble saline extracts of 14 differ- 

 ent lutjanid species with the goal anti-L. griseus 

 66 kDa protein IgG. 



antiserum with insolubilized L. apodus, the 

 immunoblots in Bl and B4 would have been greatly 

 reduced or negative with the extract of L. griseus, 

 but they were similar to the preadsorption blots. 



The 66 kDa protein was purified separately from 

 soluble extracts of the three species as described in 

 Table 1. Figure 5 shows the proteins after SDS-PAGE 

 and Western blots. The only discernible difference 

 was that the 66 kDa protein from L. griseus migrated 

 slightly slower than the proteins from L.jocu and L. 

 apodus, indicating a slightly heavier molecular mass. 



Next, it was determined that all three proteins are 

 glycosylated when the method of O'Shannessy et al. 

 (1987) is used. Separately, each of the purified pro- 

 teins was oxidized with Na periodate. Available car- 

 bohydrate moieties were then labeled with biotin- 

 hydrazide. After SDS-PAGE, glycoproteins were de- 

 tected with peroxidase-labeled streptavidin and 4- 

 chloro-1-naphthol (data not shown). 



Finally, N-terminal amino acid sequences of the 

 three proteins were obtained for comparative stud- 

 ies. Table 3 shows that the sequences from L. griseus 

 and L. jocu for the first ten amino acids were ap- 

 proximately the same (residue no. 4 was not inter- 

 pretable for L. griseus). Comparison of L. griseus with 

 L. apodus showed that only as many as 3 of 10 amino 

 acids were the same. In addition, only 3 of 10 amino 

 acids of L.jocu were the same as L. apodus. Thus, as 



