Schultz et al.: Immunologic methods for species identification of early-life-stage lutjanid fishes 



739 



12 3 4 



12 3 4 



A B 



Figure 4 



Western blot analysis of saline extracts from three closely related 

 lutjanids, L. griseus, L.jocu, and L. apodus. The goat anti-L. griseus 

 66 kDa protein IgG was used before (A) and after (B) adsorption 

 with 100 mg of insolubilized L. apodus saline extract for 24 h at 

 4°C. Lane 1 = purified L. griseus 66 kDa marker protein, 1 ug total 

 protein; lane 2 = L. jocu extract, 235 ug total protein; lane 3 = L. 

 apodus extract, 235 ug total protein; and lane 4 = L. griseus ex- 

 tract, 235 ug total protein. The arrow marks the position of the 66 

 kDa protein. 



far as the first ten N-terminal amino acids, the 66 

 kDa protein of L. apodus was different from that of 

 L. griseus and L. jocu. 



Because other studies have shown L. griseus, L. 

 apodus, and L. jocu to have a close phylogenetic re- 

 lationship (e.g. Sarver et al., in press), we wanted to 

 determine if there were obvious differences in their 

 respective protein profiles. Fifty percent of A.S. pre- 

 cipitates of saline soluble extracts of each species 

 were prepared at the same time; the precipitates were 

 solubilized and dialyzed with saline, and they were 

 compared in SDS-PAGE. Figure 6 shows that the 66 

 kDa protein was present in each of the three species 

 (open arrow). However, there are at least three pro- 

 teins that were present in L. apodus that were not de- 

 tectable in L. griseus (closed arrows). Although other 

 lutjanids have not been screened for these proteins, 

 this procedure increases the possibility of isolating spe- 

 cies specific proteins for antibody production, and even- 

 tually for species identification of early life history 

 forms. Even if one or more of the three proteins of L. 

 apodus were present in other lutjanids, some of the 

 determinants on each protein, including different amino 

 acid or glycan substitutions, or both, would probably 



be structurally different and thus would increase the 

 possibility of producing specific monoclonal antibodies. 



Discussion 



A trace single-chain 66 kDa glycoprotein was puri- 

 fied to homogeneity by FPLC technology, starting 

 with soluble extracts of adult L. griseus (Table 1; Fig. 

 1). The glycoprotein was not detected in Western blots 

 with the biotinylated goat anti-whole L. griseus an- 

 tiserum, but it was identified in extracts on SDS- 

 PAGE (Fig. 1). This finding is not unusual because 

 trace proteins in human serum frequently are not 

 reactive with an antiserum produced in animals to 

 whole human serum (senior author's unpubl. obser- 

 vations). A polyclonal antiserum was produced to the 

 purified glycoprotein in a goat, and after removal of 

 nonspecific reactivity by adsorption with insolu- 

 bilized snapper and nonsnapper proteins, Western 

 blots were carried out with the IgG fraction of the 

 antiserum and with soluble extracts of 14 other 

 lutjanid species that were available to us. Because 

 of a comparatively strong reaction in Western blots 



